Links 11/17: Sri Linka

Paul Palaiologos Tagaris, Byzantine con man, “was appointed an Orthodox bishop, pretended to be the Orthodox Patriarch of Jerusalem, switched from Greek Orthodoxy to Roman Catholicism and back again, supported both the See of Rome and the Avignon anti-popes in the Western Schism, and managed to be named Latin Patriarch of Constantinople”.

Inside the cutthroat cash-flush dog-eat-dog world of…online mattress reviews?

Canadian FDA-equivalent bans Soylent for violating standards. Particular standards weren’t mentioned, but jimrandomh’s research suggests it was a regulation saying that all meal replacements have to be low-fat – which is not supported by research and might be actively unhealthy. I’m glad I still live in a country where…

…hold that thought. FDA is trying to ban kratom, an herb many people use as a chronic pain treatment or to help overcome opiate addiction. Previously the DEA tried the same, but backed down after an outcry from kratom users and their families who say they rely on it to stay well. Conspiracy theorists note that Trump’s FDA head Scott Gottleib took $100,000 from GlaxoSmithKline, which is working on a synthetic patent-protected version of kratom that will become people’s only option if the real thing is banned.

Related: how have insulin prices increased by 10x in the past twenty years, and what does that mean for diabetics? Also: Trump nominates man whose firm tripled price of insulin to regulate drug companies. Some of the discussion I saw around this article recommended mentioning that Eli Lilly has a patient assistance program to help poor people afford their insulin, though obviously this isn’t anywhere near a solution to the real problem.

The Olympics used to be more willing to deviate from their core mission of sports, and people have competed for Olympic medals in activities from town planning to sculpture to music.

Geolibertarians frustrated with early 20th-century society’s failure to adopt Georgist land taxes built their own private intentional community, which continues to exist in the present day. Delaware’s Odd, Beautiful, Contentious Private Utopia. Interesting look at its century-long history of trying to resist assimilation and stick to its principles – including having the first desegregated schools in the region.

New poll analyzes Americans’ views of global warming, with some surprising results – only 13% of people don’t think global warming is happening, and only 30% don’t believe it’s human-caused. 23% of people think scientists mostly believe global warming’s not human-caused, suggesting that most skeptics aren’t disbelieving scientists so much as unaware of them. Also, 39% of Americans say that there is a greater-than-even-odds chance that global warming will cause the extinction of the human race.

Man speaking in slang says “give me a lawyer, dawg”; judge rules he was not exercising his right to ask for a lawyer because he actually asked for a “lawyer dog”, which does not exist.

Did you know there’s a whole field of empirical software engineering studies? Recent claimed findings include eg: “there is no evidence that [industry] experience is a significant factor in either quality of productivity [of programmers]”. Study describes itself as “exploratory” and I am totally unqualified to judge.

@beachdeath: “The CIA is releasing tens of thousands of files and videos from bin laden’s compound today, except his DVDs of ‘home on the range’ and ‘ice age: dawn of the dinosaurs’ and his copy of final fantasy vii, because those are copyrighted” is not a sentence i ever thought i would type, but 2017 continues to be full of surprises”. And a list of some of Osama’s DVDs and computer games.

Redditor asks why Google Home gives such weird readings when asked “what is the temperature inside”; top commenter notes that depending on your tone of voice, Google Home will answer this question with the temperature in the city of Side, Turkey.

A group of scientists including Friend Of The Blog Stephen Hsu launch Genomic Prediction, a company that uses genetic testing to helps families select embryos for IVF. It’s a natural outgrowth of existing tests that check for Down Syndrome and other very serious genetic diseases, but the exciting new part is where they can analyze risk for some polygenic diseases (ones that depend on contributions from hundreds or thousands of genes) as well as the usual simple stuff. Important because several socially-important traits like height and intelligence are polygenic so this advance essentially places science at the point where it could select for these traits if it were considered legal/ethical to do so (in practice, current height selection algorithms would probably do a good job; current intelligence selection algorithms are still very limited but advancing quickly). Given that Hsu has said a bunch of times that his end goal is using genetics to increase human intelligence, there’s no way him setting up a company that has exactly the right technology to do so is a coincidence, even if it’s not their first product or likely to happen any time soon.

Related: most people are really averse to genetic testing and embryo selection, even to the point where they will (in real life, not hypothetical questions) choose to give their kid a 50% chance of a horrible and invariably fatal genetic disease rather than use it.

Some of the ads Russia bought on Facebook for the 2016 election are hilarious. Some of these have a sort of “HOW DO YOU DO, FELLOW KIDS AMERICANS” vibe to them, but maybe not much more than a lot of real political ads do

Catholic readers: is this a ridiculous misinterpretation of something, or did the Pope really say that married priests might be a good idea?

Reminds me of the recent discussion of Confucians vs. Legalists on whether people should be allowed to know laws: Georgia’s laws are available only if you pay for access; the state pursues people who publicize them for copyright violation. ACLU is on the case.

This Reddit post reminded me of my essay on Kolmogorov complicity. Many US states have legalized medical marijuana, which in practice means a boom industry of special-purpose clinics where unscrupulous doctors give medical marijuana cards to everyone who comes in, claims to have a symptom, and pays them a fee. But these doctors can’t say “We are frauds who give everyone marijuana cards”, so lots of marijuana-wanting and even marijuana-needing people won’t go to the doctor because they’re afraid they’ll get turned down.

Clash of civilizations: Armed militia shuts down comic book convention in Libya.

The Intercept: Four Viral Claims Spread By Journalists On Twitter In The Last Week Alone That Are False

Gene Expression: “In Kenneth Pomeranz’s The Great Divergence: China, Europe, and the Making of the Modern World Economy he argues that the difference in per capita economic wealth between Europe and China is a relatively recent phenomenon. One of the major arguments he makes is that one has to make an apples-to-apples comparison. Comparing Northwest Europe to China is not apples-to-apples, but comparing Northwest Europe to the lower Yangzi Delta region of Central China is apples-to-apples. Using this measure Europe and China are roughly comparable up until 1800.”

Free speech watch: woman who gave President Trump’s motorcade the finger in her spare time is fired from her marketing job. This is the world that all of you “free speech only constrains the government” and “it’s just people think you’re an asshole and are showing you the door” people have built for us.

More claims that increased health spending does not increase health outcomes.

After years of self-driving cars being five years away, there’s now a date for a self-driving car being available to ordinary people for a commercial purpose: next year. It’s pretty minimal – some cooperating ride-sharing passengers in Phoenix will get self-driving cars without human backup drivers – but it’s happening.

Zvi Mowshowitz, Vladimir Slepnev, and Paul Christiano have announced a $5000 prize for “publicly posted work advancing understanding of AI alignment”. An example of an existing submission, in case you’re wondering what an entry would even look like. I know the sponsors and can vouch that they’re good honest people who are actually going to pay out the money.

Stuart Ritchie and Elliot Tucker-Drob: How Much Does Education Improve Intelligence: A Meta-Analysis. It gets a number (between 1 and 5 IQ points per grade year), but it seems kind of an uninterpretable aggregate (surely educating a naturally-IQ-100 person for fifty years wouldn’t give them an IQ of 150 to 350) – the take home point is that it’s a positive and significant number. There’s been a lot of work on “gains to IQ test scores” vs. “gains in g” which I don’t know enough about but which is probably relevant here; unclear how much past work has to be reconsidered in that light. There’s some good discussion between Gwern and Stuart (one of the authors) here.

Sci-Hub loses another lawsuit, apparently on legally shaky grounds, and now might get banned by ISPs or search engines or something. It sure would be bad if this kind of lawsuit led to a Streisand effect that made even more people aware that sci-hub exists and is a website with almost all academic papers on it available unpaywalled for free.

The Trump/GOP tax reform plan will hit grad students super-hard, maybe so hard as to make graduate studies financially impossible unless universities immediately change their compensation structure. The problem seems to be that it counts tuition waivers as taxable income, so if a college pretends it’s charging grad students a $20,000 tuition but waives the fee, those grad students will have to pay taxes of (their tax rate) x ($20,000) without actually seeing any of that income. I don’t understand why universities maintain the fiction of charging tuition and then waiving it, so I’m not sure if they can solve the problem just by not doing that. Some people think this might have very long-run positive effects of forcing universities to actually pay grad students a decent salary, but until then it may be only slight exaggeration to call this the destruction of graduate education in the United States. If you’re a grad student, contact your program to see if they have any ideas for what to do.

Related: Trump tax plan to hit colleges by changing some of their weird tax exemptions. Taxation has always been a little about punishing your party’s political enemies and giving tax breaks to its allies, but this is some next-level stuff here and it’s really blatant how much the new code shifts tax burden onto traditionally Democratic constituencies.

Related: It will also be really terrible for startup employees.

Modern people’s jaws are aligned differently due to their different eating habits. More interesting than it sounds. Also good example of nominative determinism, featuring orthodontics researcher Charles Brace.

The Open Philanthropy Project gives MIRI a $3.75 million grant, the largest it’s ever received. Some commentary here on why this is unexpected. There’s more complicated political background which I don’t think is fully written up but which this post at least hints at. Overall I view this as a really positive development.

Related: an attempt to make neural nets more transparent by investigating what pictures maximally activate each neuron of an image classifier. There’s something very creepy about this, like dissecting the world along some mysterious dimension into incomprehensible conceptual primitives. Also, some neural net is very convinced that “either an animal face or a car body” is a fundamental concept that cleaves reality at its joints, and now I’m questioning how I know for sure that it isn’t.

Related: Katja Grace of AI Impacts finishes her report on recent trends in the cost of computing and top supercomputers.

Related, and maybe this is that “negative effects of already-existing AI” I keep hearing we should worry about: Israel Arrests Palestinian Because Facebook Translated “Good Morning” As “Attack Them”

Blue collar wages are…actually doing pretty well right now? But see comments and caveats on the subreddit.

80000 Hours: What Are The Most Important Talent Gaps In The Effective Altruism Community? Good news for those of you who majored in “good calibration, wide knowledge and ability to work out what’s important”.

Cordelia Fine is good now? She describes James Damore’s Google memo as “more accurate and nuanced than what you sometimes find in the popular literature…[some of his ideas] are not seen as especially controversial”, and declares it “quite extrarordinary about someone losing their job for putting forward a view that is part of the scientific debate”. Interesting how hard it is to find anyone familiar with the science of gender, even the most blank-slatist and furthest left, willing to endorse the narrative treated as 100% proven and obvious in the popular media.

Yemen Facing Largest Famine World Has Seen For Decades, Says UN Chief.

Speculative, but why do so many trans people dye their hair unnatural colors?

New study: naltrexone as good as suboxone for opiate addiction. Also in the same genre of “studies saying an obviously worse drug is as a good as an obviously better one” – Tylenol/ibuprofen as good as opiates for acute pain relief. And heck, let’s throw in this study showing antihistamines work better than benzodiazepines for anxiety relief. I don’t know what’s up with any of these.

Also, niacin-based skin test has decent specificity (and, I’m guessing, no sensitivity) in identifying schizophrenics (vs. mood disorder). Especially interesting if it leads to understanding the etiology or ontology of some schizophrenia subtypes.

This week’s ridiculous non-controversy: Christians are boycotting British food producer Greggs for making a nativity scene with a sausage roll as Jesus. Thanks to LukeBBZ on Twitter for pointing out the kabbalistic implications: “Lord Jesus” spelled backwards is “Susejd rol”, which I guess is close enough.

This entry was posted in Uncategorized and tagged . Bookmark the permalink.

598 Responses to Links 11/17: Sri Linka

  1. Taymon A. Beal says:

    Scott Aaronson, on why universities can’t just abolish graduate tuition if it’s just an accounting fiction:

    “But isn’t there some workaround?” Indeed, financial ignoramus that I am, my first reaction was to ask: if PhD tuition is basically an accounting fiction anyway, then why can’t the universities just declare that the tuition in question no longer exists, or is now zero dollars? Feel free to explain further in the comments if you understand this stuff, but as far as I can tell, the answer is: because PhD tuition is used to calculate how much “tax” the universities can take from professors’ grant money. If universities could no longer take that tax, and they had no other way to make up for it, then except for the richest few universities, they’d have to scale back research and teaching pretty drastically. To avoid that outcome, the universities would be relying on the granting agencies to let them keep taking the overhead they needed to operate, even though the “PhD tuition” no longer existed. But the granting agencies aren’t set up for this: you just can’t throw a bomb into one part of a complicated bureaucratic machine built up over decades, and expect the machine to continue working with no disruption to science.

    It does sound like it’s something of an open question how quickly grantmaking agencies would be able to adjust.

    • stucchio says:

      What this ignores is the actual amounts being taken by universities.

      In my grad school, I got $20-25k of a $65-70k grant. Universities needed the remainder to keep running?

      I knew a guy who’s university took $400k out of a $1M grant in return for some windowless basement rooms which housed a $600k machine. A commercial landlord might have charged $20-30k at most.

      The university doesn’t need to take this money to continue doing scientific research. Science would be profitable at much lower overhead rates.

      Relatedly: https://slatestarcodex.com/2017/02/09/considerations-on-cost-disease/

      • thirqual says:

        You are confused if you think the $400k is just for renting the basement room.

        The overhead/indirect costs (which are almost always a percent of the total grant money) covers the rooms but also in most universities the air conditioning, the power bill (except in special cases of very consuming equipment), often stuff like replacing desks, minor office equipment, library services, cleaning the facilities, equipment depreciation, and the salaries for the administrators (gnnnnnn) and support staff.

        (I do think that overhead is usually excessive)

        • stucchio says:

          I know, the justification is that this is stuff too small to be itemized. Plus some benefits routed through university HR.

          Commercial landlords/Wework/etc or Zenefits/similar companies manage to itemize all this stuff or just include it as part of the cost of rental. They typically charge an order of magnitude less than universities. The idea that these extra services are some unknowable quantity, therefore we’ll charge you 40-60% is just insane.

          Universities can still profitably do science even if overhead is cut drastically.

          • thirqual says:

            So you know, but you prefer to write something erroneous and confusing.

            How often do commercial landlords provide an itemized bill for the library access by the way?

          • albatross11 says:

            Imagine if there were some organizations whose prestige made it a lot easier to get grants from NIH. You could sort-of imagine those organizations finding ways to collect rents based on that prestige….

      • rdplatypus says:

        (Long-time lurker, first time commenter)

        I know that _feels_ wrong, but that kind of overhead rate sounds not that bad. A near “real-world” commercial analogue I can think of is the defense industry: fair to high amounts of lab / capital investment, ample bureaucracy (you think grant applications are bad? Try cost-basis contracting!), etc.

        Because so much defense contracting is cost-basis, they’re required to scrupulously account for overhead costs in order to bill the government appropriately. The first google hit for “defense industry overhead rates” shows a GAO report from 20 years ago that found rates of ~150% during the late-coldwar heyday and ~175% in the 90s drawdown. I don’t know what things are like now, but they’re probably somewhere in that range.

        $1.50ish of overhead for $1 of labor costs would translate to $600K of that $1M grant. I don’t know how much of the remainder would actually be used for labor vs. equipment, but then again cost-basis contracting sometimes also includes certain equipment purchased on-contract as well as use of company capital equipment (test sets, etc.). So my back-of-the-napkin estimate would say that those numbers, egregious as they may sound to an academic, are actually in-line with similar costs calculated by “real-wold” enterprises.

        Maybe somebody knows something about Pharma overhead rates less marketing as another comparison?

        • stucchio says:

          The contracting you describe would be comparable to universities lending out professors. But this is something very different.

          I’m very explicitly comparing the cost of lab space + cost of administering a health plan + similar things to the cost of WeWork + Zenefits.

          • rdplatypus says:

            Then I must be missing something. It would seem to me that “lending out professors” is basically comparable to performing grant-based research, except that it’s grad students that are being lent out.

            Are you trying to argue that there’s a different arrangement in which science could be done more efficiently? Because I was trying to illustrate that a somewhat-similar industry has a somewhat-similar cost structure, which would suggest that the overhead you’re trying to enumerate is being resonably estimated by the existing system.

          • stucchio says:

            It’s not comparable at all. The closest analogue is probably WeWork + Zenefits + other misc itemizable services (e.g. library).

            The reason a contractor charges a lot for labor is that the contractor needs to be able to provide labor whenever the military wants. When the military is done with them, they’re still on the contractor’s payroll. And while some downscaling can be done, the contractor still needs to be able to respond quickly when they get called up.

            The university model is not keeping a cohort of grad students around, just waiting for a professor to get a grant. The university model is literally professor gets a grant, then the professor hires a grad student, and when the grant runs out the grad student is fired.

            Similarly, the university doesn’t keep expensive labs and equipment around just waiting for a professor with a grant to rent. They just provide real estate, the grant provides the rest.

            I’ll also point out that the majority of the people defending universities here would probably not defend military contractor overhead as necessary. Of the folks criticizing Trump here, who would oppose President Bernie cutting military contractor payouts?

          • Ilya Shpitser says:

            “Lending out” professors is comparable to outside consulting. Which professors do a fair bit of.

      • Chalid says:

        Would overhead also go to shared facilities? The supercomputer cluster where you can reserve time, the machine shop that everyone shares, the room full of seven-figure NMR machines the whole chemistry department uses, etc. How about the cost of maintaining the kind of emergency response team that is skilled and brave enough to deal with a fire in a lab working on god-knows-what?

        Actually, for that matter, is it funding basic maintenance of things like the roads on campus, or is that coming from other sources?

        • slightlylesshairyape says:

          Chalid, the NMR and the computer cluster are definitely paid directly out of grant money and excluded from overhead as durable purchase. The power and A/C to run the cluster come out of university operating funds.

          As for the rest, money is fungible. That tuition waiver goes into the university general fund and becomes indistinguishable from every other dollar spent on books, roads, campus police, football coaches, and IT staff.

          • Watchman says:

            So once the grant money runs out, the NMR and computer clusters are not funded? Universities invest a hell of a lot in equipment budgets, at times paying someone to basically keep a unique, often technically obselete, machine going full time.

            You are correct though that the rest of the money is fungiable, and as is is funging (?) in a university, it will be spent on the various things that universities spend money on which makes them universities and not just research labs.

          • slightlylesshairyape says:

            You can definitely pay for a technician salary out of a grant.

            Mundane operating expenses like power/AC to the cluster or generalized IT (networking?) are paid out of the university budget, which anyway already got overhead from the grants for it.

          • Eltargrim says:

            @Watchman: Not the US, but the Ultrahigh-field NMR Facility in Ottawa (housing a 900 MHz/21.1 T magnet, highest field magnet in the country) was absolutely facing closure due to budget cuts. Eleven million dollar investment facing closure over the loss of a $100 000/year grant.

    • IvanFyodorovich says:

      At least in biology a lot of government graduate fellowships or training grants pay tuition (for example, NSF GRFP gives $12k toward tuition, NIH training grants give up to 16k/60% of tuition), so if the university stops charging tuition, the university loses that money. In theory that wouldn’t hurt science as a whole, since that NIH or NSF could do equally good things with that money. However, international tuition payments also exist. I have a colleague from China, and a Chinese grant pays part of her American graduate tuition. If we drop tuition to zero, the university/American science loses that money. If we follow the tree up further we can say that Chinese science gets to keep the money, but it certainly hurts American academia.

      Also, I have less of a sense of this, but I think there are people in the humanities who really do pay these tuition amounts (certainly in Master’s programs) and actually take a lot of classes that justify charging them. Selectively abolishing tuition payments for research-oriented grad students might be politically impossible at universities.

      • Rosemary7391 says:

        It’s very normal in the UK for postgraduate rates to vary from taught programs to research programs. Also, my tuition fee is (IIRC) around £4k, so my first reaction was “surely it can’t make that much difference?” – apparently US universities operate very differently…

        I really cannot imagine that universities will let grad students starve or seriously decline in number. Could they up the payment to the student by the amount of the new tax? Or treating them as employees with the tuition becoming employment overhead?

        • IvanFyodorovich says:

          Of course they could up the payment, but the universities are looking down the barrel of 1) Having to up the payment 2) A tax on endowments 3) The abolition of the estate tax, which encouraged rich people to donate to universities and 4) Stagnating research funding. Hard to see how they avoid cutting research.

          • stucchio says:

            It’s hard to see why they would cut research.

            Research is currently a gravy train where they take 50% overhead, drastically in excess of their costs. If this is cut marginally (and make no mistake, this is a marginal cost), it’ll still be drastically in excess of their costs.

            No matter what your cash crunch is, it’s pointless to cut a profit center. That’s true even if the profit center is marginally less profitable.

          • IvanFyodorovich says:

            When endowments tank/costs rise they do cut research. Just ask anyone who applied for a faculty job c. 2009.

            You are Rockefeller University. You have no undergrads and are a pure research institute. You get money from grants and by spending X% of your endowment per year. Now your endowment is being taxed, you have to pay your grad students more, and you are getting fewer bequests. What do you cut? You hire fewer new faculty, don’t build a new building full of labs, give less generous startup packages, cut intramural grants, cut core facilities etc. And the calculation isn’t that different for say a medical school attached to a larger university.

          • Watchman says:

            stucchio,

            Overheads are so high at UK universities that most research is probably a loss leader, unless negotiated with commercial partners (who happily cover the overheads interestingly – maybe something to consider in your argument). Because research funding is bascially a monopoly (national government, unless you are in Germany or a couple of other federal states I believe), it means the price paid for research can be dictated because there is not really another option. Although the US federal government is famousily poor with money, all I have seen suggests the same will apply there.

            Incidentally, if the US is going to lose research (and it will if they lose PhD students – the need for bodies to do the research is high) then the UK stands to gain quite nicely. It helps that UK universities are charities and therefore tax exempt on their core business (education and research).

    • slightlylesshairyape says:

      Except that universities are already double-taxing that grant money.

      First, the university takes a percentage of the grant total[2] (usually around 33%[1]) as ‘overhead’. This is intended to pay for facilities, offices, etc. Then, the grant money is used to pay tuition waivers for the grad students, which again goes to the university.

      I wonder if some intrepid blogger wrote about the spiraling expenses of universities and how no one is trying to control them.

      Source: was grad student.

      [1] They denote this 50%, because the NIH grants you $X and then the university takes a fraction of that on top. So a 50% overhead amounts to 33% of the money going to the university. Weird and totally not the way anyone else would write that down . . .

      [2] Minus certain excluded expenses like durable equipment and tuition waivers. Apparently there is a limit to the chutzpah since the universities don’t pay themselves $50k/yr/student in ‘waived tuition’ and then claim overhead on that as well.

    • Controls Freak says:

      slightlylesshairyape has a good point about how they’re taking two cuts.

      I don’t actually work for a federal funding agency, but I’m regularly a reviewer for proposals and have been on selection boards for an agency close to ours that does do external funding. We can figure out how to appropriately fund research. One of the things we do is analyze the proposed budget, and that generally involves looking at the proposed tasks alongside it. “Ok, Prof X just had Task A, and his budget is about the size of one grad student; that’s probably good. Prof Y is on Tasks B, C, and D; her budget is about one/two post docs and two/three grad students; makes sense. Prof Z is also contributing substantially to Tasks A and B, but he barely has enough for half a grad student and a month of summer salary; that doesn’t seem sufficient.”

      We’re also well aware of which universities have high overhead rates. That’s a pretty secondary consideration compared to our valuation of the technical merit and our faith in the profs’ abilities. (Ok, I’ll admit that it’s actually lower down in the list. We definitely exhibit the, “It’s just money, and it’s not mine,” phenomenon when we decide how to spend the government’s money.)

      Anyway, if universities move some of their overhead cut out of tuition (and tuition waiver tax breaks) and into the explicit overhead line-item, it’s not going to do much to affect our ability to fund research. If anything, it just makes the sticker price a little more accurate. I don’t see how that’s a bad thing… especially since we’re not particularly thrifty with gov’t funds anyway.

      • youzicha says:

        Anyway, if universities move some of their overhead cut out of tuition (and tuition waiver tax breaks) and into the explicit overhead line-item

        But I guess Scott Aaronson’s original point was that they might not be able to do this, because grant agencies have rules about the maximum allowed explicit overhead.

        • Controls Freak says:

          I’m not aware of anything like that coming up for our agency (or the related, external funding agency). I just looked at NSF, and it looks like they have no hard cap; universities would just have to negotiate a higher rate. Poor university.

          Economically speaking, others have pointed out that the incidence of this tax is unlikely to fall on grad students (due to their higher relative elasticity). If universities find it difficult to be explicit about the price of things (because doing so would hit some max overhead number I haven’t yet found), then they’ll just have to find some other accounting fiction to hide it. Poor… poor university.

          But really, they could just be explicit about the price of things. They could just raise the salaries of grad students to cover the increased taxes. Then, folks like us will look at the budget proposal and still say, “Ok, they really need X grad students and Y post-docs for this problem, so $Z is about right, plus whatever the lawyers have agreed to on overhead.” (And seriously, we don’t care about overhead negotiations. I’m sure Scott Aaronson couldn’t tell you much about overhead negotiations, either. We all put it squarely in the Let The Lawyers Do Lawyer Things category.) The added bonus is that if the sticker price goes up and the overhead rate remains fixed, they get more overhead. Poor, poor, poor universities.

        • Brad says:

          But I guess Scott Aaronson’s original point was that they might not be able to do this, because grant agencies have rules about the maximum allowed explicit overhead.

          Here’s a crazy idea–maybe universities should cut overhead.

    • JohnBuridan says:

      I have heard from my agricultural entomologist friend that taking that “tax” off grants is a huge problem, and that at least his university was very predatory on the sciences, using the “taxed” money for nefarious purposes, like tennis courts.

    • outis says:

      I wrote a big reply here, but it disappeared. I think something went wrong when I tried to edit it (or maybe it was deleted intentionally?). If it’s the former, please consider this a bug report. If it’s the latter, never mind.

      I’m kind of sad because I thought it was a good comment, and I spent quite a bit of time on it. On the other hand, if this causes me to spend less time commenting in the future, it’s probably for the best.

      • Brad says:

        If it was caught by the spam filter it could be pulled out by Scott. If it was due to the use of a forbidden word, I think it is just gone.

        • outis says:

          It showed up on the page, initially (or at least, I could see it). I edited it to remove a line break, which I think showed up, and then again to make one single word bold, which did not show up (as in, the page still showed the previous version). I think that was when it disappeared, and it would not have been there if I had reloaded the page.

          Does that sound like spam or forbidden word?

  2. Mary says:

    Catholic readers: is this a ridiculous misinterpretation of something, or did the Pope really say that married priests might be a good idea?

    I can’t speak to accuracy of the news story — doing my best to fight the Gell-Mann Amnesia effect; however, there is no doctrinal reason why a husband can not be ordained a priest. The choice to select priests among the single (and widowed) who agree to remain celibate in the Latin Church is a matter of discipline, not dogma.

    I note that other Catholic rites besides the Latin do ordain husbands.

    • manwhoisthursday says:

      Right. Priestly celibacy is not a doctrinal matter. In fact, several of the Eastern churches affiliated with the Roman Catholic church already have married priests. As well, there are some married Catholic priests who were formerly Anglican priests.

      • Mary says:

        , several of the Eastern churches affiliated with the Roman Catholic church

        The technical terms for this is “rites”

        • SEE says:

          Actually, no, it isn’t. The technical term is “autonomous particular church”.

          The term “rite”, while in fairly common popular use, is explicitly depreciated in technical use, because it suggests that the major difference is the liturgy used. But autonomous particular churches each have their whole independent hierarchies, while in a number of cases sharing rites. On the other hand, there are a variety of rites in use inside the Latin Church itself (Roman rite, Anglican and Zaire use variants of the Roman rite, Carthusian rite, Ambrosian rite, several others I can’t recall).

        • rmm4pi8 says:

          In this case ‘church’ is more correct. They are sui iuris churches, which means they are competent in their own affairs, and not part of the Roman Catholic Church (because though they are Catholic, they are not Roman, though they do recognize the authority of the Roman See). Part of that is being liturgically separate, which is why they have their own rites, but here the question of law doesn’t regard liturgy. Sui iuris churches may actually share a rite while being different in other ways. Also, there are multiple rites *within* the Roman Catholic Church besides the Roman rite, so again ‘church’ is more clear.

    • Jack V says:

      Last I heard, it was discussed, many people thought it might be a good idea, but the vatican decided against. But nobody suggested it would be impossible, merely politically impossible. IIRC there are already exceptions for anglican converts who are already married and ordained in the anglican church. So I’m not surprised the Pope would float a trial balloon, but I don’t know how likely it is to happen soon.

      • Deiseach says:

        I am vaguely sceptical because part of the call for married clergy on the part of some groups was due to the thinking “Well, in this day and age no parish is going to contribute enough to keep a married priest with ten kids, so the married priests are going to have to use some kind of contraception and the natural forms are not reliable enough so that means artificial methods, which means the Vatican will be forced to scrap the teaching forbidding contraception (and maybe even abortion further down the line) for the laity, so finally Sexual Revolution, here we come!”

        Also, the idea that there are tons of guys out there who would jump at the chance to be ordained if they could just get married doesn’t seem feasible. From the Protestant side of the fence, being a clergy wife is not fantastic for the spouse due to the expectations of former days still being imposed on modern women, and I imagine the same would hold true for Catholic women, who might not particularly want to be the priest’s wife and have the responsibilities that go with that in a parish. The priesthood is a tough job, a Catholic priest is not at all in the same position as the pastor of a local church (especially one of the pastors who can set up their own church or megachurch), we don’t have the same idea of tithing so the salary is based on the Sunday collections and contributions by the congregation (e.g. Christmas and Easter dues) and trying to keep a wife and family on that would be really hard.

        Anyway, while we’re discussing Catholicism, those Russian ads leave me feeling offended and personally attacked 🙂 The “Jesus armwrestling Satan” one is put out by a crowd ostensibly called the “Army of Jesus” but the icon used is the imagery of the Divine Mercy! So they’re ripped that off (literally, taking the head of Jesus) and wrapped it in an American flag! Hey, stick to insulting non-denominational American Protestantism and leave Catholic iconography alone, you godless (former) Commies! 🙂

        • quaelegit says:

          >we don’t have the same idea of tithing so the salary is based on the Sunday collections and contributions by the congregation (e.g. Christmas and Easter dues) and trying to keep a wife and family on that would be really hard.

          Would a Catholic priest’s wife be allowed to do secular work and bring in income that way?

          • Deiseach says:

            No reason why she wouldn’t and indeed they probably would need a two-income household set up, but that brings its own share of problems (there will always be those in the parish ready to tut-tut about how she should be at home raising the children herself instead of selfishly putting a career first etc.)

          • Anonymous says:

            I wonder how the Eastern Catholics do it. I have heard that the EC married priests are more business-like about donations, with a price list for stuff like burial and baptism and all.

        • albatross11 says:

          In my parish, there are three deacons[1] as well as two regular priests. It seems plausible that in a world where priests could be married, one or more of those deacons would themselves become priests. (Right now, our parish has a fairly young priest, and watching him celebrate Mass alongside a grizzled old deacon makes me think of a second lieutenant standing next to sergeant.)

          I believe the rule for deacons is that they’re allowed to be married when they become deacons, but not to remarry if their then-current wife dies (or leaves, I guess, but that’s probably already baked-in, since they’d surely be married in the Church, and so they’d need an annulment to remarry anyway).

          This makes me suspect that allowing married priests would expand our set of priests somewhat. I don’t think this is a cure-all for all the problems of the Church, but it doesn’t seem crazy to me.

          [1] Deacons are allowed to do many of the things you’d normally associate with priests, like performing baptisms and weddings. They can’t hear confessions or celebrate a full Mass, but they can lead a religious service and can distribute already-consecrated body and blood to the people there. [Edit] Note that you don’t have to be a deacon to distribute consecrated host to people; I can do that, like any other EM. You’ll sometimes see EMs going to visit sick people in the hospital and bring them the Eucharist, and we’re neither priests nor deacons.

    • Watchman says:

      Aren’t married priests in the Catholic rite meant to be celibate though?

      • randallsquared says:

        Googling suggests that the answer is basically “no”. Married Latin Rite priests are expected to fulfill their spousal duties as well as their priestly duties. When I started looking, I did not realize that for Catholics, “celibacy” is literally about marriage, not sex, while “chastity” is about not having sex (but only outside of marriage). I did not find a Catholic word / concept for refraining from sex even in marriage, and Fr. Dwight Longenecker, a married Catholic priest in Greenville, SC (just down the road from me, coincidentally), has given a number of interviews wherein he indicates that refusing to be intimate with one’s wife would be considered wrong (a sin? Not sure…).

        • rmm4pi8 says:

          Celibacy: not getting married
          Chastity: practicing virtue with regard to sex (extamarital sex is obviously out, but also contraception, etc)
          Continence: not having sex (“sexual continence” can be specified if there’s any doubt about confusing it with other kinds; “complete continence” indicates on a permanent rather than temporary basis, e.g. natural family planning calls for periodic continence)

          Whether all priests are called to continence is a matter of some historical dispute; recent suggestions have been “no” but it’s not quite so obvious–there’s certainly a reason that wives have to consent to the ordination of their husbands.

        • RDNinja says:

          1 Corinthians 7:5
          “Do not deprive one another except with consent for a time, that you may give yourselves to fasting and prayer; and come together again so that Satan does not tempt you because of your lack of self-control.” (NKJV)

          Paul suggests elsewhere that some of his teachings on marriage are practical advice rather than doctrine, so it may not be a sin, but definitely not recommended.

    • rmm4pi8 says:

      The Pope has floated this idea, particularly around remote Amazon regions where priests are very difficult to find, so like with the Eastern Rites or former Protestant clergy, this would be a relatively limited exception. Which is not to say that a broader change in discipline couldn’t happen, but I doubt there’s a sufficiently strong constituency for it at this time.

      • RDNinja says:

        I’ve also read that many African cultures won’t listen to a man who has no wife or children, so celibate priests there have a hard time acting as leaders.

  3. MawBTS says:

    Stuart Ritchie and Elliot Tucker-Drob: How Much Does Education Improve Intelligence: A Meta-Analysis. It gets a number (between 1 and 5 IQ points per grade year)

    It should be emphasised that they only make claims about one school year.

    The gains seem to last – half of the effect size is still there by age 70 (p11). That isn’t what I expected. Maybe they’re being pulled upwards by smaller outlier studies – most of the bigger ones trend towards the lower end.

    But it does seem like a solid finding.

    Modern people’s jaws are aligned differently due to their different eating habits. More interesting than it sounds. Also good example of nominative determinism, featuring orthodontics researcher Charles Brace.

    I’ve always felt like my upper and lower jaw don’t interface properly. They feel just a little bit wrong for each other. So this is habitual?

  4. Jayson Virissimo says:

    Any other SSC/LW people working in the online mattress space, or am I the only one?

  5. Clegg says:

    Blue collar wages are…actually doing pretty well right now? But see comments and caveats on the subreddit.

    Labor force participation of prime-age males is around 89%, which is indeed a historic low, but has also been flat since the peak in unemployment following the Great Recession.

    From a few Wall Street Journal articles in the past year, it seems there’s a shortage of workers with a high school education, which is making hiring hard for high-tech manufacturers. There are plenty of people with no college and a lot of them even have high school diplomas, but there aren’t a lot of people with sufficient math and reading ability to read blue prints, follow detailed written instructions, etc.

    This seems plausible, since anybody who does well in high school gets pushed towards college and a white collar career.

    • Wrong Species says:

      Some people act like the decline in male prime-age lfpr is proof in itself that something is wrong with the labor market but there are both supply and demand reasons why that it could be. There’s even some reason to believe that it’s supply side issues like the fact that, controlling for education, immigrants have higher working rates than natives or that only a small percentage of those out of the labor force give “unable to find a job” as their reason. Either way, that one number doesn’t say anything in itself.

  6. skef says:

    The problem seems to be that it counts tuition waivers as taxable income, so if a college pretends it’s charging grad students a $20,000 tuition but waives the fee, those grad students will have to pay taxes of (their tax rate) x ($20,000) without actually seeing any of that income. I don’t understand why universities maintain the fiction of charging tuition and then waiving it, so I’m not sure if they can solve the problem just by not doing that.

    Graduate students get tuition “remission” for doing what the university wants them to do. So, when one is on a university fellowship or has a TA or RA position, one pays no tuition. Otherwise, one does. Since the tuition amounts tend to be quite large, grad school tends to be a bad arrangement unless you’re playing ball.

    There are other reasons having to do with charging on outside grants, for the usual STEMy suspects, but they could come up with some other solution for that stuff.

  7. Nabil ad Dajjal says:

    Edit: ninja’d by skef

    I’m not an expert on the economics of graduate schools, but one important point to remember with tuition is that a lot of fellowships and training grants will cover tuition.

    Right now the NIH is paying for my tuition and most of my stipend. So if my graduate school reduced or eliminated my tuition, that would cost them tens of thousands of dollars. I don’t think they’re going to do that if they have any other choice available.

    • dansimonicouldbewrong says:

      I note that you don’t give any reasons why the university *should* be getting those thousands of dollars–all you point out is that the university *can* get thousands of dollars by maintaining this accounting fiction, and therefore obviously *would*, because, well, that’s what universities do.

      And they’re not alone–researchers themselves do this sort of thing all the time. (That’s why conference spam exists–fake conferences offer researchers the opportunity to spend grant money on vacations instead of research.) In fact the entire academic research industry is shot through with so many layers and varieties of fraud that anything that reduces its volume, this tax law change included, is pretty much guaranteed to be a net gain for the nation.

      • Watchman says:

        I suspect that the university should be getting the money because the university has to continue to exist, and it will not if it incurrs costs such as teaching Nabil (requiring staff pay, facilities and equipment and consumables) without getting income in to cover this and the overheads of continuing to be a university?

        It is what universities do. I think you’ll find that they are required to do so by their charters or the laws setting them up, since no-one thinks financial black holes are a viable way of conducting higher education and research.

        I am happy to accept a debate about the role of universities, as the idea that they are the natural locus for research for example could be happily challenged. But such a debate has to start from the point of view that universities are autonomous corporations and therefore will seek to exist as part of their mission, so you can’t really question why they get paid…

        • stucchio says:

          So lets put some numbers on this. Back when I was a grad student on my Ph.D. adviser’s NSF grant, I was taking home about $20-25k/year on an approx $65-70k chunk of my adviser’s grant. The university took the rest.

          What this paid for was 12 or so research credits, which consisted of me and my Ph.D. adviser sitting in his office doing whiteboard math as well as a few leftover desktop boxes I used for computing.

          Can you tell me what costs I impose that approach even $10k, let alone $40k?

          You get similar figures if you compare the 40-60% grant overhead taken for physical sciences. A windowless basement office suddenly costs $400k/year to rent if you stick a $600k electron microscope in it, as compared to maybe $1000-3000 from a commercial landlord.

          Basically universities have set themselves up as gatekeepers and pilfer about 50% of the funding that congress allocates for “science”. Lets not kid ourselves – if any grad students suffer, it’s because universities chose to make them suffer in order to create pressure to give them back this revenue stream.

          • baconbacon says:

            So lets put some numbers on this. Back when I was a grad student on my Ph.D. adviser’s NSF grant, I was taking home about $20-25k/year on an approx $65-70k chunk of my adviser’s grant. The university took the rest.

            What this paid for was 12 or so research credits, which consisted of me and my Ph.D. adviser sitting in his office doing whiteboard math as well as a few leftover desktop boxes I used for computing.

            Can you tell me what costs I impose that approach even $10k, let alone $40k?

            So if all you and your Ph.D. adviser are doing is sitting in his office doing math, why do you deserve $75,000 for this?

          • stucchio says:

            Baconbacon, that’s the point. The cost to the taxpayer should not be $70k. The university is overcharging.

          • JulieK says:

            Apparently the academic thinks it’s worthwhile, since he doesn’t leave and rent a place from a commercial landlord.

          • dansimonicouldbewrong says:

            I can assure you that funding agencies–whose allocation of research funds is guided overwhelmingly by academic researchers–are not in the habit of handing out grants to independent researchers operating out of commercially rented office space. In fact, they are, for the most part, in the habit of handing out grants according to the same academic-political peer review “Survivor” game that governs every step of an academic’s career, from program acceptance to graduation to publication to promotion to tenure and beyond. Researchers who decline to play the game–kickbacks to host universities and all–are quickly voted off the island.

          • stucchio says:

            JulieK, the professor is not permitted to take the money and buy his own services. I’ve spoken to several who said they would do exactly that if they were permitted to.

            Another excellent way to address this issue would be for the NSF/etc to eliminate this requirement, and allow funded research to happen outside universities (plus a few national labs).

          • John Schilling says:

            I can assure you that funding agencies–whose allocation of research funds is guided overwhelmingly by academic researchers–are not in the habit of handing out grants to independent researchers operating out of commercially rented office space.

            SBIR grants are exactly that, and while they and the related STTR grants make up approximately 3% of Federal R&D spending at present there’s no reason that couldn’t be stepped up if the R1 universities fumble on their end. And there’s a fair bit of government R&D that’s funded through not-small businesses as well.

            There are also funding lines where academics inside government are given a pot of money and told to divvy it up among their colleagues still in academia on the theory that something good will come from academics doing random bits of science. Which is probably true, but A: it’s not the whole of government R&D and B: it’s not clear why it should get special tax privileges.

        • Nabil ad Dajjal says:

          I don’t think that my university is in any danger of ceasing to exist or even cutting back on their core facilities. I don’t want to dox myself but let’s just say that they have more than enough money as it is.

          I’m just less interested in the ethical questions around what they should do than the extremely important practical question of what they will do. I suspect that they’d hang me out to dry tax-wise if it lets them keep the training grant money but as I said above I’m not a finance guy.

          • slightlylesshairyape says:

            There was a fascinating post on MR about this.

            The conclusion seems to be that only the most prestigious places will be able to ‘make up’ the difference in some way (i.e. have the burden fall on the school) rather than on the student. I suppose the “evade the law without outright breaking it” option is still there too..

        • slightlylesshairyape says:

          Right, but at the same time we can say that universities have not demonstrated excellent financial discipline in the past few decades. Their profligate spending is well enough documented that society should take notice rather than give up and say “Well they must continue to exist”.

          We can also say that the institution of the university is comprised of many different (competing) missions and that for the university to take huge fractions of grants and put it in the “general fund” might in fact be a bad deal for research and a good deal for the non-research-missions of the university.

          That is, my impression as a grad student was that the university spent less on our department than we kicked to them in tuition/overhead, and that this went into the fiscal black hole rather than in any kind of improvement we shared in.

        • Brad says:

          I suspect that the university should be getting the money because the university has to continue to exist, and it will not if it incurrs costs such as teaching Nabil (requiring staff pay, facilities and equipment and consumables) without getting income in to cover this and the overheads of continuing to be a university?

          It is what universities do. I think you’ll find that they are required to do so by their charters or the laws setting them up, since no-one thinks financial black holes are a viable way of conducting higher education and research.

          I am happy to accept a debate about the role of universities, as the idea that they are the natural locus for research for example could be happily challenged. But such a debate has to start from the point of view that universities are autonomous corporations and therefore will seek to exist as part of their mission, so you can’t really question why they get paid…

          The armies of deanlets and deanlings (and vice presidents and …) that consume ever larger portions of the annual budgets of universities have nothing to do with education *or* research. And they aren’t necessary for the continued existence of the university. This is proved by the fact that universities didn’t use to have them but nonetheless existed.

          Neither tuition payers nor grant payers ought to be expected to support in great style endless numbers of highly paid functionaries.

      • Nabil ad Dajjal says:

        The thing is, if I’m understanding the situation correctly, this tax change won’t stop them from claiming unnecessary money from the NIH. They’ll still get that full amount but now I’ll be on the hook for it in taxes.

        That doesn’t sound like a net gain for the nation. There’d be just as much fraud as before and I would be broke. Unless me being broke is a top national priority I don’t see how this is supposed to help anyone.

        • slightlylesshairyape says:

          I mean, see the MR post linked above. You have to solve for the equilibrium, which is likely a combination of less ‘tuition waiver’ money being spent out of grants, lower grad take-home pay and higher stipends.

          How this gets allocated will depend on the various marginal forces at work, plus the ability of the universities to devise clever ways to get money from the grant into their hands.

    • jimbarino says:

      The main problem i have with this: if my employer pays my tuition for a class I take, I have to pay taxes on that amount. If they give me a a company car, I have to pay taxes on the amount that I use it for personal driving. Why do universities get a different set of rules?

      Note that I don’t necessarily support this proposal, which would be extremely disruptive to a lot of people. But I have increasingly been viewing the modern university like the late-medieval monestaries – institutions that started out as godly institutions dedicated to poverty and good works, but increasingly amassed power and wealth with no accountability until they were forcibly stripped of it by the emerging state.

      Whether this passes or not, you can guarantee that something like it is going to pass eventually…

  8. rahien.din says:

    Regarding opioids, Chang et al :
    1. Examined acute extremity pain, which is not, for instance, cancer pain, a broken femur, or other categories of disease that cause enduring severe pain.
    2. Allowed their study docs to give some discretionary opioids when they deemed it necessary. They attempted to correct back to intent-to-treat via a multiple imputation model (which is not fully explained and which we may be skeptical of).

    So their study amounts to “if you are willing to give some opioids, then non-opioids work as well as opioids for sprained ankles and dislocated shoulders.”

  9. Sniffnoy says:

    The jimrandomh link points to a non-public Facebook post. Is this written up anywhere else you could link to? Or could jimrandomh repost it here?

  10. Concerned Peterson Fan says:

    Related: anti-communist icon Jordan Peterson asks whether it’s morally okay to doxx communists. Because, you see, he thinks Communists are really bad. This is why I keep saying “But my outgroup really is really bad, so it’s okay to doxx them!” leads to a bad place.

    I just made a quick throwaway account since this is the first time I have commented and I expect to just continue lurking after this.

    Jordan Peterson was not asking if it was okay to dox someone. Doxxing is the act of sharing personal information of someone, generally someone who wanted to remain anonymous, which isn’t was he was asking about. He was asking if it seemed okay to bring attention to the personal sites of people who publicly demonstrated. By demonstrating publicly and without hiding their identity, they are doing nothing to attempt to remain anonymous and many who do this likely do want to be associated to the demonstrations they perform. They also often talk about their demonstrations on their own pages.

    Whether or not you think it is wrong to do that, it seems to me to be qualitatively different from doxxing, especially since he is asking about linking to people’s pages which they have control over, so he isn’t bringing attention to any information that they haven’t themselves made public.

    If Jordan Peterson gave a talk somewhere, and in response his political opposition brought attention to his youtube or facebook page, would that be wrong? I don’t think so. He has control over those pages. If there is information that he wouldn’t want to be publicly known, then he wouldn’t put them on his pages.

    I don’t see how bringing attention to personal sites that the person in question has control over can be considered doxxing.

    • manwhoisthursday says:

      Did Scott delete this?

    • Harry Maurice Johnston says:

      Previously discussed here.

    • harland0 says:

      CNN intimidated a man into silence and deleting his online persona with threats of doxxing. If it’s OK for the media to do, it’s OK for anyone else. I didn’t exactly hear about a big outcry when they did it.

      • Yosarian2 says:

        I don’t think it’s accurate to call a news source reporting on a story as “doxing”.

        I also don’t think that was CNN’s goal; my impression was that they were trying to justify their decision to not release the person’s name because they thought that was what they would be criticized for. Granted they should have been more clear, and I certainly understand why many people read the CNN announcement that way.

        • Ratte says:

          They looked into the background of a nobody who (incorrectly, as it happened) claimed to be the source of a meme the President retweeted. What, exactly, was the story supposed to be? Why would they be criticized for not doxxing some rando Redditor?

          • Matt M says:

            Because their major constituency consists of hardcore left-wingers suffering from Trump Derangement Syndrome who wanted to see that person publicly shamed, fired, and probably physically assaulted.

      • MugaSofer says:

        I definitely recall outcry surrounding this, unless you’re referring to a different incident to the one we’re all thinking of.

    • Brad says:

      I agree we need to push back against ever broader definitions of doxxing, as well as allowing the unconditional support for anonymity and obscurity to leak from the internet into the real world.

  11. somervta says:

    Man speaking in slang says “give me a lawyer, dawg”; judge rules he was not exercising his right to ask for a lawyer because he actually asked for a “lawyer dog”, which does not exist.

    The concurrence doesn’t actually say this – it’s a short opinion, which *only* says that he was not exercising his right to a lawyer, not saying the crucial word ‘because’;

    In my view, the defendant’s ambiguous and equivocal reference to a “lawyer dog” does not constitute an invocation of counsel that warrants termination of the interview and does not violate Edwards v. Arizona,

    The more plausible reading of the opinion is one that would be the same if the word dog/dawg was removed; that the defendant’s request was “ambiguous and equivocal” because it was inside a conditional ‘If you think I’m guilty, why don’t you get me a lawyer’. Orin Kerr defends the claim that this interpretation is a correct statement of the law here.

    • Svejk says:

      State Supreme Court Justice Scott J. Crichton wrote:

      “In my view, the defendant’s ambiguous and equivocal reference to a “lawyer dog” does not constitute an invocation of counsel that warrants termination of the interview and does not violate Edwards v. Arizona, 451 U.S. 477, 101 S.Ct. 1880, 68 L.Ed.2d 378 (1981).”

      If the author of this concurrence wishes to make the case that the use of the conditional in conversation during a stressful experience (interrogation) was sufficiently “ambiguous and equivocal” not to stop the interrogation to provide a lawyer, how much higher should the bar be set for the written words of a sitting judge in a concurrence to a denial of a writ of certiorari not to invite this sort of speculation? The idea that the “lawyer dog” phrase was the source of the ambiguity seems to be a plausible interpretation of this statement. Otherwise it sounds like he is mocking the defendant.

      The Court ruled 8-1 to deny the writ, and it is plausible that most of the Court based its decision on the use of the conditional rather than “lawyer dog”, but this parallel ambiguity in the writings of a Justice who felt strongly enough about the case to pen a concurrence suggests that the bar for clarity of speech in requesting an attorney is set higher than the average citizen would expect. Even Orin Kerr mentions that the use of the “polite conditional” is common in making requests in real conversation.

      The relevant precedent in Davis seems to consider the problem of the conditional, and dismiss it in favor of demanding a direct request, so the decision appears legally solid. It’s good that it is getting attention, because I’m sure most citizens were not aware that you have to phrase your request for counsel like a command in an 80s text adventure game to get results.

    • Watchman says:

      I can’t help but wonder why those doing the interview were not required upon hearing a conditional and unclear instruction around lawyers to actually clarify whether a lawyer was required, perhaps seeking a clear affirmative answer or otherwise being able to ignore the comment. It would seem to be a sensible way to conduct a system where a person being interviewed is entitled to a lawyer.

      And google image search suggests lawyer dogs are a thing to certain people at least.

      • theory says:

        We don’t live in a system where the police are obligated to make every effort to help the suspect. They are only obligated to not affirmatively screw the suspect. If we wanted to change the system to impose on police the same rules that exist for prosecutors, that would have very far-reaching ramifications on law enforcement as we know it, especially once suspects adapt to and exploit the new rules when they play the game.

        • Watchman says:

          We do (or at least those of us in the UK do, and I am led to believe that theoretically you in the US do) live in a system where there is a right for a suspect to have a lawyer though. This is not a change in the system. What you seem to be saying is that ensuring clarity as to whether a suspect requires a lawyer would be changing the system and have far-reaching ramifications. I don’t see this – it would surely just avoid messes like this where people have to fall back on the judgement of a group including an idiot who clearly has an issue with a widely known slang formation for reasons I do not know but have a worryed feeling about.

          Also, I am not sure how suspects would be able to play this game. If the police in an interview are building up pressure in a way that can be broken simply by asking for a lawyer, not only is this available to the suspect at the moment, but it suggests the police are not interviewing correctly and are instead trying to coerce an admission. After all, the US, unlike the UK, actually allows you to not incriminate yourself, so any pratice to force an admission seems somewhat suspect in context. And whilst a suspect could perhaps try and use a conditional clause around a lawyer all the time to avoid actually answering questions, there seems to be no problem with an interviewer stating after say the second occurence of this that should the suspect require a lawyer they will be allowed one if they state so unconditionally, but if they are not sure the police will continue the interview until this is clearly stated. It’s hardly beyond the wit on man to sort something like this out.

        • Jiro says:

          I count maliciously and intentionally misunderstanding the suspect as “affirmatively screwing the suspect”.

          The problem isn’t that the suspect said something truly ambiguous. The problem is that the suspect said something that any reasonable officer would understand, but the officer deliberately misinterpreted it so that he could keep asking questions.

          The purpose of making the officer ask for clarification would be to prevent malicious misunderstanding. Not to help suspects who really don’t know how to speak clearly.

      • Lambert says:

        There are specialists in the law as it relates to dogs, at least.

    • C. Y. Hollander says:

      I’d add that, even were it not linked to a conditional, “Why don’t you just give me a lawyer?” would not technically be a request for a lawyer. It seems to have a similar level of certainty to “Maybe I need a lawyer”, which was the article’s example of something too ambiguous to be considered a definite request for a lawyer.

      • pdbarnlsey says:

        I think language has evolved to the point where “why don’t you [action x]” is not a request for information but a suggestion that action x be undertaken.

        Even Bieb’s agrees:

        • C. Y. Hollander says:

          All right, let’s call it a suggestion. That’s still not quite a request. Would “I suggest you give me a lawyer” count as a definite request for a lawyer? Perhaps so, perhaps not. Even if it does, I’d say that the “Why don’t you…?” formulation is a little weaker, being expressed in questioning form.

          It seems to me that the law in these cases has evolved towards requiring the suspect to be very explicit in his request for a lawyer, to the point of making it crystal clear that he has no doubt in his mind that he wishes a lawyer present. Even the hint of such doubt can be enough to invalidate his suggestion/request.

          • Matt M says:

            When you need a lawyer to help you craft the statement consisting of the exact combination of words in the right order which will compel the police to allow you to speak with a lawyer.

          • Nornagest says:

            The ACLU used to literally give out cards with the phrasing you need to unambiguously invoke your right to a lawyer, drafted by its lawyers. Probably still does.

      • Nornagest says:

        “Why don’t you get me a lawyer?” seems to be one of those sentences where the meaning changes depending on where you put the emphasis. “Why don’t you get me a lawyer?” vs. “why don’t you get me a lawyer?” vs. “why don’t you get me a lawyer?”, etc.

        • pdbarnlsey says:

          I think that’s true, but you’d have to emphasise “don’t” for it to naturally read as a request for information. All the others are imperatives of varying force.

    • The Nybbler says:

      I believe the recommended language in the vernacular is “I ain’t saying shit until you gimmee an <pick expletive from menu> lawyer”. You then should follow up by “not saying shit” until you have an <expletive> lawyer. This flowchart referenced here recently goes through the whole thing. I don’t know how you invoke your fifth amendment rights in slang, but according to the flowchart, asking for a lawyer and remaining silent is sufficient in a police interrogation.

      • Jiro says:

        No it isn’t; read carefully. If they inform of your Miranda rights, you have to explicitly invoke the right to be silent. If you just fail to answer questions they are permitted to use your silence against you.

        • The Nybbler says:

          If you invoke the right to a lawyer after they Mirandize you without invoking the right to remain silent, you end up at the green box in the lower right: “Well done, they CANNOT use your silence against you. They’re never going to bring in a lawyer during interrogation, so just KEEP YOUR MOUTH SHUT & GOOD LUCK!” But you do have to ask for the lawyer.

          If you explicitly refuse to answer questions AND invoke your right to a lawyer, then in a custodial interrogation you don’t have to use the magic 5th amendment language, you end up in the white “CONGRATULATIONS” box in the lower left. Of course you’re still spending the night in a cell.

      • Brad says:

        Here’s the key language:

        “YOUR magic words: ‘I’m not answering any QUESTIONS and I want a LAWYER”.

        If the man in question had said “I ain’t answering no questions and I want a lawyer, dawg.” that would have been more than sufficient.

  12. Sniffnoy says:

    Reminds me of the recent discussion of Confucians vs. Legalists on whether people should be allowed to know laws: Georgia’s laws are available only if you pay for access; the state pursues people who publicize them for copyright violation. ACLU is on the case.

    Probably worth saying more here. So, for those who haven’t heard about this before — laws, being written by the government, are necessarily in the public domain, right? So what’s going on here?

    What’s going on here is that annotations on the law, which collate judicial rulings on the matter to clarify the law, are not public domain, since those are not written by the government but by, well, whoever writes the annotations, which is not generally the government. And Georgia has declared one particular set of annotations to the law to be official; effectively making it the law… even though it’s copyrighted.

    Yes, this is messed-up.

    • AlphaGamma says:

      I have seen similar arguments where the law requires certain technical standards (set by private organisations who copyright them) to be followed.

      • CatCube says:

        Most building codes are like this. So far as I know, American Institute of Steel Construction, the American Concrete Institute, the International Code Council, the American Society of Civil Engineers and the National Fire Protection Association get no money from the state and local governments that implement their model codes as law. And I was tracking that once implemented, the codes themselves are in the public domain (though still difficult to get your hands on a PDF for free, except for AISC which has the steel design codes on their website.

        One wrinkle that I’m not sure of is how codes incorporated by reference interact with copyright. For example, Oregon implements the International Building Code from the ICC as statute law, though they make some amendments to the model code. This therefore means it’s no longer under copyright (and you can find an extremely crippled PDF version on the State’s website, and the ICC’s website has a version that is only in a JavaScript viewer my browser can’t use). However, the remainder of the model codes are not implemented themselves; they’re simply enacted by the IBC as (e.g.) “Structural concrete shall be designed and constructed in accordance with the requirements of this chapter and ACI 318 as amended in Section 1905 of this code.” I don’t know if that then technically puts ACI 318 into the public domain.

        Where these organizations make money is in two ways: 1) memberships (though AISC is a little strange in that their primary memberships are companies that make steel, rather than individuals in the construction and engineering industry) 2) selling hard and soft copies of the codes and associated manuals. AISC, for example, gives the code away for free, but sells the extremely useful AISC Steel Manual, which has a hard copy of the code in the back and a bunch of extremely useful design tables and commentary on design procedures and details forming most of the book.

        So, basically, the government farms out most of the cost of writing the codes to these organizations with the expectation that they will actually make up the money through selling the code to professionals and companies that need to use it.

        This does save the cost of maintaining organizations that do code writing at state and local levels, and makes building components cheaper by not having wide variations in requirements between states that manufacturers will have to have a whole bunch of different products to ensure compliance nationwide. It’s also easier for those of us in the engineering business because we don’t have to worry about massive differences in acceptable design methods.

        The big downside is that this makes the process extremely opaque to the general public. Did you know about the massive controversy a few years back about classroom barricade devices? I actually agree with prohibiting classroom barricades for the reasons stated, but the language used by the code development organizations and fire marshals is still one of the most infuriating things I’ve ever read. Ohio school safety policy should be barricaded from politics. Really? REALLY? You’re proposing legal requirements that will cost public schools 850 motherfucking dollars per door to implement and you have the unmitigated gall to think that the elected officials representing the citizens of the state–WHO HAVE TO PAY FOR THIS–should stay out of it? The enforcement officials and code development officials who spend all of their day thinking about these issues should be carefully listened to by elected officials–that’s only common sense–but for those people to basically say “How dare you think about overruling us?” to state senators? Fuck off.

    • LHN says:

      Federal law is in the public domain. Whether state law is has some murky aspects, and many states or the publishers they subcontract to have asserted copyright in their laws, or the codified arrangement thereof. (So feel free to publish bills as they passed, but not the codified version that arranges them by subject that everyone cites to.)

      http://www.ncsl.org/documents/lsss/Copyright_Statutes.pdf

    • Watchman says:

      As far as I know, most western law is still envisaged in the pre-internet era, which means that it is regarded as available if it can be consulted at the library. I suspect one can in fact consult the laws of Georgia at a library, so from that point of view this is a non-problem.

      Personally I’d support a principle that anything incorporated into law becomes public domain, with the copyright holder compensated by the relevant government for the usurption of copyright if they had not submitted the relevant material for inclusion in law themselves. This might even encourage the development of model laws available to governments to buy, whilst meaning that government could take things from campaign groups and institutes seeking to have their standards imposed but that those seeking to influence government directly do not get rewarded for this.

      • random_eddie says:

        One can indeed consult the laws of Georgia at a library, as many of them (certainly any law library, and most large general-purpose libraries) will have a copy of the OCGA.

        However, you can also consult the laws of Georgia from the comfort of your own computer. The laws of Georgia are published online by legal publishing giant LexisNexis and are available to the public free of charge at https://www.lexisnexis.com/hottopics/gacode/

        That site sucks. It’s an awful, awful way to publish, well, ANYTHING, but particularly a codified body of law. Fortunately, because the laws of Georgia are not copyrighted, they have ALSO been published in a MUCH nicer format by the legal marketing company Justia at https://law.justia.com/codes/georgia/ . As with LexisNexis, Justia makes the laws available online to the public free of charge, although they do run ads for attorneys in the sidebar.

        If anyone feels horribly offended by the awful Lexis interface (as I do) and the Justia sidebar ads (as I do not), they are completely free to publish the laws of Georgia, including their official codification, copied directly from the Official Code of Georgia Annotated, and may do so in any manner as they may see fit. They can do this because the laws are not copyrighted. They are in the public domain.

        The annotations in the Official Code of Georgia Annotated are NOT the law, ARE copyrighted, and may not be published without the consent of the copyright holder – which, unusually among states, in this case happens to be the State itself.

    • random_eddie says:

      And Georgia has declared one particular set of annotations to the law to be official; effectively making it the law…

      No. The annotations are not the law. The law is the law and it is not copyrighted.

      The annotations in the Official Code of Georgia Annotated have no more force of law than the annotations in any other annotated compendium of law.

      Georgia has not declared one particular set of annotations to be official. They have declared that one particular codification of the enacted law is official, and they have contracted with one particular publisher to compile that codification and to publish it in annotated form, with the annotations produced by the publisher.

      If the annotations had been “made the law”, then not only would “the law” be copyrighted, “the law” would also have been written by a private party rather than enacted by a legislature and signed by the Governor. Happily, this absurd state of affairs is NOT the case.

      The only thing “messed up” about this situation is the common perception of the situation, which is a) wrong, b) ill-informed, and c) driven by media pieces pushed by the ACLU (recently) and by Carl Malamud (two years ago).

  13. K says:

    Regarding naltrexone, it could be a statistical power problem, but it’s definitely an intention to treat problem – the people who didn’t initiate either therapy should have been included in the analysis, which would presumably have raised the relapse rate in the naltrexone group higher than the subox group.

    Edit: sure enough, from the abstract, “Among all participants who were randomly assigned (intention-to-treat population, n=570) 24 week relapse events were greater for XR-NTX (185 [65%] of 283) than for BUP-NX (163 [57%] of 287; hazard ratio [HR] 1·36, 95% CI 1·10–1·68), most or all of this difference accounted for by early relapse in nearly all (70 [89%] of 79) XR-NTX induction failures. Among participants successfully inducted (per-protocol population, n=474), 24 week relapse events were similar across study groups (p=0·44).”

    So the study is fine, it’s just dodgy reporting.

    • Dog says:

      You’re right, but I still find the similar effectiveness for people who make it through induction somewhat surprising. I would expect lower relapse rates period on an agonist, even comparing people who successfully initiate the treatments. Other things that might be involved:

      1) This is not really apples to apples, in that the naltrexone is extended release injections given every 28 days, while the buprenorphine must be taken daily. The fundamental problem with treatments like this is compliance, and every time the patient has to consciously choose to comply there is another opportunity for failure. I would expect a depot formulation of the buprenorphine to be a significant improvement?

      2) Relapse while still on the naltrexone is going to be pointless since it completely blocks opiate effects, and it takes a long time to wear off, so some of the relapse in this case could look like “patient stops showing up for the study and waits for the naltrexone to wash out” instead of them actually reporting a relapse. And in fact the naltrexone group has 283 initially with 78 lost to early termination, compared to 287 and 62 in the buprenorphine group. Not a huge difference but still noticeable.

  14. Sniffnoy says:

    More in interesting links: Ketamine lifts mice’s mood only if administered by male researchers. (Just one of a number of “mice react differently depending on the sex of the experimenter — more specifically, based on their smell” results. Here, if the mice were place in fume hoods, the experimenter effect went away; the ketamine no longer worked for any of the mice. But putting them in a fume hood along with a T-shirt that had been worn by a man, then it went away in the other direction — it worked for all of them.)

  15. Alex Zavoluk says:

    The tax on graduate student “income” that isn’t really income is stupid, but basic economics predicts that the actual incidence of the tax depends on the supply and demand curves for graduate education. If grad students have pretty much no ability to pay the tax, you would expect an exodus of students from the market, but if the number of qualified students drops too low, universities will take up the tax burden to keep them around (whether by paying them more, or dropping the nominal tuition fee and eating the cost of taking fewer fees out of their grants, or whatever).

    • Jaskologist says:

      Isn’t it really income, though? I am not a tax lawyer, but I do know that there are all kinds of tax laws making sure that if my employer provides with any sort of benefits that might count as monetary, I’ll get taxed. Cursory research indicates that this includes tuition reimbursements over $5,250. Why should this be any different for university employees?

      • mostlyharmles says:

        Technically, no. The part that currently doesn’t get taxed is tuition reduction, which officially doesn’t count as income. Also, tuition assistance from an employer (which is different, but seems to be what you’re referring to) is tax deductible if the education “meets the express requirements of the individual’s employer… as a condition to the individual’s retaining an established employment”. This suggests that even if universities gave tuition assistance to their RAs and TAs instead of tuition reduction, under current law the students would still not be taxed because being enrolled in the PhD program is a job requirement. (I am not a tax lawyer, just a concerned grad student who has done some research. The quote is from https://www.thetaxadviser.com/issues/2012/aug/clinic-story-10.html)

        On a less technical note, I’m in a PhD program in a STEM field, and I only know one person who would have gone to grad school if they had to pay their own tuition (that person is independently wealthy). Furthermore, if the house tax plan passes then I will not be able to afford rent without taking out a loan. Grad school is a job, and a low paying one at that. If anyone thinks that a PhD student has a real annual income of $60-70,000, they are delusional.

        • outis says:

          If anyone thinks that a PhD student has a real annual income of $60-70,000, they are delusional.

          But they do, it’s just that they waste most of it on inflated tuition. The fact that they (and, more importantly, their professors) are completely insulated from the insanity of tuition costs is actually a problem. It does for them what guaranteed student loans do for naive undergrads.

      • skef says:

        When the amounts in question almost never actually change hands, that’s a good if not perfect indicator of an accounting fiction.

        The problem here is sorting out what “benefits” should “count as monetary”. If it’s just what happens to have a bullshit dollar amount slapped onto it, then slapping a different bullshit dollar amount onto it, like zero, should fix the problem.

        If the idea is what someone could slap a bullshit dollar amount on, then good lord. Should apprentices be taxed on the skills that they gain? Should raises be monitored and employees be taxed on the value of skills gained on the job?

        • rdplatypus says:

          Loan forgiveness is taxable income too. Isn’t that similar to tuition waivers? They at least rhyme.

        • Jaskologist says:

          When an accounting fiction is taking place, that’s a good indicator that a tax dodge is taking place.

          Frankly, it sounds to me like the universities are basically paying you guys in company script, to avoid having to treat you like real employees. I’ve noticed this a lot in grad student complaints about universities: they seem to have recreated a lot of the worst bits of capitalism from the 19th century, bits which the rest of us discarded long ago. But then you guys defend the people who are screwing you over.

          If I would get taxed for $20k in “tuition reduction” (and I’m pretty sure I would), why should university students be treated better? Why should the universities keep a loophole which most certainly benefits the privileged instead of doing their part in keep tax revenues up?

          • It seems to me that part of the objection people is raising is that they don’t really believe a dollar of tuition is buying a dollar’s worth of education. If it is, then treating a grant for tuition as income seems perfectly reasonable.

            I don’t think anyone has discussed the implication of the same logic for undergraduate scholarships. If we assume that the education is worth the same amount to all students, some of whom pay full freight, then scholarships look like income, or at least gifts. If we assume, instead, that scholarships are really price discrimination, charging a lower price to customers to whom the product is not worth the higher price, then they are neither income nor gifts and shouldn’t be taxable.

            Consider the case of a law school that gives a full tuition scholarship to a student not because he is needy but because he has a high LSAT and GPA. The school does it in order to raise its rating by getting a high bar pass rate. It also hopes that a talented student will end up as a rich and generous alumnus. That looks rather like hiring the student for the benefit of the school.

          • Jaskologist says:

            I hadn’t considered the application to undergrad scholarships; that’s a good point.

            I’m trying to figure out why my intuition considers those different. I think it boils down to whether or not the university expects you do perform labor in exchange for the scholarship/tuition reduction. If so, that’s income even if they structure it weird, and it’s past time to close the loophole.

            Otherwise, they look to me more like the usual price discrimination, of which I believe coupons are the classic example.

            (This lead me down a rabbit hole of research in which I learned that there are tax implications for sales which are advertised “By One Get One Free” as opposed to “Two for the Price of One!” which I now share with you.)

        • outis says:

          What if the universities put a reasonable price on grad school tuition, such that the grad students accepted it as a sensible estimate of the value of the education they’re getting? Let’s say it’s $5k, or $10k, or whatever.

          Well, then the universities would have to explain why they charge five times as much for undergraduate education, which if anything ought to cost less…

      • theory says:

        Like most of the tax reforms proposed by Congress (e.g., elimination of state/local tax deductions, reducing mortgage deduction, etc.), it has a solid logical theoretical basis.

        The reason people are mad is because the bill seems to specifically slanted to removing illogical benefits that favor traditionally left-leaning taxpayers (grad students, New Yorkers, etc.), while ignoring and doing nothing about the illogical benefits that favor traditionally right-leaning taxpayers (LLC passthru taxation, estate tax, etc.). The fact that the proposed reform is eliminating estate taxes WHILE preserving basis-step-up-at-death is the most potent example of such absurdity.

        • Brad says:

          The fact that the proposed reform is eliminating estate taxes WHILE preserving basis-step-up-at-death is the most potent example of such absurdity.

          I’m surprised this is as obscure as it seems to be. I’ve been bringing it up and almost universally getting blank stares. People talk about Peter Thiel’s Roth IRA — and with good reason. But this amounts to the same thing for every wealthy family.

          Instead of the double taxation people semi-accurately complain about we will have zero times taxation.

          • Iain says:

            Can you elaborate? I can’t find a good explanation online.

          • Brad says:

            Say 30 years ago your grandfather bought $50,000 worth of apple stock. Maybe he had just got an inheritance from his grandfather. For simplicity’s sake let’s say he didn’t reinvest the dividends. Today his shares would be worth just over $6.6 million.

            If he sold those shares he’d owe long term capital gains on them:
            $6.6MM – $50k (basis) * .238 (LT cap gains rate) = $1.5MM in taxes

            He’d be left with $5.1 million in cash.

            If instead of selling them he were to die and pass them to you, and assuming he had no other assets, the estate would owe:

            $6.6MM – $5.49 million (estate tax exclusion) * (graduated estate tax rates topping out at 40%) = $444,000

            leaving you with $6.1 million worth of apple stock. Importantly under currently law your basis in this apple stock would be the value of the stock on the day he died. So if you immediately turned around and sold that stock you’d get to keep the full $6.1 million worth of cash.

            The proposed law eliminates the estate tax, so no estates have to pay the 40% rate, but it keeps the step up basis so no one has to pay the 23.8% rate either. That massive capital gain just disappears as far as the IRS is concerned.

            As it stands the step-up basis at death rule is a massive loophole for every estate that isn’t subject to the estate tax (i.e. worth less than $5.49 million for an individual or $10.98 million for a couple) but for estate over that amount it makes some sense in that they are already paying high taxes on those amounts.

            If there was a deal to eliminate the estate tax in exchange for eliminating the step up basis I’d think the Democrats should think long and hard about accepting it. But eliminating the estate tax while keeping the step up basis is an insane giveaway.

          • Iain says:

            Thanks.

          • David Speyer says:

            Thank you for the explanation. Could I also make an editing suggestion? Your formulas could use parentheses:

            ($6.6MM – $50k (basis)) * .238 (LT cap gains rate)

            ($6.6MM – $5.49 million (estate tax exclusion)) * (graduated estate tax rates topping out at 40%)

          • Brad says:

            It’s long past the edit window, but your parenthesis are better than mine and I endorse them.

        • Nornagest says:

          If we get rid of some of the absurdity that benefits left-leaning constituencies this administration, and we get rid of some of the absurdity that benefits right-leaning constituencies next administration, maybe in twenty years or so we’ll have something resembling a sane tax code.

          I wouldn’t bet on it, though.

      • benwave says:

        My favourite solution to this problem – in the Netherlands, PhD students are treated not as students but as employees of the university. This makes a lot of sense, because that is how they actually act day to day. The supervisors set research goals, the students work on them, and they produce outcomes, from which the university benefits. It also simplifies all of the weird loophole nonsense immensely.

        It sounds from other comments as though in the USA there’s some weird tax code reason for universities to offer things the way the do in the first place though? So possibly not an immediate solution to you guys over there.

        • Svejk says:

          This system is found in much of Europe. Students are term-limited contracted employees of the university with statutory benefits and responsibilities. EU grants generally have line items to pay grad students’ salary, and costs of employment (health insurance, unemployment, etc.) are either enumerated according to a formula or attributable to the grant overhead (normally a competitive grant is around 10% overhead). Sometimes these costs are expected to be borne by the University/employer or taken as deductions from salary.

          European PhD programs tend to be shorter and involve less coursework (there is usually no inbuilt Master’s degree), and tuition fees are much lower, at least in continental Europe. Sometimes students choose their own project (with varying degrees of freedom), sometimes they apply to work on a specified project supported by a particular grant.

          • Watchman says:

            The UK system varies in that PhD students are normally paid for from a grant, which in its full form will include living costs. The University will be responsible for passing the living costs to the student, but this is part of their responsibilities under the grant arrangement and is not a form of salary. I suspect this is non-controversial because registered students do not pay income tax anyway.

            PhD students who are teaching are paid extra by the university, as that is not something the grant awarder wants (not that they don’t want the student to get the skills, but they’re just paying for the research). As we also have the shortest PhD requirement (three years, as opposed to the more common four and the US six plus) with a one year masters before if required, it does look like the rational response to this sort of legislation is to seek to fund PhDs in the UK (which is also the only educational system with a comparable ranking to the US, so there is no decline in standards).

          • AlphaGamma says:

            @Watchman on income tax: Registered students do pay income tax if their taxable income exceeds the personal tax allowance- though they are exempt from other taxes like council tax (for Americans: a local tax on residential property. A property occupied entirely by students is exempt). It’s just that a PhD student’s stipend does not count as taxable income.

            Money earned by a student from teaching or other work *is* taxable in the unlikely event that it exceeds the personal tax allowance (currently £11,500 pa). One of my labmates during my PhD was actually employed as a research assistant rather than receiving a student stipend because that was the only way our PI could fund her (this student is a Chinese citizen, a lot of stipends are restricted to EU). Her salary was set to a level such that after tax, she took home the same as those of us with stipends.

      • tmk says:

        Recent grads working at, say, a tech company is also gaining valuable experience and getting valuable mentoring from a supervisor and colleagues. Just look at how much higher salary someone with a few years of working experience can command compared to a fresh grad. Should employees pay tax on the monetary value of this mentoring?

        I can see a comparison to an undergrad working part-time to pay for tuition and does have to pay tax. But consider a computer science grad choosing between a PhD and taking a job at Google. Both will give them valuable knowledge, but only the PhD will be taxed as tuition.

        • John Schilling says:

          Should employees pay tax on the monetary value of this mentoring?

          Is there such a thing as a mentor-for-hire with an established market rate? I don’t think I’ve seen such a thing advertised anywhere, at least. Meanwhile, we have in many industries unpaid internships where mentoring is provided without any money changing hands and with at least in theory no work of value to the employer being done, which suggests that the established market rate for mentorship is zero.

          Income taxes are for several good reasons based on the market rate, not the value of a thing to its recipient.

          • David Speyer says:

            (This answer refers graduate school in STEM programs in the US, at fairly high ranked institutions, as that is what I know.) I don’t think there IS a market rate for graduate tution. Graduate schools try very hard to make sure that the student is never actually paying or taking on debt. Tuition is charged to outside grants if possible, or university funds if not, or paid off in labor which, if our sole job were to hire labor, could be bought much more cheaply. Actually putting a student in a position of paying tuition is a forceful way of communicating “you should leave by now”.

            The only comparable thing I can think of are individual hospital fees and prescription drug costs, which are set in order to be marked down for insurance companies/the federal government. And, of course, those are also their own special tax category.

            Moreover, a graduate education isn’t available for simple purchase. If you want to get a Ph. D., you have to apply; you can’t simply skip applying and write a check. From the faculty’s perspective, we are finding a cohort of apprentices we want to mentor, and which we will support as best we can, not hiring employees.

            The European system which benwave and others describe does seem a lot more logical, and maybe we should be heading there. But I hope this explains why the slogan of “tax graduate tuition at market value” doesn’t make sense.

          • John Schilling says:

            (This answer refers graduate school in STEM programs in the US, at fairly high ranked institutions, as that is what I know.)

            I concur with this assessment for engineering and the hard sciences at least, with the caveat that there may be a significant population of East Asian students whose parents are paying the nominal tuition for their favored sons to get STEM degrees from prestigious US universities. Other than that, there’s enough industry and government money to go around. I am in the position of being able to offer student loan forgiveness for anyone who comes to work for me, and find it to be of zero help in recruitment because everybody I want to hire is already debt-free.

            Liberal and Fine Arts, there are definitely people paying graduate tuition out of pocket. MBA programs as well, and Law and Medicine. I don’t know about the soft sciences and education. But I’d be reluctant to generalize from the STEM experience, because so much of that is being funded by industry/government players with deep pockets and a specific interest in either A: the student they are paying for coming to work for them when they graduate (if not sooner) or B: the targeted research that student will be doing along the way. The fields where the student is traditionally likely to hang out a shingle as a sole practitioner, that funding model isn’t there. And the fields where the major careers are inside academia, obviously need someone paying tuition for their books to balance.

          • quanta413 says:

            which, if our sole job were to hire labor, could be bought much more cheaply.

            I’m confused about how to parse this. If tuition is included as a cost to the university (which it isn’t in STEM unless the student is doing no research, although tuition for a student doing research does cost a PI), it makes sense. But the tuition isn’t really a cost to the university, I’m paid ~$20,000 a year and would make easily triple that or more for the same workload outside of the university. On average, I take one course every year or two (the rest is research credits). And as a bonus, the university gets an extra $50,000 from the NSF.

            If I only count students who are TA’ing the statement roughly makes sense, the university is paying ~20,000-30,000 a year (including benefits) for part time labor and the student is probably consuming some fraction of resources taking a couple courses (but probably an order of magnitude less than $50,000). This isn’t bad, but it’s worse per hour in some cities than can be made by tutoring (but better than others). I have difficulty believing you can get much cheaper. Maybe you can shave off 10% of the cost or something. I think maybe you could get big savings, but it would require much more significant changes in how universities work.

          • David Speyer says:

            Disclaimer: Although I work at UMich, I am ignorant about many aspects of my university and there may be basic errors here.

            TLDR: Paying a lecturer versus paying a grad TA are roughly the same payment per course to the student, and thus roughly the same cost to the university. However, when we include the tuition cost, which goes from the department to university, a lecturer is much cheaper for the department.

            A UMich math grad student stipend is roughly $19K per year, not counting summer salary (citation). A UMich math Lecturer I not working in the summer makes $50-58K (citation). Grad students normally teach one course per term; lecturer’s generally teach three (estimate formed by searching names here).

            So it is roughly three times the pay for three times the “work”. (In quotes because a grad student’s main work is research.)

            Both jobs come with benefits. I don’t know how to estimate the impact of this — three grad students might use more healthcare than one lecturer, but then again, lecturers are older and more likely to have families.

            Graduate candidates pay $6K per term, so $12K per year. (Go here and click on “College of Literature, Science and the Arts”, “Rackham and Interdepartmental”.)

            Tuition for a grad TA comes out of the departmental budget and goes back into the university’s general fund. So, if the department just wanted to teach the courses, we would definitely save by hiring lecturers.

          • quanta413 says:

            @David Speyer

            Thanks for the response! That makes a lot of sense.

            I didn’t realize that the department had to shift money from its budget to the general budget for each TA it has. I thought that only happened for RAs when grant money was being collected. That obviously invalidates my mental model since staffing decisions are made at the department level.

          • David Speyer says:

            According to this estimator, for a single person in Michigan, a change in nominal gross income from $19K to $31K, without actually getting $12K more, means a change in take home pay from $16K to $13K.

    • onyomi says:

      Universities will pay the lawyer burden to figure out how to basically keep doing what they’re doing but call it something different/configure it differently.

    • John Schilling says:

      The tax on graduate student “income” that isn’t really income is stupid,

      No more stupid than, e.g., the tax on employer-provided meals and housing. The whole concept of an income tax becomes basically unworkable if employers are given free latitude to identify huge chunks of what their employees would otherwise be spending their wages on and saying, “look, we’ll buy this for you directly and the IRS won’t take a cut”. And maybe I’m kind of sympathetic to the idea of making the income tax unworkable, but I don’t describe the IRS or the administration as “stupid” for trying to put an end to that crap.

  16. Sfoil says:

    The insulin price article didn’t even speculate about why prices have risen so much. Regulatory barriers to market entry, I assume, but then what changed?

  17. Sfoil says:

    Scott Aaronson kind of needs to get a grip, but I agree that the law is probably intended to harm institutions of higher education. I don’t know why he’s surprised though; universities openly discriminate against the sort of people supporting the measure and more subjectively university representatives of all stripes treat them with contempt (Aaronson’s own article being an example, of course). If graduate student employees are so valuable then universities can start paying market rates for them. Maybe some colleges will have to face a choice between staying open and sending their diversity commissars off to find more useful means of employment, if so then good news.

    • Aapje says:

      Aaronson goes Hulk when anything threatens academia. There seems to be this general tendency in the US on both sides of the political to excuse discrimination and other unfair treatment of the outgroup when parts of the ingroup do it & see it as a sign of near apocalypse when parts of the outgroup do it.

    • Matt M says:

      Maybe some colleges will have to face a choice between staying open and sending their diversity commissars off to find more useful means of employment, if so then good news.

      No chance. They’d more willingly fire their entire physics faculty than give up their Title IX administrators.

      • Watchman says:

        To be fair that’s a rationale decision in most cases. The amount of money lost through non-compliance with Title IX is likely in any comprehensive university at least, and most science colleges, to be far greater than the income of the physics faculty.

        It is not universities that cause these stupid posts to proliferate, but (mainly) the requirements of government. They basically totally distort the market to the point that a bureaucrat has more value than a marketeer or an academic. And in relation to the overheads conversation above, the bureaucrat who ensures compliance with Title IX has to be funded from other activities because Title IX does not bring funds, just entitles the college to existing funding. And Title IX is only one of a number of risks that need addressing. And we’re in cost disease territory again…

        • Matt M says:

          As I’ve said in multiple OTs, I disagree with this interpretation. This is the story the universities tell for the purposes of plausible deniability whenever they get negative publicity for doing something egregiously unjust. But the reality is that the crazy Title IX stuff is not some onerous government requirement they grudgingly comply with against their will. This is stuff they enthusiastically embraced. This is everything they’ve always wanted to do anyway.

          And we can verify this by looking at their reaction to Betsy Devos’ announcement that she’s rolling back the worst parts of it. Were they relieved? Did they express enjoyment and approval? Announce that they’re super excited to be able to dedicate less resources to bureaucratic enforcement that can now be applied towards better educating students?

          Absolutely not. They called her names, denounced her position, declared their intention to operate as per the old guidelines, and in some cases threatened to file a lawsuit to ensure the old interpretations continued to be enforced.

          • Alphonse says:

            I think your point makes even more sense when shifting down a level of abstraction — the “university” isn’t really an entity that exists; it’s all just people. And the people who write the statements expressing the university’s position regarding the rollback of Title IX requirements are disproportionately likely to be the “experts” on Title IX requirements . . . i.e. the people whose jobs might disappear if the reach of Title IX actually shrank meaningfully.

            Once the Title IX administrators exist, the principal agent problem ensures that the “university” will oppose rollback of Title IX requirements, even if that change would actually benefit the “university,” since it would hurt the individuals in charge of creating the university’s response to such policy changes.

            (Obviously this applies equally to other regulatory regimes, not just Title IX. Repeat this a few times, stir with some endlessly available federal student loans, and cook for a few decades to get $50k/year tuition.)

      • Kevin C. says:

        No chance. They’d more willingly fire their entire physics faculty than give up their Title IX administrators.

        I think that Bisexual Asian Studies should have its own building. The question is who goes? The math department or the hockey team? … I think hockey, call me about it.

        —Port Chester University President Garcia-Thompson (Jessica Walter), PCU (1994)

  18. jhertzlinger says:

    The tax on grad student salaries might be a way to crack down on a Ponzi scheme.

  19. drethelin says:

    So do I need to keep giving to MIRI or should I try to find something else now that they have a good solid hunk of cash?

    • Rob Bensinger says:

      Malo mentioned on the MIRI Blog that we’re planning to talk more about this at the start of our December fundraiser. Short answer is that we’re in a similar position to 2013, our last big windfall year: we have more runway than usual at the moment, but we’re also gearing up to make a large number of hires over the coming months/years, especially engineers.

  20. onyomi says:

    I’m surprised people oppose embryo selection when would-be mothers looking to patronize a sperm bank want all kinds of information about the potential father. How is being picky about which particular sperm and egg you bring to term among you and your partners’ own sperm and eggs any different from being picky about the donor of your sperm? That is, you don’t prefer the luck of the draw in picking a partner or sperm donor, why would you prefer the luck of the draw when it came to your own sperm or egg?

    • Murphy says:

      I think people view the former as more “natural” or it’s closer to things that are currently happening since it’s comparable to simply finding a partner with X traits you like.

      If, by some cultural happenstance, it was the case that every day the people of new york gathered in time square to sacrifice an infant to the rain gods I have a strong feeling most people would find ways to argue that it’s not bad because “it’s something we’re already doing and anything that implies everyone is doing something bad must be wrong hence this is ok.” or similar

      Since they’re not consciously selecting sperm already it gets the full brunt of “but that’s eugenics” while merely selecting an attractive astronaut to be the dad gets a free pass on the eugenics front.

      • onyomi says:

        It’s really pretty insane, in general, that “eugenics” has become a scare word for a species that has, throughout its history, consciously selected plants, animals, and mates for desirable characteristics.

        • johansenindustries says:

          There are an awful lot of things that we do to livestock and plants, that we wouldn’t be OK with people forcing onto us. That doesn’t strike me as insane at all.

          • Murphy says:

            Don’t forget the “and mates” bit.

            the thing is: shopping around for an attractive mate or sperm donor:

            http://www.smbc-comics.com/comics/1469199447-20160722.png

            Gets a free pass.

            While parents freely choosing to shop around for the sperm and egg carrying the genes they most want to pass on gets immediately compared to Nazis even though it has basically nothing in common with anything nazi-ish except for scraping in under the definition of eugenics right next to shopping around at the sperm bank.

            Someone picking a sperm donor with perfect pitch in the hope of having a child more likely to have perfect pitch: Fine and dandy.

            Someone picking from among their own individual sperm in the hope of having a child more likely to have perfect pitch: basically hitler.

          • Lambert says:

            It’s not even about perfect pitch.
            It’s about wanting their child to have a happy, normal, long, not incredibly painful life.

          • Doctor Mist says:

            we wouldn’t be OK with people forcing onto us

            Who said anything about force?

    • gwern says:

      If you read articles about sperm banks, you’ll see that people do virtue-signal hard about how offensive and eugenics it is. The great innovation of modern sperm banks is to provide the eugenics substance without the wrapper; The Genius Factory argues that this was the biggest legacy of the Nobel sperm bank – providing detailed profiles and medical information about the sperm donors rather than just the previous paternalistic practice of providing sperm with no information other than a unique ID number. It failed for several reasons, including backlash for being explicit about eugenics, but its revolution was immediately picked up by all succeeding sperm banks. Of course, they never say that any of it is genetic but mothers aren’t stupid, and so yes, they use the information the sperm bank just happens to provide and just happen to pick the sperm donors who have the most education, are tallest, and sound the most sane in sample interviews, etc. Some sperm banks will even charge for access to more detailed dossiers about specific donors; that’s quite amusing. (This is why I say that probably the most important problem in embryo selection is figuring out how to cater to Homo hypocritus.)

      An example of the virtue-signaling: https://www.washingtonpost.com/national/health-science/donor-eggs-sperm-banks-and-the-quest-for-good-genes/2017/10/21/64b9bdd0-aaa6-11e7-b3aa-c0e2e1d41e38_story.html

      “It’s a little unsettling to be marketing characteristics as potentially positive in a future child,” said Rebecca Dresser, a bioethicist at Washington University in St. Louis and a member of the President’s Council on Bioethics under George W. Bush. “But it’s hard to think on what basis to prohibit that.” And so, Dresser said, “what we have now is prospective parents making judgments about what they think ‘good’ genes are” — decisions that are literally changing the face of the next generation.

      What a pity. She can’t think of a way to criminalize it. Perhaps we need more funding for ‘bioethics’ so they can finally come up with a pretext to ban mate selection rather than having sex at random; for the greater good, you know.

      • Anon. says:

        >a bioethicist

        Of course.

      • Deiseach says:

        What a pity. She can’t think of a way to criminalize it.

        Ah, come off it; there’s a reasonable interpretation there which is “We know fuck-all about the real effects when it comes to a living baby instead of ‘in theory X should mean Y’, and if parents are choosing these methods to have ‘smart, pretty, tall, athletic kid’ they may end up with some or even all of that but also ‘has asthma/rare degenerative disease/weird new problem not previously seen/hi mom I’m a serial killer’ side-effects, then it will be “why didn’t somebody stop us???”

        Best case scenario is that you end up with a healthy baby despite the selection processes because they’re peddling more than they can deliver and your kid turns out the same as they would have turned out if you’d stuck to getting pregnant the old-fashioned way, because they can’t give you the smart, attractive, athletic, tall baby they promised.

        Worst case scenario? “Hello, Mr and Mrs Jones, the amniocentesis results are back and well….” discussion that ends up with “termination of pregnancy or delivery of a child that can’t live past birth or if it does survive is massively handicapped?” choices. And if there are enough worst case scenarios instead of best case ones, you bet the same public which wanted the choice to shop for a super baby will be screaming about “the government should have done something to prevent this!”

        • Nabil ad Dajjal says:

          IVF and preimplantation genetic diagnosis (PGD) aren’t risk free but you’re blowing this way out of proportion.

          The perinatal death rate for IVF is higher and birth weight is slightly lower, but the children who survive to birth are otherwise perfectly healthy. Many of them are now adults and the biggest difference that’s been identified is that those conceived through IVF are a bit taller than average.

          PGD is newer but not that much newer: it’s been around since 1991 and so there’s a fair amount of data available on risks. Biopsying and freezing embryos destroys about 20% of them, but the kids that are born are indistinguishable from other children except for their birth weight.

          Being able to avoid monogenic diseases like cystic fibrosis and aneuploidies like Down’s syndrome is well worth that risk. Screening for polygenic disorders like ASD, or for polygenic traits like height and intelligence, is more difficult but potentially has enormous benefits as well.

        • gwern says:

          there’s a reasonable interpretation there

          No, there isn’t. Especially not for a card-carrying bioethicist quoted as a critic in an article whose entire tone is that this is a bad thing. There just is not. To think there is requires desperate evasions and denial of both connotation and denotation and a willful forgetting of everything one knows about the English language. You could not ask for a more clear expression of a prewritten bottom line than a quote like “But it’s hard to think on what basis to prohibit that.” This is why it’s such a striking line: it’s one of the most naked admissions of prejudice I’ve seen in a long time without a figleaf of cost-benefit reasoning – it doesn’t matter that she has no evidence of harm, she already knows the desired conclusion and the question is merely what pretext they can come up with to ban or regulate it out of existence.

          • Toby Bartels says:

            FWIW, when I read that quote, I immediately interpreted it to mean when I would expect an academic to mean by it, rather than what I would expect a politician to mean by it (I know she has been both): that it’s hard to think of a reason why it should be prohibited, so it probably should not be prohibited, that is, it should be allowed. That seems to differ from everybody else’s interpretation.

          • Deiseach says:

            Right, first I should clear up that you were talking about sperm banks and I didn’t address that, I was talking about the proposed gene selection/embryo selection/what have you putative techniques.

            As for sperm donation and sperm banks, hasn’t there been discussion about “regression to the mean” on here? The donor fathers may all be excellent and superior specimens, but if the mothers are average, that means the resulting offspring are less likely to be as impressive in the desirable traits as their fathers. And even if the mothers are equally exceptional, regression to the mean means that the kids are likely not to be “Mom and Dad both 150 IQ, therefore Junior 150 IQ”. So it’s still as much of a gamble as racehorse breeding, which despite all the refinements cannot absolutely guarantee “yes this stud and this mare will give you a Gold Cup winner”.

            Now, on to the methods that are causing me more concern if touted as “gonna give great results, guaranteed!” Embryo selection is still on a very crude basis; going for “select optimum sperm and optimum ovum and then select out optimum embryo for implantation to result in guaranteed tall, attractive, smart, athletic, socially gregarious baby” is not going to work for the first X number of cases.

            Now, if we’re talking about breeding and discarding lab mice until we get it right, that’s one thing, but we’re talking about human infants who may not exhibit “uhhh – that’s not something we were expecting” until after birth and into infancy/toddlerhood/early childhood, at which time you can’t simply just euthanise them and start again.

            If you can guarantee that this will be run on the basis of “the first thousand couples signing up to use this procedure all realise, understand and accept that ‘past performance is not a guide to future returns’ and just because it worked on mice doesn’t scale up to humans, and if the kids turn out not to be the super-geniuses or super-athletes expected (never mind if it turns out that some kind of negative effect accompanies high IQ or other high performance genes) they will not immediately run to the courts to sue”, then fire ahead with your mass social experiment. But I don’t think real world people, especially those wanting the future Nobel Winner or Major League Superstar Payday Winner so badly they engage in something like this, are going to behave like that.

            And I’ll admit, I also have secondary and unrelated directly to the possible procedures themselves worries, and it’s not about eugenics: it’s about the way those children will be raised. I’m afraid of the Stage Mother phenomenon, of the ‘our love is conditional on you continuing to get the highest possible grades for the next twenty or so years of your life, and one bad test result will mean criticism, anger, and clear disapproval of you as failing to perform to the potential you were engineered to have’ outcome.

            If the parents will love and support their children the same if they turn out to want to be artists, musicians, or plumbers instead of tennis stars or Nobel winners, that is great – but I do worry.

            And we’ve previously discussed the Polgar Method on here, and how the originator claimed any average child could be steered towards being a high achieving prodigy – but somehow it does not seem, even if it’s an easy and clear method, to have achieved widespread use as an everyday technique. He was able to back it up by producing three high achieving chess playing daughters, but even with that empirical evidence, the Polgar Genius routine is not in common use with ambitious parents striving to improve their children’s chances in life.

            So I think the attempts are always going to be oversold as “this can make a HUGE and DEMONSTRABLY PROVEN difference!” but will always fall a little short, and the risk is that the falling short will result in human misery. I’m old enough to remember the publicity about Doron Blake and at the time the mother struck me as – to be frank – a fruitcake. Visibly ambitious to vicariously achieve through her offspring, I did wonder what kind of a life the misfortunate child would have. He seems to have turned out relatively okay, and did achieve, but has not turned into the World Changing Household Name Superstar his mother seemed to be aiming for. In fact, I would say that the fact he turned out to be as normal as could be expected after all the publicity and brouhaha and his mother’s flaky child-rearing methods is a great thing and not a testimony to “see, this has no downsides at all!”

          • Murphy says:

            @Deiseach

            “regression to the mean” does not mean you get no gains at all, merely that they’re not just a straight average of the parents.

            Life is easier if you’re smart, healthy and pretty. If you can find an extremely smart healthy and pretty sperm donor, even if the gains are only small your child gets to live with the difficulty slider nudged down a few points for pretty much whatever they want to do in life.

            Ultimately every change ever involves some kind of risk. the first people using sperm banks at all were taking some steps into the unknown, the first people using IVF similar. Decades later it’s just another boring way to have kids. There’s nothing particularly special about this.

            before they ever use this kind of stuff on humans it’s going to have been tested a hell of a lot on many species of animals and the early versions will involve selecting against polygenic genetics diseases and such. At some point you reach a stage where the method is solid and all you’re doing is selecting on a different trait and little enough risk for the kids involved that the potential benefits to them are likely greater.

          • Aapje says:

            @Deiseach

            The mothers obviously don’t want the best kid in general, they want the best genes to mix with theirs. They aren’t optimizing for society, but for themselves.

            As for “regression to the mean,” this just means that the higher the IQ of a person is, the higher the contribution of non-genetic causes is to that IQ on average. However, it also true that the higher the IQ of a person is, the higher their genetic IQ will be on average. So it’s still a good idea to pick the IQ=150 guy over the IQ=120 guy, even if the average genetic gap between people with those IQs is less than 30.

        • Yosarian2 says:

          We’re just talking about choosing one embryo (with PGD and IVF) or choosing what healthy father you want to have your child from with sperm donation. There is no risk here that isn’t already present in any other case where someone has a child. This is just choosing one of the children you might have had anyway, and there is no reason to think you are increasing other risks. (In fact, it’s more likely you are lowering other risks.)

          If we were talking about using CRISPR to add de nova genes into embryos that had never been seen in humans before your concern would make sense. But here, it really doesn’t.

        • The Nybbler says:

          Best case scenario is that you end up with a healthy baby despite the selection processes because they’re peddling more than they can deliver and your kid turns out the same as they would have turned out if you’d stuck to getting pregnant the old-fashioned way, because they can’t give you the smart, attractive, athletic, tall baby they promised.

          No, the best case scenario is that they DO get the smart, attractive, athletic, tall child they promised. Of course this is the same as the chance of getting that if they’d gotten pregnant the old fashioned way _with the same donor_, who probably wasn’t available.

          Worst case scenario is they don’t have enough money, compromise on “attractive and tall”, and end up with a smart, athletic, cynical internet troll. But I’ve never made a deposit so they’re reasonably safe 🙂

    • MugaSofer says:

      Do you really have trouble understanding why people would be more squeamish about “let’s kill all the sub-par embryos” than “let’s be careful to only make high-quality embryos”?

      (Of course, all IVF includes abortion, but they’re usually able to de-emphasize that when selling it. With embryo selection the abortion is the primary “benefit” offered.)

    • INH5 says:

      I’m surprised people oppose embryo selection when would-be mothers looking to patronize a sperm bank want all kinds of information about the potential father.

      The key words there are “looking to patronize a sperm bank.” Most single mothers become pregnant using far cheaper and less selective methods of obtaining sperm. It is, in fact, not unknown for single mothers to be uncertain about who the fathers of their children are.

      There are somewhere around 40,000 donor-conceived babies reported born every year in the US. Even if we assume that there are another 60,000 unreported annual births, that’s still a tiny fraction of the 1.6 million babies born to unmarried mothers every year.

  21. honhonhonhon says:

    Could somebody with a tumblr account pastebin the jaw article?

  22. Douglas Knight says:

    The tax change would tax (vesting of) stock, but do startups give out RSUs? I thought that they gave out options, instead. (which probably wouldn’t help against this tax)

    I’ve been meaning to ask: why do startups all give options, not stock? Is it for this exact reason, that the stock would be immediately taxed at the current valuation, while the option is nominally worthless because it requires the payment of the current valuation?

    • Brandon Berg says:

      I suspect that it’s because there tends to be fairly high turnover. Many employees will leave without exercising the options. When this happens, they can give the options to another employee. If employees were simply given stock, many would hold on to it after leaving, diluting the ownership of the investors and remaining employees.

      Also, if employees do exercise the options, the start-up gets some cash, which they usually need pretty badly.

    • actinide meta says:

      Yes. Once upon a time, the tax treatment was even more favorable: basically companies could give out options with almost zero strike price, economically equivalent to stock, without creating any immediate tax liability. It’s not hard to see how a form of compensation with indefinitely deferred taxation became popular. Newer rules require the strike price to be market value, but due to the illiquidity and different share classes typically small startups can still get away with prices that are much lower than investors are paying for shares. For big liquid companies this falls apart, and so they have all moved to using RSUs for compensation.

      The last I had heard this part of the tax bill was probably dead, but my guess is that it was never a good cause for panic, because clever tax lawyers would find a way to adapt comp structures that wouldn’t be too painful. The market is pretty good at taking lemons and making lemonade. See also: the tuition waiver issue; global warming.

    • The Nybbler says:

      I’m not really sure of the difference with RSUs. The change was described as making the taxable event vesting rather than delivering of shares, but when I got RSUs, those events were simultaneous and the company sold off some of my stock on vesting to cover the tax on the rest. What I found annoying is that while I had several grants which vested monthly, they claimed they had to cover the tax on these separately, which due to quantization meant more shares were sold off than necessary. I don’t believe the IRS actually required that, they just weren’t set up to do it the sensible way (sell as many total shares as necessary to cover the tax on all the shares vesting at a given time without regard to which grants they came from).

      With options, this change would be a disaster for holders of low strike price options; you’d be taxed on the difference between the strike price and the market value at the time of vesting, even if you didn’t exercise the option.

      • poipoipoi says:

        Pre-IPO startups.

        I’m at a Series E with about a hundred grand in stock options vesting over the next four years. Within 3-5, we’ll IPO (or at least we’d better), but until we do, I’m basically “making” an extra $30K/year that is in no way liquid, and which I owe an extra $15,000 in taxes on under this bill.

        If the startup *works*, that’s actually pretty great (albeit deeply painful in the short-term). If it doesn’t, I’m completely fucked. And more importantly, if I’m an early-stage employee who’s seen their stock go up 19x in value in the last 5 years… I don’t make enough money to pay the taxes on $700K in stock!

        The somewhat less obvious counterpoint is this:

        Hey, remember all of that above? Yeah, it’s completely fucked.

        The basic problem is that you take a deeply serious paycut in exchange for stock, then if you leave before the company IPO’s or sells out, you have to buy your stock or leave it behind. And in 2017, companies are taking 10 years to IPO even when they succeed, and average employee tenure is 2.

        So there’s really something to be said for trying to force a new compensation method, but at the same time, you’re screwing over every existing employee of these startups.

  23. C_B says:

    Not sure if you already know, Scott, but the sequel to Worm has started: https://www.parahumans.net/

    I found out about Worm from this blog, so might be worth an entry in a links post sometime to pay it forward.

  24. Dedicating Ruckus says:

    Related: an attempt to make neural nets more transparent by investigating what pictures maximally activate each neuron of an image classifier. There’s something very creepy about this, like dissecting the world along some mysterious dimension into incomprehensible conceptual primitives. Also, some neural net is very convinced that “either an animal face or a car body” is a fundamental concept that cleaves reality at its joints, and now I’m questioning how I know for sure that it isn’t.

    Neural networks don’t know anything about objects, they know only about images. Presumably there’s some statistical similarity between images of animal faces and images of car bodies that’s discoverable if you throw enough GPUs at it.

    It’s worth keeping in mind that image-classification neural nets are working fundamentally differently from a human visual cortex. They know nothing about referents, only about pixels. Human vision is constantly trying to interpret what it sees in accordance with its knowledge about the world; NNs just calculate elaborate algebra problems over RGB values.

    • Peter says:

      There’s a lot of work that that “just” in your last sentence is doing.

      The training system as a whole has knowledge of referents; it’s in the training data, it’s the class labels of the images the system classifies. The trained system as a whole, once separated from the training data, has some knowledge of referents, in that it knows how to correctly apply class labels to images. OK, not with complete accuracy, but humans aren’t completely accurate either. Picking bits of the trained system, very roughly the early layers are almost entirely about images, the later layers are increasingly about objects, but all are sort of a weird mixture.

      I work with neural networks for a living (I’ve got one training right now while I write this comment), and I don’t see a fundamental difference; humans are opaque where ANNs are transparent, and so people willing to do an equivalent of a gods-of-the-gaps argument have plenty of gaps to populate with gods.

      Anyway, the problem with carving nature at the joints is not that there are no joints but there are too many; in a sense, everything is a joint, but some joints are jointier than others. The things a trained network picks out are likely to be far jointier than the things “picked out” by the same network before training, but still some distance from the maximum possible jointiness.

      Getting machine learning systems to actually sieze upon the more jointy joints is a range of activities that ranges from a science to a black art. One common thing that needs hand optimization is the size of the network – how many layers, how many neurons in each layer. Too small, and the network can’t find the joints it needs to do a good job, too large and it greedily seizes on unjointy joints that fortuitously resemble irrelevant regularities in the training data, just right and it does a mix of both.

      I think being somewhat nervous about your intuitions (and other’s intutions) for how jointy joints are is a good thing; on the other hand the intuition that not all joints are equally jointy I think is sound; the postmodernists that reject it can safely be dismissed.

      • Peter says:

        Come to think of it, there is a more mundane sense in which Dedicating Ruckus might be right. GoogLeNet, and other major deep learning vision systems, are feedforward – when the trained net is being used, information only goes forward. It’s fairly easy to argue that human vision isn’t – that information from the late stages of vision can be fed back to inform subsequent iterations of the early stages – and so explicit knowledge of referents can be a part of the process in the way that it can’t be in feedforward networks.

        Example: the image in this page. Once you can see the spotty dog it’s hard to un-see.

      • Dedicating Ruckus says:

        The training system as a whole has knowledge of referents; it’s in the training data, it’s the class labels of the images the system classifies. The trained system as a whole, once separated from the training data, has some knowledge of referents, in that it knows how to correctly apply class labels to images. OK, not with complete accuracy, but humans aren’t completely accurate either. Picking bits of the trained system, very roughly the early layers are almost entirely about images, the later layers are increasingly about objects, but all are sort of a weird mixture.

        Training data with category tags doesn’t actually include “knowledge of referents”. Training data with category tags includes knowledge of which (opaque to the NN) category fits which (opaque to the NN) pixel statistical patterns.

        The human visual cortex is tied in with human instinctive physics knowledge and with a human’s entire body of knowledge about the world. The bit of brain that’s actually parsing images is constantly receiving back-feedback from these other systems about how it should interpret ambiguous data. By contrast, an image-recognition NN is just a statistical function from pixels -> category. The components of that function may or may not have anything to do with actual features of the referent categories, as we can see with the whole adversarial examples thing.

  25. manwhoisthursday says:

    Cordelia Fine tends to:

    1. give a righteous thrashing to the many bad studies which supposedly support biological causes for sex differences in behaviour.
    2. ignore the many more good studies showing biological causes for sex differences in behaviour.
    3. give a pass to other highly dubious studies (especially priming studies) which supposedly support social causes for sex differences in behaviours.
    4. give ridiculous explanations for why CAH women don’t have female typical behaviour despite their female appearance.

    Despite all of the above, she does at least cite studies and give substantial arguments for her position. There are a lot of bad studies and bad arguments for biological causes of sex differences in behaviour. Fine is merely wrong, not evil, and it is a useful intellectual exercise to read her books and figure out where they go wrong.

  26. harland0 says:

    Ah, I know the hair coloring thing. Anytime anyone colors their hair in colors usually found in sno cones, it’s a desperate attempt to gain attention. “LOOK AT ME!” it shouts. It is an attempt not to fit into society, but to stick out like a dick on a pumpkin. You have to understand attention-needing people, they require attention otherwise they will be miserable with themselves. Actually getting attention due to actions is too hard, so the quick and easy path is clown hair.

    • skef says:

      Some people want attention, some people are super-cranky — it’s all a beautiful tapestry.

    • Watchman says:

      Or people who are getting a lot of attention for a certain action (changing gender) are going to be a lot less concerned about the relatively mild reaction to something else (changing hair colour) so are more likely to do it. I never did anything too ridiculous with my hair due to concerns about social reactions (pity – I would have liked blue streaks) as it would generally be the most notable thing about me. If that did not apply, if I had transitioned gender or become a major political figure perhaps, then the relatively mild decision to have fun wiht my hair would be easier.

    • James C says:

      My initial thought was that, perhaps, those with low attachment to their physical form are more likely to change or alter their appearance by either dyes, tattoos or surgery. It is less a cry for attention and more a lower boundary for making changes.

    • Ozy Frantz says:

      Is there a name for the fallacy “I pay attention to the thing you’re doing, therefore you must be doing it for attention”? It’s a surprisingly common theory of mind error.

      Anyway, people dye their hair funky colors for the normal signalling reasons that they engage in any other grooming; it’s just that people who dye their hair funky colors have a different group than you, in which funky-colored hair gets compliments and expensive watches or suits or whatever do not.

      • actinide meta says:

        Even this is a little too cynical. Believe it or not, it’s occasionally possible for people to actually like something.

        • Anon. says:

          Quoting Robin Hanson,

          Many people (including me) claim that we eat food and drink water because without nutrition and fluids we would starve and dehydrate. Imagine this response:

          No, people eat food because they are hungry, and drink water because they are thirsty. We don’t need abstract concepts like nutrition and dehydration to explain something so elemental as following our authentic feelings and desires.

          Yes hunger and thirst are direct proximate causes of eating and drinking. But we are often interested in finding more distal explanations of such proximate causes. So almost no one objects to the nutrition and dehydration explanations of eating and drinking.

          However, one of the most common criticisms I get about signaling explanations of human behavior is that we are instead just following authentic feelings and desires.

          Yes, people don’t need to consciously force themselves to express opinions on many topics. That habit comes quite naturally. Even so, we might want to explain that habit in terms of more basic distal forces.

          “I’m just doing it because I like it” is not an explanation, it just kicks the can down the road to “why do you like it?”

          • AnonYEmous says:

            “I’m just doing it because I like it” is not an explanation, it just kicks the can down the road to “why do you like it?”

            Disagree; as an explanation, “I’m doing it because I like it” excludes purely cynical and self-interested reasons. That tells you that the person isn’t particularly cynical or self-interested as a general rule, even if their underlying desires mimic this in specific cases.

            I mean, look at the example you gave. Someone who, when asked why they drink water, responds with “so I don’t die of dehydration” is probably a behavioral match with Spock or something. Someone who doesn’t even think about it, or someone who does it “because they’re thirsty”, signals a different type of person. I’m going to go drink water right now, because I don’t like the feeling of dry skin in my throat. That should tell you something about me, at least.

          • Whatever Happened To Anonymous says:

            Would “because they find it aesthetically pleasing” good enough, or do we have to go deeper?

          • Anon. says:

            “I’m doing it because I like it” excludes purely cynical and self-interested reasons.

            It just requires imperfect introspective ability. People come up with all sorts of non-cynical explanations for things their extremely cynical unconscious does. There is absolutely no reason to believe them. (Again, see Hanson)

            Would “because they find it aesthetically pleasing” good enough, or do we have to go deeper?

            Of course you have to go deeper. Why do strawberries taste nice and poop bad? Is this an aesthetic judgment you came up with independently? Did you snatch it from the aether? Of course not, it is a product of evolutionary pressures (possibly combined with cultural pressures). Without understanding these distal forces you understand nothing about aesthetic judgment.

          • Ghatanathoah says:

            “I’m just doing it because I like it” is a perfect explanation. That is the main reason people do things. It only “kicks the can” if (like Hanson) you have an extremely confused model of human motivation that fails to distinguish “desires to do something” from “explanations why our desire to do something evolved or developed.”

            You and Hanson are conflating “there are evolutionary and developmental explanations for why people developed a desire to do things” and “people have secret, hidden motives for why they do the things they do.” This is fundamentally mistaken.

            Needing nutrition is a plausible explanation for why hunger evolved. But if a need for nutrition is the reason people eat, why is there obesity? It’s because hunger is the real reason people eat. You can’t talk someone out of overeating by pointing out that the food they are eating isn’t nutritious, because the true authentic reason they eat is that they are hungry. You instead have to point out that fulfilling their desire to eat interferes with other desires, like being healthy.

            The real reason Hanson gets criticism isn’t that he’s trying to overexplain something primal. It’s that his confusion leads him to say that signalling explanations are more “authentic” than non-signalling ones, when they are not. On the contrary, while signalling is a plausible explanation for why people might have evolved to engage in certain behaviors, it isn’t the real reason they behave that way. The “real” reason is the desire itself, not the explanation for why it developed.

            In the case of the OP, where we are discussing whether someone dyes their hair for attention, or because they like the color, saying “for attention” kicks the can even more than “I just like it” because then you need to explain why someone would like attention. At some point you have to stop passing the buck and admit that people just like something. And I don’t see why stopping the buck-passing at “I like it” is any more reasonable than stopping it at “I like attention.”

          • Anon. says:

            But if a need for nutrition is the reason people eat, why is there obesity?

            The answer to that, unsurprisingly, involves evolved distal causes.

          • johansenindustries says:

            @Ghatanathoah

            Ignoring the fact that ‘attention’ is obviously more clear than simply ‘it’. Perhaps you don’t think attention is an acceptable stopping point – in which case yes you ought to go deeper – but it seems intuitive (to me at least) that desiring attention is a more fundamental concept than desiring to have one’s hair dyed a weird colour.

            There is also the point that ‘because they like it’ is a worthless response to ‘why the sudden upswing in people dying their hair odd colours’ and that the response to ‘why does he dye his hair odd colours’ ought to be a useful clue to answer the earlier question.

            Of course it could be that you think that making yourself colourful is a fundamental human desire – sort of like a peacock (yes, the peacock has his colourful tail to attract mates, but we drink to avoid weakness so its close enough) – that has been earlier suppressed by the gray mafia. But I’m pretty sure that’s not your position.

        • Ozy Frantz says:

          I notice that my fellow funky-haired people are far more likely than average to have friends with funkily colored hair, that when I dye my hair I look forward to getting compliments on it, and that it is quite common to hear people say things like “oh, I dyed my hair before the party so I could show it off to everyone.”

          Signalling– which is essentially a word for indirect communication– is not bad or inauthentic, any more than direct communication is. And just like directly communicating a joke to someone is evidence you genuinely like the joke, indirectly communicating with your dyed hair is evidence you genuinely like dyed hair.

          • m.alex.matt says:

            Looking forward to getting compliments is also called looking for attention, Ozy 😉

            The problem with the people looking for attention in the way harland is complaining about isn’t that they are looking for attention, it is that they are looking for attention from people who are not necessarily interested in giving it, so they have to kind of be forced to give it. Dying your hair because it’s what your friends are doing and you would love to get some positive attention from them is one thing, dying your hair because no one else is doing it and you want to get people who otherwise wouldn’t pay you any attention at all can become very annoying very quickly.

          • Lambert says:

            If it annoy you, then you should ignore their hair colour.
            It’s quite simple, really.

          • johansenindustries says:

            @ Lambert

            Categorical imperitive. What happens if everybody ignores the annoying but ignorable things that they do to get attention?

      • harland0 says:

        So they dye their hair for signalling reasons, which means getting attention. Gotcha.

        • cuke says:

          Stiletto heels, monster trucks, Rolex watches, alligator skin boots, bright red sports cars, large gun collections, tattoos, dramatic eye makeup, the far end of every fashion trend ever. Big hair, “platinum blonds,” mohawks, lots of piercing. Big ass houses, gold toilets, expensive wine. Low-slung baggy pants, basketball shoes worn as fashion, low-rider cars, wild colored socks with plain dark suits, jeweled cufflinks. Luggage with brand logos all over them. Talking loud, telling outrageous stories, bragging, posturing, interrupting, talking more than everyone else in the meeting. Being an ass in public.

          How would we distinguish transgender people from the parade of signaling and attention-seeking human behavior across history?

          • Ozy Frantz says:

            Your list seems to me to be weighted towards things generally considered “tacky,” which seems silly to me. Sure, the far end of every fashion trend signals things, but so does the near end. Polo shirts and khakis signal things too. So do blue jeans and a T-shirt you got for free from a company you interviewed at.

          • cuke says:

            I didn’t mean at all to convey tacky. I was reaching for things that may convey something negative to out-group members but definitely aren’t considered negative to in-group members. I was trying to address the judgmental tone of the OP in seeming to suggest that transgender people using hair color is uniquely an over-the-top kind of attention-grabbing move particular to transgender people. So I was reaching for a variety of examples that are likewise considered over-the-top by outsiders in other times and places. I am aware that everyone is signaling stuff all the time.

            I don’t consider stilettos tacky. I had big hair and huge shoulder pads in the 80s and thought I was quite fashionable at the time. I grew up in a part of the country and in a time where/when low-rider cars were cool, as were luggage items with corporate logos plastered all over them. But absolutely, Brooks Brothers button downs are the same. I’m in my mid-50s and originally from Los Angeles, so my cultural references may be different from yours (and therefore you may read them as tacky).

        • Ozy Frantz says:

          In the same sense that you’re seeking attention by commenting on a blog instead of leaving all your comments in a text file like a digital Emily Dickinson, yes. I don’t get why so many people think engaging in indirect communication means you are evil.

          • johansenindustries says:

            But he’s looking for attention by writing comments from people who want to read a comment sections. You’re looking for attention from people on the street minding their own business who would be better off not having to dedicate processing power to ‘what’s that small turquoise thing just below eye-level’.

            (And, of course, he could just be posting so we have the benefits of his words. Not seeking compliments or anything else for his actions.)

          • rlms says:

            I certainly prefer having to dedicate processing power to coloured hair than low quality comments (no implication about original comment intended).

    • Iain says:

      Scott’s link gives a perfectly plausible justification for hair dye that has nothing to do with attention from others — specifically, that a dramatic change in appearance can help reduce dysphoria when looking in a mirror.

      If trans people really were needy for un-earned attention, I might instead recommend that they try trolling comment sections online. It appears to be a successful strategy.

    • Deiseach says:

      Ah, hair colouring is for the youngsters, leave them have fun with it. I sometimes wish I’d been brave enough to do crazy hair colours but I hadn’t the confidence when I was young and I’m much, much too old now (my one attempt was going mildly red – not candy apple red, but auburn – with a hair dye. Didn’t suit my colouring, so I had to wash it out). Could have gone blonde but I hate the idea of naturally darker-haired people going fake blonde (yeah, I know that contradicts what I’ve said about it’s only fun) so I never did or would. So I’m stuck with kinda mid-brown, dull hair (weirdly enough, I haven’t gone as grey as I’d expected by now – I have a small streak of grey hair in the front and the rest is mostly still dull mid-brown).

      • Winter Shaker says:

        You mean you haven’t tried going full goth with a plausibly-natural very-dark-brown or black dye? Don’t rule it out until you’ve given it a go 🙂

        • Toby Bartels says:

          I’m having trouble imagining Deiseach as a goth. Mind you, even learning that she had brown hair is a shock to my image of her, which is basically Molly Weasley. (Hopefully Deiseach sees this as positive.)

        • Deiseach says:

          When I was younger I would have loved to try dying my hair black, but again it probably would not suit my colouration, and the thing I was afraid of was doing it, looking in the mirror directly afterwards and going “Oh crap this was a mistake” and trying to wash it out and having to do that over a week or so, in the meanwhile looking awful.

          Plus it would have been too drastic a change – everyone who knew me would know this was not my natural hair colour and I couldn’t face the idea of dealing with comments on the change and why did I want to change it and did it look good or not and all the rest of it. I try to fade into the background as much as possible when out and about in the world outside the walls of my house, and Weird Hair Colour would be much too attention-attracting.

          Going darker brown than I already am is a bit “meh” because it’s not really enough of a change (why yes, I am contrary). I suppose what I’m looking for is one of those vivid hair dyes that I could apply, look in the mirror and say “Oh that’s great” or “No, that doesn’t work” and either way then immediately wash it out with no traces left behind so I could go out to work/other necessary public interactions as normal in the morning 🙂

          • Perhaps the solution is to go a little darker brown then, when everyone has gotten used to that as your normal color, a little darker still, and eventually, perhaps over a year or so, make it all the way to black or, if you prefer, very dark brown.

    • AnonYEmous says:

      why has no one mentioned that they do it because trans people are usually intimately linked with Social Justice / Progressive culture, and dyed hair is often seen in that culture

      “but why do those people do it” probably some combination of generalized transgressivism and a backlash against beauty norms, plus an emphasis on doing what makes you feel good, all these are probably factors

      • Nornagest says:

        I do think there’s something to the idea beyond demographics. Right off the top of my head I can think of two acquaintances who came out as trans and dyed their hair weird colors within a few months, which they had never previously done. That’s a big fraction of all the trans people I’ve known before and after their transition, so it’s a big signal. Could be a coincidence but I don’t think it is if there are articles about it.

        • Le Maistre Chat says:

          Note that unless it’s socially constructed, the fraction of the population with gender dysphoria should be stable at, what, 0.2%? That’s pi*Dunbar, so… how do you end up knowing 4 or more transes?

          • Nornagest says:

            A lot of stuff might be contributing to that, but the simplest and maybe the most important is that my friends aren’t a representative sample of the population. I expect pretty much everyone on this board knows a statistically bizarre number of people with rare traits.

            Of course, some parts of my life are more representative than others. I don’t think anyone I knew in high school is now trans, for example.

          • Ratte says:

            Trans folk tend to cluster – for a variety of reasons – similar to the way gay men and women do. I dated one and went from knowing having the one in my social circle to more than a dozen in under a year.

            This might just be my experience, but the ones that ‘pass’ and the ones that don’t self-segregate and don’t associate much with the other group.

      • Deiseach says:

        If you’re trans and contemplating changing your entire presentation, messing around with your hair colour is (a) a simple and easy first step that is not too drastic of changing your appearance, to dip your toe in the water (b) really extreme candyfloss colours and effects draw attention away from ‘hey you’re dressing like a girl/boy’ to ‘hey your hair is weird’ and so again eases the way into full transition from one appearance to another.

        • Matt M says:

          Interesting. I almost see it as the opposite. Dying your hair a weird color is a way of signaling “I am a weird person interested in weird things.” Couple dyed hair with a “gender-neutral appearance” and well, while you’re not literally wearing a sign saying “I AM TRANS” you’re basically signaling a high possibility that you might be – which might be enough to attract people who are into that sort of thing, and might serve as enough of a red flag to push away any hetero-cis people from trying to pick you up.

          I will personally admit that if I’m on dating sites and I see an androgynous looking person with dyed hair, I immediately move on. Sure it’s possible that they’re just a tomboyish female with a small chest who happens to really love the color purple, but it’s not really worth my time to find out…

    • Whatever Happened To Anonymous says:

      Do you think that’s uncharitable? Because I have a far more uncharitable take

      • JulieK says:

        Going from that page to the one on Müllerian mimicry, I was quite surprised to see the claim that although it was previously thought that the Viceroy butterfly is non-poisonous but mimics the protective coloration of the poisonous Monarch butterfly, research shows that the Viceroy is also poisonous!
        (Wiki calls this a “recent” discovery, but apparently it was known as early as 1991, though I only learned the old theory when I took AP Biology in 1993-4.)

    • Walter says:

      I think you are being a bit mean.

      Whatever your explanation for any particular style of dress (goth is a fine example) is will work for the dyed hair crowd.

    • Grek says:

      There’s also the less obvious answer that being willing to dye your hair in unusual colours and being willing to transition are both caused in part by shared factors such as ‘ability to determine own presentation’, ‘access to chemical treatments’, ‘willingness to risk potential ridicule’ and ‘self confidence’.

      If you filter out everyone who is actively prevented by others from changing their own appearance, everyone who cannot afford or who lives somewhere without reliable access to a pharmacy, who is particularly sensitive to public ridicule, who believes that altering their personal appearance using cosmetics or other outside influences is unethical, or who is unwilling to make any decisions which are not immediately reversible, you’re going to be left with a population that is both more likely to dye their own hair and more likely to come out as being transgender.

  27. Squirrel of Doom says:

    39% of Americans say that there is a greater-than-even-odds chance that global warming will cause the extinction of the human race.

    1. This is an absurd fear, even if you believe some quite extreme scientific predictions.

    2. If people actually believe this, even when they don’t respond to pollsters, isn’t this an enormous mental health crisis in our midst?

    • dansimonicouldbewrong says:

      It’d be interesting to see similar polls about other apocalyptic scenarios–nuclear war, alien invasion, asteroid strikes, and so on (including, dare I say it, super-intelligent AI). My guess is that they’d all get a similar response, for a similar reason: people hear people talk about them as if they’re real and serious, and therefore assume that they’re real and serious, but also figure that (a) whatever happens will happen after they die and (b) there’s nothing they can do about it anyway. So in the end they don’t worry about it too much.

      • soreff says:

        @dansimonicouldbewrong
        Hmm – actually, I do expect somewhat-more-intelligent-than-human AI
        to have more than 50/50 odds of causing human extinction but

        (a) whatever happens will happen after they die and

        Yup (I’m 59, and I doubt that it will happen in the next 20 years)

        (b) there’s nothing they can do about it anyway.

        Pretty much true – I have little enough control over software projects
        that I directly work on…

    • Watchman says:

      Did the question have a timescale? Otherwise they’re pretty well certain to be right if we hang around long enough and forget to leave the planet.

    • actinide meta says:

      39% sounds about right for the intersection of (a) people who listen to the question only carefully enough to determine whether it is more similar to “Are you a Republican?” or “Are you a Democrat?” and (b) people who are Democrats.

      I wouldn’t worry about it, unless you still think normal people have coherent political philosophies or facial beliefs. 🙂

    • ildánach says:

      It isn’t actually completely implausible. As a small example, some modern research thinks that the runaway greenhouse effect responsible for the climate of Venus could happen on Earth more easily than we believed in the 90s.

      Needless to say, the climate is incredibly complicated, and anthropogenic effects involve many potential positive feedback loops that have never been activated in nature. If something like the Clathrate gun hypothesis (or a similar forcing element) is true, the extinction of humanity would be reasonably likely.

      From the Fermi paradox, we can extrapolate that there is a good probability that life-sustaining environments are very fragile, thus it is best to err on the side of caution.

      • rdplatypus says:

        I’d argue the Fermi paradox tells us there’s likely to be some kind of low-probability hurdle, but not that the hurdle is “life-sustaining environments are fragile.”

        In fact, the lack of even simple life elsewhere in our solar system where we think Earth extremophiles could in fact survive if not thrive (Martian subsoil, crater shadows, Saturn / Jovian moons, etc.) is circumstantial evidence toward the hurdle being in our past and not our future. If we find remains of skyscrapers under the clouds of Venus, the environmental fragility hypothesis would get a big boost.

      • 4bpp says:

        What sort of feedback loops are we talking about here? As far as I know, the current scientific consensus is that we are very far below historical peak temperatures, so unless the clathrate deposits or whatever other mechanism you posit have assumed a completely new form in the past 50 million years or so, it seems like we still have a safety margin in excess of 7K before any feedback loop that is intrinsic to the rest of the ecosystem kicks in.

  28. dansimonicouldbewrong says:

    “Software engineering” is, in my opinion, one of the most absurd fields in all of research (and that’s saying a *lot*). Based on its name, one would expect it to focus on questions like, “how can we build reliable, high-performance software from unreliable components with poor performance?”, or, “what general tools and methods can we come up with to make software run faster, more reliably and more cheaply?” This is what normal engineering disciplines do, after all. But imagine if a field like, say, aeronautical engineering were devoted primarily to (a) evaluating various ways to organize airplane factories, and (b) eliminating all flaws in individual airplane parts, on the assumption that a single tiny flaw in any single airplane part could easily be enough to bring a plane down. As far as I can tell, that’s basically what software engineering is about.

    • skef says:

      Based on its name, one would expect it to focus on questions like, “how can we build reliable, high-performance software from unreliable components with poor performance?”, or, “what general tools and methods can we come up with to make software run faster, more reliably and more cheaply?”

      It sounds to me like you have a very academic view of “engineering” in mind. The question most mechanical engineers focus on is “How do I build this, given these restrictions of package, cost, etc.?” And as part of that, “What are appropriate tolerances to put on each of these dimensions?” Tackling meta-level questions is relatively rare.

      • dansimonicouldbewrong says:

        I actually think we’re on the same page here. Software engineering research of the type that Scott linked to doesn’t actually help software developers do their job, which, as you say, is to build fast, reliable, usable software as quickly and cheaply as possible. Rather, it attempts to answer meta-questions like, “what ways of organizing software development projects tend to result in better productivity and code quality?” Tools that actually help developers write code are ubiquitous–but they don’t come from software engineering researchers, who are more interested in higher-level goals.

        • skef says:

          Well, sure, but there are probably similar studies taken up in University mechanical engineering departments, with the idea that the research will eventually wind up being integrated into classes to have some effect on practice.

          • dansimonicouldbewrong says:

            The field of industrial engineering is devoted to studying and improving the efficiency of industrial processes, but I don’t think chemical or electrical engineering is dominated by “industrial engineering as applied to our discipline” the way software engineering is.

            That’s a shame, because there’s a constant and substantial flow of new software engineering paradigms, practices and principles–almost all generated by ordinary (or particularly talented/creative) software developers, not by “software engineering” specialists–and absolutely no serious study of their effectiveness, let alone integration of them into sets of engineering principles and practices that, if learned and followed, could result in optimal code quality and developer productivity.

            Again, this is how other engineering disciplines work, and if you don’t believe that software development should work that way, then the least you could do is not call your field “software engineering”.

          • The Nybbler says:

            Again, this is how other engineering disciplines work, and if you don’t believe that software development should work that way, then the least you could do is not call your field “software engineering”.

            The train drivers and boiler plant operators are still irritated at having their term taken, and the siege-engine makers were probably irritated before that.

            As far as I can tell, academic software engineering (CMUs Software Engineering Institute being the type specimen) is mostly about producing paperwork, and processes by which “software engineers” can produce paperwork (instead of code). The value of this paperwork is not anything intrinsic to it; it’s just that other engineering disciplines are buried in paperwork, so software should be too.

          • CatCube says:

            …it’s just that other engineering disciplines are buried in paperwork…

            You’re going to have to detail this “paperwork” you think is excessive. I’m hard pressed to think of paperwork I need to do that doesn’t need to be there. Unless you’re going to go for “you don’t really need to produce or keep the calculations you use or document why you made certain design decisions.” Because as somebody who spends most of his time either inspecting or designing new features for existing facilities, I can assure you the designers should absolutely need to do that.

          • The Nybbler says:

            You’ve got your System/Segment Design Document (SSDD), your Software Requirement Specifications (SRS), your Interface Requirement Specifications (IRS), your Software Product Specification (SPS), your Software Design Documents (SDD), your Interface Design Documents (IDD), and of course you have a Requirements Traceability Matrix (RTM) to tie it all together. Once you’ve written all of those and gotten them all reviewed and made sure everything’s traced all the way from the SSDD to the SDD and back up, then maybe you can consider writing some code.

            Of course the famous line in aerospace, as opposed to software: “The airplane is ready to fly when the weight of the paperwork equals the weight of the aircraft.”

          • gbdub says:

            If you’re 20 and don’t think the amount of paperwork involved in engineering is excessive and a waste of time, you aren’t an engineer.

            If you’re 30 and haven’t yet realized that a substantial portion of that paperwork is actually necessary, or at least useful enough to make ignoring it dangerous, then you’re a fool.

        • quaelegit says:

          > Software engineering research of the type that Scott linked to doesn’t actually help software developers… build fast, reliable, usable software as quickly and cheaply as possible.

          > Rather, it attempts to answer meta-questions like, “what ways of organizing software development projects tend to result in better productivity and code quality?”

          Doesn’t “better productivity and code quality” lead to fast, reliable, and usable software (or at least help with the production of it)? I’m not seeing the objection here.

          • dansimonicouldbewrong says:

            Well, you could ask, “what color and design of window treatments leads to better productivity and code quality?”, but I don’t think you’d come up with very useful answers. Similarly with questions that focus on high-level organizational approaches, and neglect the concrete problems that developers face, and the tools and methods that software developers actually use (or could use) to solve them.

    • Bugmaster says:

      I believe this koan from The Codeless Code applies here.

      • dansimonicouldbewrong says:

        Thanks for the pointer. Basically, the parable is meant to teach the lesson that software development isn’t an engineering discipline, and never can be one. I profoundly disagree, but even if I agreed, I’d consider the supposed field of “software engineering” to be absurd–just for different reasons.

        • Bugmaster says:

          I guess it depends on what you mean by “engineering”. If you mean, “a discipline whose goal it is to build physical constructs”, then of course software engineering isn’t it. If you mean, “a discipline whose goal it is to build constructs under a set of mostly fixed constraints”, then arguable software engineering isn’t that, either. However, if you broaden the definition to something like, “applying theoretical knowledge in practical ways in order to create useful constructs”, then software engineering does fit.

          The problem with software, though, is that you’re building things out of ideas, not atoms. The short design cycles, changing technologies, fluid requirements, etc. are not random acts of misfortune — instead, they are features of the problem space, just as metal fatigue is a feature of the problem space in mechanical engineering. You might argue that software engineering isn’t as adept at solving the problems of its field as mechanical or electrical engineering, and I’d agree (it’s a young discipline, after all); but that’s not the same thing as saying that it solves zero problems in total.

          • dansimonicouldbewrong says:

            I would agree with your last, broadest definition, and my quarrel isn’t with software developers in general, some of whom might informally refer to themselves as “software engineers”, but rather with researchers who specifically define themselves as studying the field of “software engineering”. It’s not that they haven’t solved as many problems as I’d like yet–I’d be satisfied if they were merely asking roughly the right set of questions, such as “what software-building tools, methods and practices can developers use to make their software more reliable, performant and cost-effective”?

            Software developers already attempt to address this very question all the time, in ad hoc ways, producing innumerable such tools, methods and practices. One would expect software engineering experts to (1) begin by developing quantitative benchmarks by which to measure the effectiveness of these ad hoc resources; (2) proceed to identify common elements among the measurably most effective; and (3) finally, extract empirically validated guidance to developers regarding their optimal use. Instead, they’re mired in the kind of software-development-as-industrial-process thinking exemplified by the research Scott linked to.

          • The Nybbler says:

            One would expect software engineering experts to (1) begin by developing quantitative benchmarks by which to measure the effectiveness of these ad hoc resources; (2) proceed to identify common elements among the measurably most effective; and (3) finally, extract empirically validated guidance to developers regarding their optimal use.

            They did. It was called ISO-9001, and it was a disaster. If the idea of software engineering is to reduce programming to a discipline characterized by mechanical application of best practices to the problem space, it seems unlikely to work.

          • gbdub says:

            Part of the problem is, I think, the insanely broad design space that “software engineering” covers. You wouldn’t apply the same standards or processes to building a suspension bridge that you do to designing a paper towel dispenser, and yet you’ve got people trying to apply principles from Google style app development to creating ICBM flight software (or vice versa).

            Of course worst of all is when you try to apply manufacturing principles from Toyota to code development as if developing software was an assembly line where “Kanban” can mean a damn thing, but that doesn’t stop companies that build physical things that also need in-house software from trying.

            But none of this means the idea of trying to develop efficient documentation and processes for software development is a lost cause, just that it’s highly unlikely to be “one size fits all” or look much like any other sort of industrial process except in the very broadest terms.

          • Bugmaster says:

            @dansimonicouldbewrong:
            I’m not sure who qualifies as “researchers who specifically define themselves as studying the field of “software engineering””, but at least some people have managed to come up with some very real advances in the field.

            Whole programming languages (such as, recently, Rust and Go) have been invented to eliminate most common programming mistakes. Design Patterns have been codified in an attempt to classify most common solutions to software engineering problems. IDEs and compilers now include lint-checking; some IDEs can perform deep analysis of your code, highlighting potential errors in real time. Common security errors are researched, codified, and taught in software engineering classes. All of these efforts (and many others, of course), definitely provide some answers to your question: “what software-building tools, methods and practices can developers use to make their software more reliable, performant and cost-effective ?”.

            The problem with your traditional approach, i.e. “begin by developing quantitative benchmarks by which to measure the effectiveness of these ad hoc resources” is twofold. First of all, software engineering is developing very quickly, in response to rapidly changing requirements. A steel bridge is pretty much the same today as it was 50 years ago; 50 years is an eternity in the software world. Benchmarks don’t live very long.

            Secondly, there’s no such thing as “software”, just as there’s no such thing as “engineering”. A chemical engineer deals with vastly different sets of tools and constraints than an architect; similarly, a firmware engineer deals with a vastly different domain than, say, a data scientist. There are few “common elements” among their fields.

            You deride the “software-development-as-industrial-process” approach, but again, I think you may be missing some details. On the one hand, I do agree that prescriptions like “you should use message-passing instead of shared memory access” and “you should use an Agile approach” are qualitatively different. However, as I said before, software is built out of ideas, not steel girders. This means that the process of communicating and managing ideas is a key building block of software, and improving it will make software stronger.

            True, not every newfangled development methodology, programming language, design pattern, etc., is useful (in fact, most probably aren’t); but then, I’m not claiming that software engineering is somehow perfect. No discipline is.

          • Douglas Knight says:

            Academics building such tools don’t call themselves “software engineering researchers” but “programing language researchers.” But weighing the upsides against the downsides is usually an empirical question that feels more like engineering. If you make a new lint, you can measure the bugs caught and the false positives, although that’s going to vary a lot from person to person, and coding styles. Rust makes guarantees with obvious value, but is it usable? Some people reject it because of the string library, which is good to know, but doesn’t tell you whether to copy its pointer semantics in your next language. Usually people just say “It works for me.” I’m not sure what Dan thinks about those people, but I think it’s a useful thing to say and I don’t blame people for making a positive contribution and stopping there. But there are other people who, as a result of academic incentives, put in a lot of cargo cult effort to make a rigorous measurement of something. I’m pretty sure those are the people he is condemning.

      • poignardazur says:

        Thanks for the link! I’ve just started reading and I’m loving this blog.

    • Zakharov says:

      I studied software engineering, and as far as I could tell the research academics in the department were much more interested in things like making new compiliers than researching how software engineers build software. The latter is a small subfield of software engineering research.

    • johansenindustries says:

      Isn’t your aeronautical engineering example quite similar to chemical engineering? (e.g. https://www.undergraduate.study.cam.ac.uk/courses/chemical-engineering)*

      The main similarity of Software and Chemicals that aeronautics don’t show is the former are made out of the former, whereas planes aren’t made of planes.

      * I’m not a chemical engineer. Never been a chemical engineer. Know no chemical engineers etc.

  29. shenanigans24 says:

    Christians are boycotting British food producer Greggs for making a nativity scene with a sausage roll as Jesus.

    The tweets in that article sound like “Christians” are pointing out hypocrisy, not upset about the display. The U.K. actually has hate speech laws and publicly parodying Mohammad carries potential legal consequences as well as public outcry. I think they’re simply pointing out that the UK isn’t against “hate speech” they’re against speech that criticizes some views but okay with other views. So they have wrong opinion laws not hate speech laws in practice.

    • Watchman says:

      I have never seen a case of parodying Mohammad treated as a hate crime, although it has been stopped for public safety reasons (our Muslim community has a nice hard core of idiots). If you want you stores burned by intolerant biggots, you’re free to make a model of Mohammad out of a pork product. The UK’s problem is that much of our political establishment would try and ‘understand’ why this was offensive (it’s not difficult in thiscase) rather than try and eradicate the violent response to being offended.

      And the complainers were saying this was offensive and that Greggs should be boycotted. For reference, getting Christian campaigners in the UK to make this sort of fuss will produce very few boycotters (we’re mostly sort-of-Anglican or mild Catholic, or just not religious (which does not preclude turning up to church to get married or for the services with the really good songs), and mildly unhealthy food is more of a religion amongst many anyway) and a lot of publicity. After all, how many non-UK commentators on here had heard of Greggs before? How many might try them (they’re not bad for the price band) if they encountered one now?

  30. TentativeQuestioning says:

    And here I am on my Leesa mattress. I sure do feel like a sucker now.

    • Protagoras says:

      Hmmm. Consumer Reports likes Casper a lot better than Leesa, though Leesa’s scores aren’t terrible. I am under the impression that Consumer Reports is regarded as relatively trustworthy (if I weren’t under that impression I wouldn’t subscribe to them), but it’s entirely possible I just haven’t heard about their scandals the way I have about some of the online reviewing sites. I guess I’m posting this in case somebody does have a link to why Consumer Reports can’t be trusted either.

      • randallsquared says:

        I dunno about “can’t be trusted”, but it doesn’t make me trust them *more* when I look for electric cooktop reviews, and find that they’ve reviewed 27 (of far more easily available; Lowes.com alone has more than 100) and haven’t found one they like less than 78/100.

  31. shenanigans24 says:

    This is the world that all of you “free speech only constrains the government” and “it’s just people think you’re an asshole and are showing you the door” people have built for us.

    In what practical way could this not be the case though? A law that says you can’t discriminate on opinions or actions privately? That would be far more oppressive since it’s a limitation on people’s behavior and speech instead of just limitation on what laws can be passed.

    • pipsterate says:

      I think the idea is to create different cultural standards, which would hopefully make it a worse PR move for a company to fire a controversial employee than it would be to keep them around. A law wouldn’t necessarily be required to change the situation, because the law didn’t create this in the first place, culture did.

      I’m not optimistic about that happening anytime soon though.

      • Ketil says:

        Well, you could have laws protecting people from being fired without reasonable cause – and perhaps explicitly defining voicing of political opinions not to count as such. That’s how it works in this corner of Europe at least.

        • soreff says:

          Agreed. It is important to remember that management is a small
          fraction of the population. Giving them a completely free hand to
          arbitrarily fire people reduces the average person’s liberty.

          • Mary says:

            You’re not giving them a completely free hand. That’s what they have naturally. You’re forcibly preventing it by having all their decisions subject to second-guessing by an unaccountable bureaucrat with zero skin in the game. That’s the REAL way to reduce liberty.

          • soreff says:

            @Mary
            As things stand now (in the U.S.) the situation is that management
            is unaccountable to the people whose lives they twist.
            As I said before, their power to arbitrarily fire people
            reduces the average person’s liberty.

            I’d far prefer to have it as in Ketil’s corner of Europe, with
            laws preventing someone from being fired without reasonable cause.

          • Petja Ylitalo says:

            The problem with demanding a reasonable cause is that it is really hard to know for an outsider whether the accusation is true (it can even be hard for management to know how good a worker is, if they´re not working alongside him most of the time).
            So how this “cause needed to fire” actually works (atleast what i´ve seen) is that as long as the company in total makes money workers can be very lazy, working with 30..50% capacity of a worker trying his best, without getting in trouble (though often there are social incentives to not do this).
            When the company loses money it is easier, since you can just scale down your production and use that as a reason to fire those you want to get rid of.

            The downsides from this are: when buying things you lose money because everything costs more because the same work costs more for the company (which means that since competitors have the same problem they can raise prices to compensate)
            Getting hired gets much harder, since hiring a bad worker that is hard to get rid of is really expensive, so companies are very careful about hiring new people (and often prefer to pay overtime to old ones instead if they know they are trustworthy)

      • Matt M says:

        I think the idea is to create different cultural standards

        This.

        I’m confused as to why “free speech” is a good concept for the government, but supposedly a meh concept for everyone else.

        If we hold free speech up as an ideal, we should hold it up as an ideal for everyone, whether it’s “legally required” of them or not.

        • Bugmaster says:

          How exactly do you propose to “create different cultural standards” ? I don’t think this is even possible to accomplish — at least, not deliberately.

          • Le Maistre Chat says:

            Convert people to a cult with different standards. That’s worked a few times.

          • Lambert says:

            If it’s impossible to create different cultural standards, where did all the ‘whites only’ water fountains and expectations that women don’t belong in the workforce go? Was that an accident?

          • Bugmaster says:

            @Lambert:
            Yes, it was in fact an accident of sorts. More specifically, racism as a cultural standard is probably a byproduct of human evolution, although it is still quite poorly understood. The laws against sharing water fountains are not cultural standards, but merely their symptoms. Same thing goes for the expectations that women don’t belong in the workforce.

          • Aapje says:

            @Bugmaster

            I see prejudice as a very useful adaptation. If you encounter a snake, you want to default to the assumption that it is dangerous, even if the chance is relatively small.

            However, I think that we were historically used to a low information and high-risk environment, where people had to extrapolate wildly and couldn’t afford to take too many chances. So we are probably miscalibrated a bit for modern society.

        • shenanigans24 says:

          Because free speech the ideal derives from being able to own yourself and your actions if they don’t harm others. Words don’t harm other people so it is not justified to take physical violence to stop them. Laws are the threat of physical violence. There are however laws to protect people from violence for any reason including speech. That is in keeping with a cultural concept of free speech, not a first amendment concept so both exist.

          Choosing not to associate with people is also under the umbrella of free actions that are not harming people, which is also the principle free speech derives from.

          Any attempt to create this acceptance of free speech through forced coercion over voluntary coercion would violate the principle free speech is derived from.

          • Matt M says:

            Who said anything about coercion?

            My point is that when someone is fired for saying something unpleasant, there are two reactions one can have.

            1. “That was a shitty thing to do, I denounce this practice.”
            2. “They have a legal right to fire that person because they aren’t the government therefore ‘free speech’ doesn’t apply here!”

            Most people seem to act like you have to pick one or the other. My point is no, BOTH are 100% true. And strongly emphasizing 2 while failing to address 1 ends up building a culture that does not particularly value free speech.

            And we all know what resides downstream of culture…

          • Davide S. says:

            @Matt M
            Does it matter if the free speech of the employee harms the business?
            I’m suprised no one commented on the fact this was a *marketing* job.

            I see no obvious contradiction in believing that employers are justified in firing people for saying some things, but the government shouldn’t punish them. (unless the government is their employee of course).

            @shenanigans24

            I am generally pro-speech, but I DO see a contradiction in simultaneously claiming that
            1. Speech is really important.
            2. Speech can’t harm people.

            There is a quote attributed to Stalin that I feel is relevant here
            “Ideas are more powerful than guns. We would not let our enemies have guns, why should we let them have ideas?”

            So free speech is important because speech is powerful – and it’s hardly absurd to try to restrict powerful things in some ways.

            I think it’s more reasonable to admit that yes, speech can harm people, and does all the time, even if indirectly, but there might be reasons to be especially tolerant of this harm.

            (but not 100% tolerant, as that would lead to libel & defamation being acceptable)

          • Aapje says:

            @Davide S

            I don’t see how it is relevant that she had a marketing job. It may have been relevant if she was a spokesperson or had an otherwise consumer-facing job. However, most ‘scandals’ have a very short half-life, so even then an employer can often just reassign the person to a different job for some time or give them an administrative leave of absence, until the scandal blows over.

            Also, I think that shenanigans24 used a fairly restrictive definition for “harm.”

          • Matt M says:

            Does it matter if the free speech of the employee harms the business?

            Look, realistically, there is no objective standard here – these issues would be decided subjectively on a case-by-case basis.

            I admit that if someone says “I think men and women have different preferences” my reaction is “Even if you disagree, this is a legitimate opinion and firing someone for expressing this is morally reprehensible.”

            But if someone says “I think racial/ethnic/religious group X should be systematically exterminated and by the way please try the latest flavor of Pepsi because I hold an important position there!” it strikes me as entirely reasonable for Pepsi to get rid of that person, and I wouldn’t be upset if they did.

            I admit that this issue gets complicated when you have a non-trivial portion of society who thinks that there is no significant difference between “Men and women have different preferences” and “Exterminate the inferior race”

          • Matt M says:

            I don’t see how it is relevant that she had a marketing job.

            This comes up a lot actually. Justine Sacco (the woman who made a so-called AIDS joke in South Africa) was in PR.

            The logic isn’t “these people represent the company”, but rather that people in marketing or PR should be expected to know better about these sorts of things. Often you see their firings justified as “Well this just proves they are really bad at their job!” which is somewhat plausible, given that their job is to analyze and predict how customers will react to various images presented in media.

          • soreff says:

            Choosing not to associate with people is also under the umbrella of free actions that are not harming people,

            Firing someone harms them.
            That is obvious common sense.
            I’m calling bullshit on this.

          • Firing someone harms them.
            That is obvious common sense.

            Do you distinguish between harming someone and failing to help someone?

            Suppose you gave me $10 for XMas last year. This year you don’t. Have you harmed me?

            You hired me to mow your lawn last week. This week you don’t. Have you harmed me?

            There are two different concepts going under the “harm” label. One is “due to this decision, I am worse off than if you hadn’t made it.” By that the answer to both of my examples is “yes.”

            The other is (roughly) “made me worse off than I would have been if I didn’t interact with you.” By that the answer to both is “no.”

            Almost certainly, there is someone in the third world whose life you could have saved with sufficient effort and money–and didn’t. Are you a murderer?

          • John Schilling says:

            Firing someone harms them.
            That is obvious common sense.

            Divorcing someone, other than by mutual consent, also harms them. This is obvious common sense, and for the most part it’s the same common sense that says firing someone harms them. And marriage, unlike employment, comes with an explicit lifetime commitment.

            So what’s your position on unilateral no-fault divorce?

          • Harry Maurice Johnston says:

            Isn’t that why alimony exists?

          • soreff says:

            @DavidFriedman

            Look, at every company that I’ve worked for
            the company itself refers to termination
            as a threat or punishment in its code of conduct.

            Suppose you gave me $10 for XMas last year. This year you don’t. Have you harmed me?

            This is literally orders of magnitude away from a manager
            threatening a worker’s livelihood. Treating the two as
            analogous is ludicrous.

            What prompted this whole discussion was a firing,
            not the presence or absence of a $10 gift between
            individual people of perhaps roughly equal power.

          • soreff says:

            @John Schilling

            You have an interesting point. When my first wife divorced me,
            yes, she harmed me but
            a) She did have to file papers in a family court –
            Unlike a manager, she could not act completely arbitrarily,
            the judge acted as a check on anything unconscionable
            b) There wasn’t, and often wouldn’t be, a huge disparity
            in power between the spouses, while even a small
            company typically has financial resources an order of
            magnitude beyond what a typical employee has

          • I wrote:

            Suppose you gave me $10 for XMas last year. This year you don’t. Have you harmed me?

            Soreff responded:

            This is literally orders of magnitude away from a manager threatening a worker’s livelihood.

            That is relevant to the question of how much one is harmed. It isn’t relevant to the question of what counts as harm and what doesn’t.

            If the woman I am in love with refuses to marry me, the size of the effect on me is large. That doesn’t tell me whether I can legitimately claim that she harmed me by doing so.

          • Brad says:

            I guess it is mostly semantic, but I’d say it’s harm, she inflicted it, but she was entitled — legally and morally — to inflict it.

          • soreff says:

            @DavidFriedman

            It isn’t relevant to the question of what counts as harm and what doesn’t.

            Firing someone is harming them.
            Ask anyone outside of a libertarian bubble.

            As I said previously
            The companies themselves describe termination
            as a punishment that they use.

            Use common sense.

            Even if you think that completely arbitrary firings ought to be
            legal, for you to deny that they harm the person fired is
            asinine.

          • skef says:

            Do you distinguish between harming someone and failing to help someone?

            Suppose you gave me $10 for XMas last year. This year you don’t. Have you harmed me?

            You hired me to mow your lawn last week. This week you don’t. Have you harmed me?

            There are two different concepts going under the “harm” label. One is “due to this decision, I am worse off than if you hadn’t made it.” By that the answer to both of my examples is “yes.”

            The other is (roughly) “made me worse off than I would have been if I didn’t interact with you.” By that the answer to both is “no.”

            However, firing someone is plausibly harming them by either of these standards, if by never having “interacted” with the company that fires you, you were to obtain a job at a company that wouldn’t fire you under the same circumstances.

          • Jiro says:

            It requires unique activities and expenditures (on both sides) to start hiring someone which are distinct from the ones in keeping someone continuously hired. It also requires unique activities and expenditures to fire someone.

            If something requires unique activities and expenditures in order to start, and more of them in order to stop, most people consider stopping it to be an action, even though stopping things that don’t have such expenditures is inaction.

            If you keep saying “normal people think this way; I can’t make any sense of that” you may be advised to try modelling normal people better.

          • John Schilling says:

            When my first wife divorced me, yes, she harmed me but
            a) She did have to file papers in a family court –
            Unlike a manager, she could not act completely arbitrarily,
            the judge acted as a check on anything unconscionable
            b) There wasn’t, and often wouldn’t be, a huge disparity
            in power between the spouses, while [corporations got $$$]

            On point b), the power disparity should only matter if there’s going to be a fight. If divorce is a matter of “I divorce thee x3”, if termination of employment is “I quit!” or “You’re fired!”, then it doesn’t much matter who has the big legal staff.

            If instead you insist that nobody be fired except for due cause as demonstrated via due process of law or the equivalent, then you put the lone employee up against Corporate Legal, and when they lose it isn’t just “you’re fired!” but “here’s the formal legal ruling that you are a Bad Employee, on your permanent record”.

            As for point a), with no-fault divorce the paperwork and family court appearances are going to be mostly pro forma. I agree that having a judge sign off on even pro forma paperwork can be a useful safeguard (but see above), if not one that’s going to make a real difference very often. So maybe we can talk about a similar requirement for employment contracts.

            Are you OK with having to go before a judge to get their OK before you walk out on your job the way your first wife walked out on you? Should the judge be able say, “waiting until a week before the project deadline and demanding a $10k bonus or you’d walk is unconscionable, so now you have to pay your employer’s penalty fee for their late delivery”?

          • However, firing someone is plausibly harming them by either of these standards, if by never having “interacted” with the company that fires you, you were to obtain a job at a company that wouldn’t fire you under the same circumstances.

            Hiring plus firing you is harming by that standard. But doing that requires your consent to the initial hire. If the consent was obtained fraudulently, by claiming that they never fired except for causes x, y, or z and they then fired you for a different cause, you have a legitimate complaint. If not, it’s a harm you consented to when you accepted the job.

          • skef says:

            a harm you consented to

            Well, now we’re getting rather abstract.

            New example: I quit on a Tuesday with no notice. Have I harmed my employer?

            I think many people in the U.S. would say “yes”, and that I should give at least two weeks notice. But doing that isn’t (generally) a requirement, it’s a social convention. If it isn’t in the hiring agreement, they presumably “consented to” it when they hired me.

            So it seems that either can’t be a harm, or such social conventions can be relevant to what constitutes a harm. If the latter is the case, it’s hard to see what would screen off firing someone for political views as not being a harm, whether or not you “consented to” it.

          • or such social conventions can be relevant to what constitutes a harm.

            As I said earlier, consent fraudulently obtained isn’t really consent. What is or isn’t fraud depends in part on the understanding of the parties to what they are agreeing to, which depends in part on social conventions. Language itself, after all, is a social convention.

          • soreff says:

            @John Schilling

            I agree that having a judge sign off on even pro forma paperwork can be a useful safeguard (but see above), if not one that’s going to make a real difference very often.

            Thank you.

            Actually, even in an amicable divorce, there was some degree of
            negotiation on terms, so a large power imbalance would have
            mattered.

            If instead you insist that nobody be fired except for due cause as demonstrated via due process of law or the equivalent, then you put the lone employee up against Corporate Legal, and when they lose it isn’t just “you’re fired!” but “here’s the formal legal ruling that you are a Bad Employee, on your permanent record”.

            But the point of limiting firings to due cause is to prevent the
            unjustified firings from happening. The unjustified firings would
            turn into either cases where the management has to grit their
            teeth and let the employee stay, or have a formal legal ruling where
            management is convicted of an unjustified firing.

            Are you OK with having to go before a judge to get their OK before you walk out on your job the way your first wife walked out on you?

            Yes, but only if the circumstances allowed for firings were very
            tightly restricted. If workers’ protections were approximately
            equivalent to the protections that tenured faculty used to have,
            then I’d be ok with needing a judge’s ok to leave a job.

          • Brad says:

            @skef

            I think many people in the U.S. would say “yes”, and that I should give at least two weeks notice. But doing that isn’t (generally) a requirement, it’s a social convention. If it isn’t in the hiring agreement, they presumably “consented to” it when they hired me.

            So it seems that either can’t be a harm, or such social conventions can be relevant to what constitutes a harm. If the latter is the case, it’s hard to see what would screen off firing someone for political views as not being a harm, whether or not you “consented to” it.

            The answer is right there. The prevailing social convention / norm / whatever-you-want-to-call-it in the US allows a company to fire an employee for having gone to a neo-nazi rally or for there being a viral picture going around showing the employee giving the President the middle finger.

            If you think that’s a bad thing, that’s fine, there are prevailing social conventions I don’t like either. But if you hope to convince people then you likely need to make some sort of argument. Not just write indignantly about how could people possibly be okay with this. That sort of thing could work if you had enough people with enough status doing it, but that just moves the problem back one step. You need to convince them.

          • soreff says:

            @Brad

            The prevailing social convention / norm / whatever-you-want-to-call-it in the US allows a company to fire an employee for having gone to a neo-nazi rally or for there being a viral picture going around showing the employee giving the President the middle finger.

            The currently prevailing laws in the US do allow a company
            to fire an employee under those circumstances.

            Even if general social norms considered this to be an ethical
            action by the company, and I’m not convinced that they do,
            that is a different question from whether John Q. Public in
            the US would consider firing an employee to harm the fired
            employee.

            Regardless of whether the firing was justified, I’d be
            astonished to hear that the general public did not view
            the firing as harming the employee.

          • Conrad Honcho says:

            Late to the party here, but I think it’s also worth noting that she worked for a government contractor. So, she was giving the bird to the head of her company’s biggest (only?) customer.

            I’m very much in favor of not firing people for their opinions and expression thereof, but I’m not an absolutist about it. This seems like a gray area for me. If you work for a PR firm and Coca Cola is your biggest customer, and when you’re off the clock you tell the president of Coca Cola how much you think Coca Cola sucks, I can’t be that surprised when you get fired.

        • Brad says:

          I’m confused as to why “free speech” is a good concept for the government, but supposedly a meh concept for everyone else.

          It’s a category error to equate the government punishing someone for something he said and a private entity reacting to what someone said. Your “confusion” is akin to asking a libertarian how he could be opposed to taxes when he is perfectly fine with people going into a store and buying things.

          This idea that somehow people should be able to say anything they want to anyone else in any circumstance and no one else should be allowed to treat them any differently because of it is frankly bizarre. Contrary to all the claims here to the contrary, it describes exactly zero periods in zero societies in all of human history. That is to say — it isn’t any kind of existing norm that is being defended from barbarians but rather an extraordinarily radical new norm that advocates aren’t even offering any kind of justification for but instead merely insinuating that it is obviously a good idea and disingenuously pretending was the status quo in some vague but recent golden age.

    • birdboy2000 says:

      There is a categorical difference between the “freedoms” of the bourgeoisie and the freedoms of workers, restricting the former is necessary to ensure the latter can exercise any meaningful freedom at all. If the law says “you can’t fire me for X” (which it already does in some circumstances, just not political speech) most people will experience more freedom. And if said laws are imperfectly enforced – why, that’s just a further argument against at-will employment being compatible with genuine democracy.

      • soreff says:

        Well said!

      • Zakharov says:

        Bear in mind that the freedoms of the bourgeoisie don’t always conflict with the freedoms of the workers. A law stating that only the hereditary aristocracy may hire workers restricts the freedom of both. Making it easier to fire people reduces the risk of hiring people, which increases the bargaining power of labor. Laws ought to be evaluated on specifics, not broad principles.

        • Harry Maurice Johnston says:

          Making it easier to fire people reduces the risk of hiring people, which increases the bargaining power of labor.

          An interesting theory, but I don’t think it works that way in practice.

          • Viliam says:

            Making it easier to fire people reduces the risk of hiring people (if you change your mind, you can still fire them later).

            The impact on bargaining… I can imagine forces working in both directions, so the question is which one is empirically stronger.

            Making it easier to fire people decreases their bargaining power. (“My way or the highway.”) However, if hiring more people leads to fewer people being unemployed, that in turn increases the bargaining power. However, if making it easier to fire people increases the amount of work you can squeeze from them (“if you are unwilling to do unpaid overtime, you are free to go… no, I won’t put ‘they refused to do unpaid overtime’ on paper, I will just say ‘fired without any reason whatsoever’, because that is perfectly legal”), then maybe the employment will not increase. Etc.

  32. Philosophisticat says:

    most people are really averse to genetic testing and embryo selection, even to the point where they will (in real life, not hypothetical questions) choose to give their kid a 50% chance of a horrible and invariably fatal genetic disease rather than use it.

    Philosophical nitpick – this is an inaccurate description. Choosing an embryo with a genetic disease is not giving anyone a disease who could exist without it, since choosing a different embryo would result in a different person being born. A lot of philosophers, myself included, think that this is morally significant, at least in cases where the child with the disease still has a life worth living.

    • Winter Shaker says:

      I think Eneasz Brodski has a good reply to that (his old website seems to be broken but scroll down to “How to think about non-existent people” on the Google cache).

      …choosing to bring one child into existence is denying life to another child that would have been born in their stead […] And since almost nothing can be known of someone before they are born, in the aggregate it’s most accurate to think of the potential future-children of any given couple as undifferentiated entities. The replacing-person is best modeled as the same as the replaced-person EXCEPT for the things that can be known about them before they are brought into existence. If a genetic test shows that the egg released this month will give you a child with blue eyes, and the egg released next month will give a child with brown eyes, the question is not “Should we murder the child with the blue eyes or the child with brown eyes?” Because it is impossible to birth both of them. The question is more accurately modeled as “Do we want a baby Eneasz (or baby Steph) with blue eyes or brown eyes?” Think of the two potential children as the same potential person, differing only in the characteristics that can be determined beforehand. Thus, the question isn’t “Should we murder Mary Sue with Downs Syndrome to birth non-Downs Sally May?” it is “Should we birth Mary Sue with or without Downs Syndrome?”

      Edit: You are technically correct that it’s not the same person, but I don’t see how you can derive the claimed moral significance without counterbalancing the moral significance of *not* bringing into existence the other potential child, who would have a better chance of an enjoyable life.

      • Philosophisticat says:

        A very plausible thought about morality is that in order to do something wrong, you have to wrong someone. If you bring into being the less happy of two potential children, have you wronged anyone? Well, you haven’t wronged the child you brought into being, since they would not have existed otherwise, and they are happy to exist. Have you wronged the child who was not born? I think you have not, because someone who does not exist cannot be wronged.

        Now, one might want to claim that you can wrong people who do not exist, and that in fact if you fail to bring someone into existence who would have been happy you wrong the purely hypothetical person. But a) this is counterintuitive in its own right, and b) this has even more counterintuitive implications, like that deciding not to have children (and indeed, a great many children) is wrong.

        Some people will instead deny the idea that to do something wrong you have to wrong someone – the moral significance of failing to have the happier child doesn’t come from wronging anyone, but from not making the world as good as it could be or whatever. That view has its own set of problems. I recommend David Boonin’s “The Non-Identity Problem and the Ethics of Future People” for a longer (pretty thorough) discussion.

        It’s a difficult and in my view extremely fascinating topic.

        • Murphy says:

          Lets imagine I fly over a town spreading some chemical with horrible effects on released but not yet fertilized human eggs. Lets say it dramatically ups the chances of being born with many horrible painful deformities, learning difficulties and other problems. It also slightly alters which sperm is likely to get lucky so the people who end up existing have a different genetic makeup as well.

          When I acted the child brought into being didn’t exist.

          Who they are as people later in life depends on the brains they ended up with and those are only as they are because of my spraying.

          Very very different people would end up existing had I not sprayed that chemical. Those people ended up not existing.

          have I done anything wrong?

          If you can look at a room full of damaged children and agree that as they are screwed up because of what I chose to do but also declare that I in no way wronged them…. then you let a divide by zero error slip into your ethical calculations somewhere and ended up with absurd results.

          it implies that people are playing semantic games roughly equivalent to (and as valuable as) the troll-pi “Proof”

          https://qntm.org/files/trollpi/piequals4.png

          • Philosophisticat says:

            These are roughly the kinds of cases Parfit discusses to challenge the “to do something wrong you must wrong someone” principle. I do think that the view gets counterintuitive implications in those cases. It’s one of those problems where down every road is something counterintuitive. I’m with Boonin that ultimately this is the right bullet to bite.

          • JohnBuridan says:

            I think I’d bite the bullet on that one too.

            You did something that caused horrible defects for future generations. But the moral failure of that is your malevolence/ignorance, not your chemical spraying.

            I don’t blame the people of the past for using lead-based paint, even though that caused horrible defects in my neighborhood. But I do blame them if they knew the effects and used lead it in order to damage the kids of the neighborhood.

            Would you agree @Philosophisticat?

            I guess I’m willing to blame Moloch on this one…

        • Mary says:

          “If you bring into being the less happy of two potential children, have you wronged anyone?”

          Moot point. Both of them unquestionably have being at the moment of choice.

      • shenanigans24 says:

        The article gave an example where 13 couples agreed to terminate the pregnancy if the testing showed a disease. All 13 changed their minds. The moral question is too kill the child in pregnancy or let the disease kill it later in life. Apparently that question is easier prior to pregnancy.

        I don’t think Brodski addressed this point since he appears to be talking about a choice between getting pregnant with or without a child with a disease as a choice. I’m sure most people would choose without. But the choice they are faced with isn’t choosing which embryo is fertilized, it’s once fertilized what to do. That seems identical to the choice, once a baby is born, if you know it has a disease what should you do?

        • Murphy says:

          It’s only identical if you consider the blastocyst a person. If you don’t believe a person exists yet it becomes a matter of whether you choose to continue to the point where you have created that person with that condition.

          • Janet says:

            Well… any actually-existing embryo is certainly an individual, otherwise we wouldn’t be talking about its individual characteristics (such as specific genetic traits). Likewise, we wouldn’t be having this discussion if the embryo was not alive, or was not human (we do know specifically who are the parents, after all). So, the only remaining question is, will we, or will we not, extend the moral and legal status we give to other living, human individuals to this particular human individual?

            I don’t think that there’s any plausible way to say that some other person/people “create” a person with a particular trait (good or bad) at some time after conception. The individual human begins to exist at conception, and has (or does not have) the trait in question from that point forward: zygote, embryo, fetus, neonate, child, youth, adult; this is true whether or not the conception was the old-fashioned way, or in a petri dish. You can allow this individual’s biologic processes to continue to their natural conclusion, or you can destroy that individual (which is what’s in view here).

            That’s morally quite different than trying to ensure that sperm or eggs with desirable characteristics fuse– again, whether that’s the old-fashioned way of picking a desirable man/woman to have a child with, or by using some medical technology in the lab. Sperm and eggs are genuinely hypothetical individuals, not actual individuals, and so the moral reasoning above is feasible.

          • shenanigans24 says:

            Depends on if you considered a born baby a person also.

          • Janet says:

            Depends on if you considered a born baby a person also.

            Well… the law certainly does; and the great majority of the population finds infanticide to be a particularly heinous form of murder. You can have any interior considerations you’d like, I suppose; but if you want to stay out of prison, you’d best make your exterior actions conform seamlessly to “born babies are people”.

          • Toby Bartels says:

            @Janet : I don’t understand why a fertilized egg is absolutely an actual person, while an unfertilized egg is absolutely not an actual person. A fertilized egg is simply an unfertilized egg that has had something happen to it, namely, that it has incorporated a sperm, which in my view is just one step on the long road towards becoming a person. (To be sure, it’s a step with a profound effect on that hypothetical person, belied by the tiny size of the sperm, but that’s all.)

          • Murphy says:

            @Janet

            You’re just stating your cultural position as if it’s fact.

            Some historical cultures didn’t extend personhood to infants until some point some time after birth.
            Some cultures withheld personhood until the child was given a name. Our culture happens to use breathing air for the first time as the line in the sand. Otherwise it is not special.

            being human is not the same as being a person. My toe is human, 100% made of human, cells extracted from it could be used to grow a clone but it’s not a person.

            The potential for a person exists when 2 horny people lock eyes across the room. (remember, cockblocking your mates is thus murder since it could be preventing potential people from being born)

            Post fertilization the mothers body tries remarkably hard to murder the shit out of the zygote and stop it from implanting. (funny side note, in mammals the endometrium is about the most hostile tissue in the human body for a fertilized egg, they’re even more likely to implant an survive in brain tissue)

            Why isn’t implantation the special moment then, it’s when it’s passed the first test? (implantation being the line in the sand the catholic church chooses)

            Then the mothers body tries it’s best to starve the little fucker while the zygote tries to hack the control systems of the local tissues to force them to release more nutrients.

            And it’s not even certain how many individuals you’re going to end up with at conception, you could get 1,2,3 identical twins/triplets from the single fertilized egg…. or zero if any of the test are failed.

            Your line in the sand is not particularly special and there’s about a hundred other perfectly reasonable lines in the sand that we can draw and which many cultures in the past have drawn.

          • Janet says:

            @Toby Bartels:

            What I wrote was, the zygote is an actualindividual, whereas the unfertilized egg or sperm is merely a potential individual. That is, the zygote is biologically distinct from either mother or father from conception (for example, by having different histocompatibility than either one). It is not a tissue/organ/component of either the mother, or the father. Assuming a suitable environment and sufficient nutrition, the zygote will independently proceed by its own biologic processes through the various stages of development, over a period of years, up to adulthood (and, presumably, making new eggs or sperm for the next generation).

            None of that is true of unfertilized eggs or sperm; they are a tissue of the man or woman respectively. No matter how favorable the environment, they won’t develop by themselves into anything else (and in fact will die in a matter of hours to days). So, they are not individuals distinct from the man or woman, and Philosophisticat’s distinctions about harming individuals are relevant here.

            That’s all biology, and I’d say, fairly obvious at that. Now, the question of what is a human person, as distinct from a human individual, is a philosophical issue; whether the laws that protect other human individuals from being harmed, should or should not apply to these human individuals, is a legal and political question. But I do think that we need to be clear on the observable facts, before we go off philosophizing, or politicizing.

          • Murphy says:

            “None of that is true of unfertilized eggs or sperm; they are a tissue of the man or woman respectively.”

            Place a sperm and unfertilized egg in a suitable environment together and with sufficient nutrition and you’ll get a person.

            If you’re going to use histocompatibility as a measure of when someone is a separate person you’re opening a whole new can of worms.

            Or maybe 3 persons/individuals. Because it’s not definitely going to be just one person at that point. Or maybe it’s one of the zygotes that would have failed one of the tests before birth and been killed and absorbed by the mothers body. Or maybe that fertilised egg will turn into 2 persons but one individual body.

            (is it murder if a mother takes a drug that makes it less likely for a zygote to split into twins? they’re preventing potential individuals after all)

            Or maybe they’ll just become a teratoma with brain tissue, eyes and teeth sticking out at odd angles.

            Your line in the sand is arbitrary and not special in any way.

            When people throw the term “obvious” at things they’re often trying to make a supposition or opinion sound like something more concrete than it is.

            Funny side note: you’d be surprised how long a room full of biology professors can argue about seemingly obvious questions like “is X alive” or “is X one or 2 organisms”.

          • Janet says:

            @Murphy

            As in my reply to Toby, above, I’m not making a cultural statement, I’m making a biological statement.

            I do think it’s very interesting that both you and Toby switched my wording from “individual” to “person”, without seeming to notice. In fact, you’ve gone quite around your elbow to avoid addressing the issue of individuality.

            So, your argument about toes– toes are not individuals, they are component tissue of an individual; I think that the biology, philosophy and legal status of toes is pretty uncontroversial, and also pretty irrelevant to the specific issue we’re talking about here.

            Philosophisticat’s original argument was, among other things, the rebuttal to your “cockblocking” joke– you can go re-read it, but no, there’s no such thing as murdering a potential person. (Edit: also your subsequent comment about preventing a zygote splitting– you can’t murder a hypothetical.)

            Implantation is an important practical hurdle for the zygote to get over; but it shouldn’t be any surprise that a hostile environment can kill humans– including you and me, right now. Implantation does not, however, change anything about the individuality of the zygote, any more than escaping Hurricane Irma changed anything fundamental about the individuality of anybody in Texas.

            Yes, a zygote can split into two halves; usually that’s fatal, but if both parts manage to reestablish themselves as independently functioning individuals, then… you have two individuals. Their biologic paths will begin to diverge from each other at that point, due to differences in their environment; one can survive while the other does not, etc. That is, they are independently developing biologic organisms from that point, and we need to recognize that reality.

            Yes, different cultures have different laws and customs; but all of us humans share the same biologic facts, which is what I’m emphasizing here. Different cultures withhold legal or social recognition of human individuals for many reasons– but I think recent history should make you pause before you jump on that train. It’s going someplace you don’t want to go, I think.

            (One more edit: no, it wouldn’t surprise me to hear biologists arguing about minutia– I work with them all the time! But if any of us actually thought that the zygotes in question here weren’t alive, and human, and capable of developing independently, we wouldn’t be having this discussion.)

          • Murphy says:

            Sure, as long as we’re clear that individual isn’t being used to attempt equivocation between “individual” and “person”

            You were responding to a post talking about personhood that didn’t talk about individuality injecting the term “individual” as if you considered it synonymous.

            Best to clear that up because it’s standard pro-lifer tactic to use “individual” and “human” interchangeably with “person” and then pretend that everyone has agreed that there’s a person.

            The point with twins is that there isn’t one single “individual” from the point of conception onwards. N potential individuals where N is zero, 1, 1.something ,2 or more.

            There’s still only potential people. And as mentioned there’s no such thing as murdering a potential person.

            The potential is only slightly increased from the point where you had 1 egg and one sperm swimming around and we can still talk about the genes carried by that single egg and sperm that the individual could have.

          • The original Mr. X says:

            Some historical cultures didn’t extend personhood to infants until some point some time after birth.
            Some cultures withheld personhood until the child was given a name.

            And some cultures withheld personhood from humans with the wrong ethnic background. What’s your point?

          • JohnBuridan says:

            @Murphy

            It doesn’t seem Janet is stating a “cultural position” as fact. I think the idea is that fertilized embryos have genetic trajectory into adulthood, i.e. that is we can measure the probability of certain genetic diseases obtaining only once we have a ready-to-grow embryo. We can study the actual reality of that embryo’s genetic trajectory presumably(?) because it genetically distinct from its parents.

            This makes sense to me. But maybe there is something I don’t know I don’t know about The Process.

          • Murphy says:

            @JohnBuridan

            “genetic trajectory”

            That’s not all that interesting in it’s own right since the same can be said of a single separate egg and sperm sitting next to each other before they fuse.

            But of course that’s not a reason to break into a rendition of Every Sperm is Sacred.

            @The original Mr. X

            That it’s an entirely subjective matter when a society decides something counts as a person with many lines and everyone seems to think the random line they’ve settled on is special in some way. none of them are all that special and many get deluded that they’ve picked up on something profound.

          • The original Mr. X says:

            That it’s an entirely subjective matter when a society decides something counts as a person with many lines and everyone seems to think the random line they’ve settled on is special in some way. none of them are all that special and many get deluded that they’ve picked up on something profound.

            So if a society decided that, say, Jews weren’t people and it was OK to kill them, you don’t think they’d be wrong to do so?

          • Toby Bartels says:

            @Janet :

            >What I wrote was, the zygote is an actual*individual*, whereas the unfertilized egg or sperm is merely a potential individual.

            An unfertilized egg is an individual; it can be distinguished from all of the other eggs and from (the rest of) the person that it’s in (if it’s in a person at all). A fertilized egg is in exactly the same position. (I write more on individuality below.)

            >Assuming a suitable environment and sufficient nutrition, the zygote will independently proceed by its own biologic processes through the various stages of development, over a period of years, up to adulthood[.]

            And an unfertilized egg, given a suitable environment (including suitable nutrition, and including a suitable sperm[^1]) will potentially grow into a person (or up to 6 persons so far, I think), just like a fertilized egg will. (It’s a bit much to say that either of these will do so ‘independently’. The fertilized egg will only do so if it takes resources from the person (or test tube, in some cases, briefly) in which it grows, after which it will still need to use additional resources for several years to grow to adulthood; the unfertilized egg will only do so if it additionally takes a sperm first. But stuff all of this into ‘suitable environment’, and then it’s technically true in both cases.)

            [^1]: I’m inclined to include the sperm within the nutrition, since the egg consumes the sperm. It’s true that the sperm’s DNA is incorporated directly into the egg rather than broken down for spare parts or energy, but that’s pretty much the same way we treat essential fatty acids, and those are still usually thought of as nutrition. The process of eating the sperm is a unique event, and it makes sense to treat it as its own special thing, but nutrition or not, in the end it’s still a part of the egg’s biological process of interacting with its environment.

            >I do think it’s very interesting that both you and Toby switched my wording from “individual” to “person”, without seeming to notice.

            I did that because personhood seems to me to be what is relevant, while individuality is essentially meaningless. I realize now that you said ‘individual’ quite deliberately, so that was my mistake. (On rereading your first comment, I can clearly see the significance of your word choice.) So as to individuality:

            Individuality distinguishes one thing from another. An unfertilized egg is an individual unfertilized egg, a fertilized egg is an individual fertilized egg, a foetus is an individual foetus, a person is an individual person. This gives no information about their moral worth. If the question is whether an unfertilized egg is the same person as the one whose ovary it came from (or is still in), then I answer that it is not a person at all, and the same for a fertilized egg. If and when it becomes a person (or multiple persons), then of course it is a different person (or persons) from the parents. But personhood is the issue.

            As to whether it’s a separate biological organism (which I gather is what you mean by ‘human being’), then the problem is that reproduction blurs the concept of individuality (and contradicts the etymology of the term): what was one organism becomes (with the brief help of another) two (or more), and this is a process that takes place over a period of time. If biologists find it convenient to draw a hard dividing line at the moment of fertilization, then I can see why that would be useful, especially since so much depends on an organism’s genetic makeup. But that can’t be all that there is to it, since this then allows for one individual to become two (or more) again, as with identical twins, or even for two to become one, as with a genetic chimaera. So it might be simpler to draw the dividing line somewhere after these shenanigans are possible, perhaps at implantation. I don’t know what the practice among mammalian biologists is, but I don’t particularly care, since this is not the issue.

            To see again that biological standards of individuality for organisms is not what matters, I *do* know that the practice among herpetologists is to speak of a two-headed snake as one snake. I don’t know if primatologists do the same for two-headed monkeys[^2], but it doesn’t matter: the Hensel twins are two separate *persons* either way. If I discovered that primatologists use the same language as herpetologists, I would not conclude that it’s OK for one conjoined twin to kill the other, or that a third party who killed one would be guilty only of battery and not murder. It’s not just that this information about the practice of biology would be outweighed by other considerations; it would be completely irrelevant, since they are still separate persons. So likewise, if I discovered that biologists universally think of a fertilized egg as a separate individual organism, that would be irrelevant to moral questions around embryo selection and the like, since a fertilized egg is still not a person. (If selection is done at a later stage, then to answer moral questions, you have to examine what exists at that stage. A blastocyst is definitely not a person, and indeed I’d be shocked if anything is that can live in the proverbial test tube, but I’m willing to look at any evidence you might have regarding pain sensation, capacity for thought, or whatever. I don’t believe that late-term abortion is morally neutral, but that’s not what we’re talking about here.)

            [^2]: I tried to find out, but my Google results kept getting invaded by news about Sergio Canavero. But it’s fitting that I don’t know, since my point is that it doesn’t matter.

            Anyway, all of this is what I avoided saying by changing ‘individual’ to ‘person’. I thought that I was simply replacing the wrong word with the right word, but now I see that you used the wrong word deliberately, to refer to the wrong concept, so now you have three paragraphs (and a footnote) about why it is, in my opinion, an irrelevant concept (and also why it is not as straightforward as you said either).

          • pdbarnlsey says:

            The individual human begins to exist at conception, and has (or does not have) the trait in question from that point forward: zygote, embryo, fetus, neonate, child, youth, adult;

            And yet I’ve never seen a blue-eyed embryo.

            Likewise, we wouldn’t be having this discussion if the embryo was not alive, or was not human (we do know specifically who are the parents, after all).

            Arguing “any living thing with human parents is a human” just begs the question. It might well be, or we might insist on a more demanding definition of humanity than “is alive and is made of humans”.

          • The original Mr. X says:

            Arguing “any living thing with human parents is a human” just begs the question. It might well be, or we might insist on a more demanding definition of humanity than “is alive and is made of humans”.

            Perhaps “we” could try solving this question by seeing if there are any biologists out there who would deny that the living offspring of two humans is itself a human.

          • Creutzer says:

            I have a very strong linguistic intuiton that an embryo is human, but not a human. It seems plausible that the difference is genetic material vs personhood.

          • Murphy says:

            @The original Mr. X

            “living offspring of two humans”

            I’m not a biologist but I am a bioinformatician and I’d be fairly happy to make the statement that some things , for example a teratoma, even a diploid teratoma that may be nominally living and made up of 100% human tissue nominally with genetic material from 2 human parents isn’t “a human”, even if it’s one of those really horrifying ones with sort of pulsing deformed heart tissue and some lumps of neural tissue. I absolutely do not consider such things to be “a human” at all and any definition which makes them such I would consider faulty and not merely culturally subjective.

            Cases of complete acephaly (infants born without a brain: http://4.bp.blogspot.com/_sWAF7gFdwW4/S_w2Nh0r6aI/AAAAAAAAABE/yYSXgd_xohc/s1600/med_sketch.gif ) I’m also reasonably comfortable considering it human but not “a human” since even if they’re born breathing they’re effectively born a corpse-doll missing a brain and the only moral issues related to them revolve around the effect on living humans with brains. If the parent wants to play dress up with their living corpse-doll to help deal with the emotional pain of what is effectively a breathing still-birth fine and dandy, if they want to donate it’s organs to some other infants or throw it in a wood chipper, also no moral issues there.

            But the latter may be more a case of the difference between [biologically living] which would include a corpse braindead but on life support or an infant born with complete acephaly vs [living:legal status] which may or may not include either of those.

        • Davide S. says:

          @Janet

          Infanticide is a crime, but while the general population might consider it an especially heinous form of murder, the actual law seems to disagree:
          https://en.wikipedia.org/wiki/Infanticide#Current_law

          However the justification for this doesn’t seem to be that the baby is less of a person, but that many (most?) infanticides are done by mothers who are presumably not of sound mind.

          • Douglas Knight says:

            The reason that England treats infanticide lightly is that juries refuse to convict it as murder. I think the general public seeing infanticide as especially heinous is a recent development, though maybe it’s cheap talk.

    • Ketil says:

      Are we back to Copenhagen Ethics? If you meddle, you kill the would-have-been-born, but if you close your eyes and leave things alone, it’s not your fault that the other would-have-been-born weren’t?

      • Philosophisticat says:

        That’s not really it. If your failing to meddle would lead to a person being worse off than they would otherwise have been, my view is happy to blame you for it. The point here is that that’s not true.

        My view also entails (I acknowledge, counterintuitively) that it’s permissible to meddle and choose an embryo that will be worse.

  33. Роман Плотников says:

    Never tried benzodiazepines, but Hydroxyzine is a super-powerful anxiolytic and, surprisingly, just an antihistamine. Also, there is no addiction and I do not think it is possible to abuse it or overdose. Naturally, in my country it requires an official recipe, but you can freely buy flasks of phenobarbital solution for pennies and chuck them down until your heart stops.

  34. cvxxcvcxbxvcbx says:

    Re: the Georgia ACLU case, I went from mildly surprised that the ACLU apparently pursues cases in Eastern Europe to quite disappointed very quickly.

    • Winter Shaker says:

      It’s not even as if the Kartvelians call themselves anything at all to do with ‘George’. If we just changed the English name of the country to something based on the name it’s called by the people who actually live there, this whole annoying confusion could be avoided.

      • Watchman says:

        That sort of logical proposition is what takes all the fun out of life…

      • Peter says:

        Well, I think[1] some of them call it Gürcüstan or Վրաստան (Vrastan) or Грузия (Gruzia); OK they’re small minorities whose languages aren’t official, it’s not like Belgium where calling it Belgique or België while speaking English would be taking sides, but still.

        [1] As in, I did some quick web research of languages spoken in Georgia and then applied Google Translate to the problem. If I’m really unlucky I may have translated the wrong Georgia…

        • Winter Shaker says:

          Okay, but that’s like the English or Scots-speaking inhabitants of Scotland calling it ‘Scotland’ and the Scots-Gaelic speaking inhabitants calling it ‘Alba’ (apparently related to the term ‘Albion‘ used as a poetic name for England, coming from a Celtic word which used to refer to the whole of Great Britain).

          If you agree that it is most reasonable for the Georgians to say შოტლანდია (Shotlandia), then I think that something like ‘Kartvelia’ would be a good option.

          • Deiseach says:

            Speaking of Georgia, the gorgeous folk tune which I first heard in the 1979 Warner Herzog version of “Nosferatu” and sampled by Kate Bush on her song “Hello Earth”, and couldn’t find out who sang it or what it was until the Internet told me. Variant spelling of all kinds, including but not confined to “Zinzkaro, Tsintsharo, Tshintsharo, Cin Ckaro”. Version in the movie by Vocal Ansambl Gordela, version here by another (or possibly the same?) vocalist, equally beautiful.

          • Winter Shaker says:

            And none of those spelling variants even give any hint at the ejective versus non-ejective consonants 🙂
            წინწყარო – tsintsq’aro according to the transliteration system I used to learn the alphabet (please note I cannot actually ‘read’ georgian, just painstakingly figure out words letter by letter, hoping that by the time I reach the end I can still remember how it started)

          • While on the subject of Georgia …

            What struck me on my visit was that it is the only country I have been in where, judged by casual observation, female libertarians outnumber male libertarians. I suggested the export possibilities to them but don’t know if they have taken any action on them.

      • Lillian says:

        The English version of the Georgian constitution clearly states in Article I that the name of the country is Georgia. So in English, the Kartvelians do in fact call themselves Georgians.

        http://csb.gov.ge/uploads/2081806.pdf

        • Winter Shaker says:

          Presumably they thought it wasn’t worth the bother trying to insist that the Anglophone world change its name for them (though I’m not sure how hard it would actually be – it’s hard to assess off the top of my head how often I hear ‘Myanmar’ vs ‘Burma’ these days, and whether the ratio is changing).

          But I am surprised that they have an official translation into a language that (I assume) isn’t an official language within their borders. I wonder how many other countries do that.

          (My own country famously doesn’t have a single written constitution all, so even if we wanted to, it would be hard to even decide what to translate – but I’m pretty sure that if we did, we’d have it in Welsh and maybe Gaelic long before we had it in any Caucasian languages)

          • SEE says:

            And yet countries do go ahead and successfully insist the Anglophone world change its name for them all the time.

            The most extreme case here is Côte d’Ivoire, which, while not perfectly successful, has still managed to get significant Anglophone adoption of a French name with an “ô” in it to replace the far more natural-to-English literal translation “Ivory Coast”. But there’s also cases like Burkina Faso (Upper Volta), Sri Lanka (Ceylon), Iran (Persia), Thailand (Siam), Congo (Zaire (Congo)) . . . .

          • Winter Shaker says:

            Good point. Burma / Myanmar is the only one that’s happened while I’ve been alive / old enough to be aware of it. Though am I correct in thinking that all of these cases are countries that were either newly independent, or, in the case of Iran and Thailand, newly under the control of a radical ideology seeking a break with the past? I guess the Georgians missed their chance with the breakup of the Soviet Union, but hey, I’ll call it Kartvelia if you will 😛

          • Futhington says:

            @Winter Shaker

            Iran’s name change happened in ’35, when Reza Pahlavi had been on the throne for about ten years. Thailand in the late ’30’s (when the Siam name was changed) was going through a series of dictatorships, but their monarchy has remained relatively stable so I don’t know how radical things really were.

            Generally speaking I think you’re giving an unfair image of the west as not receptive to the desires of other nations to be called what they want. If Georgia and the Georgians genuinely take issue with being called that I’m pretty sure we’d alter matters overnight.

            There’s actually some fascinating ethnic politics behind the Myanmar/Burma issue that makes it controversial within their own country. Plus international issues, the US for instance withheld acknowledgement of the change over not recognizing the military government that changed it as legitimate, while the UN endorsed it within a week.

          • SEE says:

            Ivory Coast was independent for decades when it just got fed up with the huge transformations of the name depending on the language translated into. No more ‘Ivory Coast’ in English and ‘Kōtojibowāru’ in Japanese and ‘Costa de Marfil’ in Spanish and ‘Xiàngyá hǎi’àn’ in Chinese and whatever, just Côte d’Ivoire everywhere!

          • Toby Bartels says:

            On the other hand, Czechia can’t get anglophones to stop saying ‘Czech Republic’. Part of the problem here is that ‘Czech Republic’ is still the correct formal name in English (analogous to ‘French Republic’), while ‘Czechia’ is merely the official English short name (analogous to ‘France’). So it’s never wrong to use the formal name, but it’s kind of weird if you keep using the short name for every country *except* this one.

          • The original Mr. X says:

            On the other hand, Czechia can’t get anglophones to stop saying ‘Czech Republic’. Part of the problem here is that ‘Czech Republic’ is still the correct formal name in English (analogous to ‘French Republic’), while ‘Czechia’ is merely the official English short name (analogous to ‘France’). So it’s never wrong to use the formal name, but it’s kind of weird if you keep using the short name for every country *except* this one.

            To be fair, “Czechia” in English sounds almost exactly like “Chechnya”, so there’s the potential for confusion if the short name takes off. (Personally I’d call the place “Czechland” or something similar, since it’s less confusing than “Czechia” and less of a mouthful than “The Czech Republic”.)

          • Toby Bartels says:

            >Czechia” in English sounds almost exactly like “Chechnya”

            Maybe if they pronounce it incorrectly, but then ‘Czech Republic’ sounds like ‘Chechen Republic’ (although not as much so since the extra /n/ isn’t buried in the middle). Which I guess is a good reason to not use the formal name of Chechnya.

            Anyway, while the ‘cz’ in ‘Czechia’ sounds like the ‘ch’ in ‘Chechnya’, the ‘ch’ does not.

          • The original Mr. X says:

            Maybe if they pronounce it incorrectly, but then ‘Czech Republic’ sounds like ‘Chechen Republic’ (although not as much so since the extra /n/ isn’t buried in the middle). Which I guess is a good reason to not use the formal name of Chechnya.
            Anyway, while the ‘cz’ in ‘Czechia’ sounds like the ‘ch’ in ‘Chechnya’, the ‘ch’ does not.

            If the Czechs want the English-speaking world to change its spelling and orthography to accommodate them, good luck to them, but they might be waiting a considerable time.

          • The Nybbler says:

            “Czechia” doesn’t sound right to this American English speaker. I’m not sure what’s wrong with it, but it’s definitely off.

          • Creutzer says:

            I suspect there may be an issue with the vowel “e” in “Czechia” that makes it weird. I don’t know why, but I have an intuition that the vowel before “-kia” should be long, like in “Slovakia”. But then you end up with “chaykia”, which isn’t like “czech” anymore and therefore weird. But “checkia” with the short vowel is also weird. I wish I knew why.

          • Matt M says:

            I’ve literally never heard of Chechia before but I will enthusiastically adopt this immediately. Czech Republic always sounded really dumb to me.

          • Toby Bartels says:

            @The original Mr. X :

            >If the Czechs want the English-speaking world to change its spelling and orthography to accommodate them, good luck to them, but they might be waiting a considerable time.

            On the contrary, English is famous for accommodating its orthography to whatever language it’s borrowing from. We usually insist on using only our normal letters and phonemes, but we don’t require the spelling and pronunciation to match. (For example, China didn’t have much trouble getting us to change ‘Peking’ to ‘Beijing’, with pronunciation rules following Pinyin rather than normal English.) Native Anglophones may not learn other languages much, but we’re really good at borrowing from them.

            In this case, however, there’s no change. The ‘Czech’ element in ‘Czechia’ is pronounced in the usual way (for English), the same as in ‘Czech Republic’. They’re not asking us to say ‘Czesko’ (the short name in the Czech language), although I’m sure that we could if we tried.

          • Creutzer says:

            For example, China didn’t have much trouble getting us to change ‘Peking’ to ‘Beijing’, with pronunciation rules following Pinyin rather than normal English.

            To be fair, if you map Mandarin sounds to the closest English sounds, then the two pronunciation rules happen to coincide in this case.

          • Matt M says:

            And yet, Cow Cow vs Tsao Tsao is still an unending source of contention for the Dynasty Warriors video game series.

          • Toby Bartels says:

            @Creutzer :

            >To be fair, if you map Mandarin sounds to the closest English sounds, then the two pronunciation rules happen to coincide in this case.

            I’m not sure what you’re being fair about here. If you mean that English borrowed the pronunciation as closely as possible from the original, then I agree, and that’s half of my point: that English does this quite readily. The original pronunciation doesn’t always stick around, but it usually lasts at least for a while.

            The other half is that English will do this even if the spelling does not match the pronunciation by English’s normal spelling and pronunciation conventions. One would not expect this pronunciation of ‘Beijing’ from the spelling, nor the other way around, but that didn’t stop them from being adopted.

          • Jiro says:

            Dynasty Warriors is a Japanese series that contains Chinese names. Japanese uses Chinese characters and pronounces them differently. If the English translator isn’t savvy enough, he’ll translate the Japanese-pronounced version into English rather than going back to the original source. And an awful lot of English translators aren’t very savvy.

            This is even worse in the opposite direction. Translating form Japanese to Chinese and then to English is likely to mangle the names.

          • The original Mr. X says:

            On the contrary, English is famous for accommodating its orthography to whatever language it’s borrowing from. We usually insist on using only our normal letters and phonemes, but we don’t require the spelling and pronunciation to match.

            On the other hand, I’ve never, ever heard an English person pronounce Citroën or Volkswagen as they’d be pronounced in French or German.

            In this case, however, there’s no change. The ‘Czech’ element in ‘Czechia’ is pronounced in the usual way (for English), the same as in ‘Czech Republic’.

            Since the -ch- isn’t at the end of the word, the “usual way (for English)” of pronouncing Czechia would be something like “Che-chee-a”, not “Check-ee-a”.

          • Toby Bartels says:

            >I’ve never, ever heard an English person pronounce Citroën or Volkswagen as they’d be pronounced in French or German.

            Good point. I don’t think that I’ve ever heard anyone pronounce ‘Citroën’ at all, but Americans (at least) definitely don’t pronounce ‘Volkswagen’ using English versions of the German sounds, instead using a pronunciation that matches the spelling. I’d like to say that this is because the name has been around for too long, but it’s not *that* long (several decades but less than a century), and I don’t actually know how people pronounced it when it first came out.

            >Since the -ch- isn’t at the end of the word, the “usual way (for English)” of pronouncing Czechia would be something like “Che-chee-a”, not “Check-ee-a”.

            I don’t see how this makes a difference. There’s no rule in English that ‘ch’ can only sound /k/ at the end of a word, and in fact the most common position for that letter-sound is at the beginning of a word (for Greco-Latin borrowings such as ‘character’).

            The rule that I’m expecting people to apply is for the entire morpheme ‘Czech’. This morpheme keeps its pronunciation even when it’s not at the end of a word, as in ‘Czechoslovakia’. There’s no reason to treat it differently in ‘Czechia’.

          • The original Mr. X says:

            The rule that I’m expecting people to apply is for the entire morpheme ‘Czech’. This morpheme keeps its pronunciation even when it’s not at the end of a word, as in ‘Czechoslovakia’. There’s no reason to treat it differently in ‘Czechia’.

            Where I come from, the ch in Czechoslovakia is pronounced like a ch, not a k.

          • Toby Bartels says:

            Wow, I’ve never heard that! Well, that explains our disagreement then. May I ask where you’re from?

        • S_J says:

          There is a group of people who are, in their own language, the Έλληνες (approximately Éllines).

          The culture is ancient. At one time a military leader from this culture managed conquer a neighboring empire, and create systems of education that spread the language and culture across most of the Fertile Crescent and the Mediterranean basin. After the heyday of that empire, another empire sprung up. The people of the successor empire called the Éllines by the name Graecia.

          Thus, most English-speakers call the Ελληνική Δημοκρατία (Éllineke Demokratia) by the name Republic of Greece. Most Latin-derived languages use some variant of Graecia as the identifier for the nation/language.

          I wish there were a clean way of figuring out what is the proper name for each linguistic/cultural group, or associated nation.

          • Creutzer says:

            Those ancient people only very “approximately” called themselves Éllines. That is, in fact, what people who now live in the place where they originated call them. They actually had an h at the beginning of the word and something resembling the vowel in “bad”, but long, in place of the i.

          • AlphaGamma says:

            As your post suggests, the Modern Greek word for “republic” is Demokratia, which is also the word for “democracy”. This causes some issues where phrases include both words and must be translated.

            For instance, Greek discussions of US politics refer to the Democratic party as the Demokratiko Komma (a direct translation) and to the Republican Party as the Repoublikaniko Komma (an otherwise-meaningless loan word).

            The Democratic Republic of the Congo is translated into Greek as Laiki Demokratia tou Kongo (“People’s Republic of the Congo”, the same phrasing used for the PRC). The DPRK requires further tautology, so is Laokratiki Demokratia tis Koreas. This means “People-Ruled Republic of Korea”, it sounds much weirder in Greek once you consider the meaning of the word “democracy”. Laokratia (“folk-rule”, perhaps) as opposed to demokratia (“people-rule”) was the term used by the Communist EAM resistance movement to describe the political system in territories it controlled during WW2.

            Of course, Graeci comes from the name used for themselves by Greek colonists in southern Italy. The name of another group of colonists gave the name used for the Greeks by various cultures to their East. For instance, the Turkish word for Greece is Yunanistan (though the Greek people are also called Rum, “Romans”, in other words Orthodox Christians), the Persian for Greek was IIRC Yona and the Hebrew Yevanim– Yevanic is the name of the now almost-extinct Judaeo-Greek spoken by Greece’s Romaniote Jewish community who predate the influx of Sephardim in 1492.

            [NOTE: In this post, I have followed the transliteration convention used by S_J, which gives more information as to which English words a Greek word is related to, and less as to how it is pronounced today].

  35. SEE says:

    I’m ex-Catholic rather than still Catholic, but the Pope-and-celibacy story is, well, not horribly inaccurate?

    Pope Francis, approached by a Brazilian bishop on the issue, told that bishop to bring it up with his national conference of bishops first, rather than simply shooting it down. That’s not the same thing as saying that he’s actually open to it, but it would be a first step to relaxing the discipline in places where the national conference of bishops decides it would be a good idea.

    He has also said that “optional celibacy is not a solution” to the shortage of priests in an interview with a German newspaper, but he did go on to possibly hint that the class of married men currently allowed to become deacons (the viri probati) could help solve it, vaguely enough that he might be hinting that he would be willing to ordain them. Alternatively, he might just be suggesting recruiting and deploying them to oversee parishes where priests aren’t available.

    There of course are the existing provisions for the Eastern particular churches to have married priests already, and for married pastors of Anglican and Anglican-descended denominations to become Catholic priests. The Pope has relaxed the rule prohibiting Eastern Catholic priests from serving outside their home country, so there are now some more existing married priests potentially available to be assigned to places with shortages.

    • JohnBuridan says:

      Good answer. I know in my diocese there are two parishes which are testing giving deacons more authority. A married deacon runs the parish on a day-to-day basis, keeps track of grounds, payments, renting out space, employees, etc., while the priest gets a monthly report and sticks to priestly duties (Sacraments, prayers, funerals). Technically, the priest is still in charge, but he can take care of greater number of religious duties by having the managerial side of his job go to an operations manager (the deacon).

    • Anonymous says:

      the Pope-and-celibacy story is, well, not horribly inaccurate?

      Sounds like a standard piece on the Pope, then.

      • SEE says:

        Or, well, on pretty much anything else. Always remember to account for the Gell-Mann Amnesia effect.

        • Anonymous says:

          Certainly, certainly. Though I’d put “the Pope” on the media’s list of topics they are particularly egregious in botching every damn time.

          • SEE says:

            I find I can flip myself between the propositions “Yeah, it does wind up worse because the reporters know they don’t understand, say, quantum physics, but think they understand the Church,” and “No, it isn’t actually worse, it just that Gell-Mann Amnesia is a relative effect rather than binary,” fairly easily.

  36. Chalid says:

    But these doctors can’t say “We are frauds who give everyone marijuana cards”, so lots of marijuana-wanting and even marijuana-needing people won’t go to the doctor because they’re afraid they’ll get turned down.

    I guess this must vary by state and locality. I remember passing by this place on vacation a few years back. They’re not quite explicitly saying “we are frauds who give everyone marijuana cards” but anyone with a room-temperature IQ should be able to figure it out.

    (For those who don’t want to click the picture: the place is called “The Green Doctors” and has a marijuana leaf logo prominently displayed along with the text “$40 medical marijuana evaluation”. ISTR were also employees in green shirts with marijuana logos outside letting passers-by know exactly what medical conditions might justify a medical marijuana prescription.)

    Anyway, even if this sort of blatant behavior isn’t allowed in most states, surely a prospective “patient” could figure out which clinics to try with the tiniest bit of googling? MarijuanaDoctors.com is the top hit for “marijuana doctor”, and it has a “find a doctor” page with starred reviews by users and clinic self-descriptions like, for example, “The Holistic Clinic was created with one goal in mind and that was to introduce inquiring Massachusetts medical marijuana patients to the exposure of top-notch MA Board-Certified medical marijuana doctors.”

    • Ratte says:

      To be honest, it’s more than a little concerning how easily and openly these essentially fraudulent doctors can operate without censure. You’d think that blatantly advertising diagnosis-for-hire would draw scrutiny from the licensing boards, even in this context, but no. Makes me wonder what else is being overlooked (I’ve heard there’s a similar racket servicing the plaintiffs’ bar).

  37. Doug says:

    Contra the neural net link, Inference.vc made an interesting point. The maximal activation is a very close concept to the mode of a DNN classifier. We’re used to thinking of the mode of a distribution as being a pretty good representative of the class. E.g. with height/weight, 5’9″ 170 lbs seems like a very realistic example. But this is only true because we’re used to dealing with low-dimensional distributions.

    In very high probability spaces (like an image, where each pixel is a separate dimension), the mode is very atypical of normal sample from the distribution. Most of the mass of the probability distribution lives on a thin shell, which is E(||Norm||) away from the center. So even if the maximal example seems very strange, that doesn’t necessarily indicate that the underlying generative model is pathological.

    http://www.inference.vc/high-dimensional-gaussian-distributions-are-soap-bubble/

    • lycotic says:

      FWIW, the attribution algorithm that we’re currently using is in this paper: Axiomatic Attribution for Deep Networks, which integrates the gradients as the input moves from some “zero” state (a blank picture, say) to the given input.

      This carries a lot more weight than the weird deep-dream effects of looking at individual activations.

    • cactus head says:

      This weird behaviour of probability distributions in high-dimensional spaces is something that also needs to be accounted for in Markov Chain Monte Carlo statistical methods.

      Sections 1.3 and 1.4 of this paper explain the problem in a little more detail: A Conceptual Introduction to Hamiltonian Monte Carlo by Michael Betancourt.

  38. The Arcadian says:

    The sci-hub.io link already leads to a DNS failure.

  39. theory says:

    Your description of the “lawyer dawg” link is inaccurate (or more accurately, perpetuates an inaccurate summary of what happened).

    There’s a good breakdown here: https://www.washingtonpost.com/news/volokh-conspiracy/wp/2017/11/03/the-lawyer-dog-decision-isnt-obviously-wrong/?utm_term=.2e1aad239cbe

    The TLDR is that the ambiguity is almost certainly NOT the “lawyer dog” part of the sentence, but what was said immediately before it. The opinion is unfortunately ambiguously written such that people fixated on the “lawyer dog” portion. The opinion was also only a concurrence, i.e., it did not impact the majority’s actual denial of cert, which relied on other factors.

  40. Yes, the Streisand effect you are helping to produce would be “bad.” But even worse is the fact that my ISP appears to have already blocked it. Going there gets “server not found.”

    • Nornagest says:

      Try https://sci-hub.bz/. Wikipedia states that sci-hub.io exists, but can only be accessed when you configure your browser to use the custom nameservers 80.82.77.83 or 80.82.77.84, which is creative but probably ineffective if I can find out how to do it so easily.

  41. Delighted to hear that Stephen Hsu is working on the problem of purging the tall genes that have infiltrated my gene pool. Also, apropos of the previous discussion of standing in airplanes, making the population shorter is obviously a more efficient solution to the problem of airplane capacity.

    • pontifex says:

      Wasn’t there a transhumanist who was advocating in favor of creating “tiny people” in order to save on resource costs, rocket ship fuel, etc. etc.? I’m serious about this, I saw this idea somewhere.

      I’m fine with this as long as myself and my tall descendents get to rule over the oompa-loompas in the chocolate factory benevolently guide the destiny of the proletariat.

  42. Deiseach says:

    Paul Palaiologos Tagaris reminds me of the Irish version, Myler McGrath. Franciscan friar who was ordained a priest, was appointed bishop of an Ulster diocese by the Vatican but, due to the delays caused by travel times in those days, he had taken the opportunity offered by the new English regime to convert to Anglicanism (the Church of Ireland), allegedly in hopes of being offered the position of Protestant bishop of that same diocese. When this did not happen, he held on to the Catholic office and was eventually offered a bishopric in the Protestant Church of Ireland in another county. It took nine years, during which he was double-jobbing as a Catholic and Protestant bishop, before he was deprived of his see by the Vatican. Played both sides off against each other by keeping in with both the Irish (through his family connections) rebels and the English government in Ireland and in England, kept dangling the possibility of renouncing Protestantism and returning to the Catholic faith before his Irish connections (and at the same time kept implicitly threatening the English that if he didn’t get the good, valuable dioceses he felt he was owed, he’d make an embarrassing to them public return to Catholicism so they had to pay to keep him sweet), fathered a large family (nine children), and eventually died in bed at the age of 100, in which faith (or both, or neither) is unknown.

    it was a regulation saying that all meal replacements have to be low-fat

    And that is precisely why I think Nathan Robinson’s notional American Diners/Government Cafés would not work out to be “cheerful, pleasant places where you can get tasty, healthy food with a good range of choice and no stigma!” It would be the perfect opportunity for every crank and faddist and busy-body with a PhD or nutritionist or OBESITY EPIDEMIC! CRISIS!!! alarmist to stick their nose in and indulge their “the poor are too stupid and/or lazy to do what they’re supposed do, in spite of all the good advice we give them, so now we’re going to make them be virtuous” instincts. Insipid, bland, sludge where you can have the choice of the grey sludge or the beige sludge, “nutritionally optimised” to be low-fat, low-sugar, and low-everything that makes food palatable in calorie-counted, portion-controlled servings and all supplied by contractors who won the government tenders (so that means industrial agribusiness cheap and mass produced items). All dispensed in centres that will be more reminiscent of modern-day workhouses (remember the scene in Oliver Twist about the beadle and the boy daring to ask for more) to hammer it in that you are a parasite on society existing on the hard-earned taxes of good hard-working people and moreover need to be made to do what will make/keep you healthy and fit to become a productive, working citizen (if the idea is that a Republican administration would take the opportunity to enrich their cronies by giving them contracts to supply/run such Diners, trust me, the Democrat administration would equally be besieged by “this is the perfect opportunity to promote veganism on a large scale and force ethical behaviour down the throats of the ungrateful reduce the harm caused by and to animals in large scale livestock production, over-fishing, and the methane output of cattle contributing to climate change”).

    This week’s ridiculous non-controversy: Christians are boycotting British food producer Greggs for making a nativity scene with a sausage roll as Jesus.

    Oh those crazy, unreasonable Christians! Whatever is the matter with an Advent calendar depicting the traditional scene of the Three Wise Men adoring the infant saviour Greggs’ sausage roll laid in the manger? Next year, Greggs go inter-faith, with their Rosh Hashanah version featuring a happy family gathered at the table for the seder tucking into Greggs sausage rolls instead of the traditional foods such as apples in honey! And for the Muslim equivalent, end the Ramadan fast on Eid al-Fitr with Greggs’ tasty sausage rolls! Nobody could possibly object to those, or if they did, it would be a ridiculous non-controversy whipped up by a few fanatics and zealots, right?

    (Somebody on here in a previous comment thread mentioned that they should have shown the Three Wise Men bearing gifts of sausage rolls instead of gold, frankincense and myrrh, and that would have been a lot funnier and more sensible instead).

  43. Deiseach says:

    Tylenol/ibuprofen as good as opiates for acute pain relief.

    That must be one heck of a placebo effect, because personally ibuprofen does bugger-all for even ordinary aches and pains as far as I’m concerned (paracetamol/acetaminophen is middling, aspirin works best but murders my stomach).

    If I were in really bad pain and some doctor gave me ibuprofen on the basis of this study, there would be murder done.

    • Harry Maurice Johnston says:

      FWIW, ibuprofin works much better if taken regularly. It takes several days IIRC for the full effect to kick in, and you have to be sure not to miss any doses. Personally, when I was dealing with chronic pain, I found that the slow-release tablets worked best for me. I think that was mainly because they meant I didn’t need to get up in the middle of the night to take another dose, though my general absent-mindedness might also have played a role. 🙂

      (My pain wasn’t particularly severe, nowhere near the point where opiates might have been an option, mind you. And there’s a limit to how long you can take ibuprofin, although I have the vague impression that in some cases you can be given other medication to limit the side-effects.)

      • Deiseach says:

        When I have a pain in my side the severity of which is such that stabbing myself with a steak knife in that region to relieve pressure begins to seem like a reasonable idea*, “take this for three days because it won’t work immediately and it will take that long to kick in” will get me stabbing with a steak knife, sure enough, but the person who gave me that advice rather than my side.

        *This was a bout that ended in one of my Misadventures in the Local A&E Department, where I decided to go instead of stabbing myself in the side. It wasn’t appendicitis but they never did find out what it was due to – well, it’s a long story, but I only know it wasn’t appendicitis as I didn’t die from it, not thanks to any medical/surgical knowledge, and stabbing myself in the side might have been the better option after all.

        • Harry Maurice Johnston says:

          Yes, I quite agree – I just wanted to mention that ibuprofin works a lot better than you might imagine if you’ve only ever used it casually. Not really relevant. My bad.

  44. herbert herberson says:

    This is the world that all of you “free speech only constrains the government” and “it’s just people think you’re an asshole and are showing you the door” people have built for us.

    No, this is the world we have always lived in, and if you think that it would be any different if more people on the left were free-speech absolutists you’re dreaming.

    I don’t like seeing people fired from their jobs for their politics, and if all the free-speech hand-wringers want to statutorily guarantee a right to not be fired for political speech, I am behind it and I think most socialists and liberals would be as well. But (tellingly) this has not been proposed, and if all you’re offering is vague social norms I have absolutely zero confidence it will be reliably reciprocated. Certainly, the First Amendment has done very little historically to protect the left (see, e.g., Debs, Emma Goldman, and the fact that it’s still technically illegal to be a member of a Communist Party), and the GOP as it currently exists has reliably taken good faith offers to compromise or uphold principles as nothing more or less than a display of vulnerability.

    • AnonYEmous says:

      No, this is the world we have always lived in, and if you think that it would be any different if more people on the left were free-speech absolutists you’re dreaming.

      More like, “if more people on the left weren’t eager to tear down the 1st amendment”. It’s not even about being against racists fired, so much as not making websites dedicated to getting racists fired, or so forth.

      Vague social norms are probably all we have, because otherwise it’s kind of hard to prove. Sadly, I agree it probably won’t be reciprocated; it’s far too late for that. But maybe we can try, I guess? I don’t know, I go back and forth between these stances.

      • herbert herberson says:

        With the exceptions of positions where an employee has serious authority over members of the general public (cops and public school teachers) I will condemn anyone who gets anyone fired over politics. It’s nothing more or less than being a scab, collaborating with management and capital against a worker, and that doesn’t change just because the worker is a piece of garbage. Anyone who pretends to be a socialist or even just a progressive who does such a thing should be fuckin’ ashamed of themself.

        But the “ho-boy you naive lefties will be sorry if you stop respecting free speech” line of argument is totally ignorant of the history of how this country has always treated leftists, and far too credulous towards the right’s sense of fairness. Unless there’s any Jordan Peterson videos out there condemning the prosecution of the J20 I missed?

      • herbert herberson says:

        Also, while you’re absolutely right that it is extremely easy to pretend you are firing someone for a reason other than a legally prohibited one:
        a.) merely making them cover their tracks would encourage cultural change
        b.) it would be far more difficult to maintain that pretense in the sorts of public-controversy situations we’re talking about here

        • AnonYEmous says:

          but also c.) people would say it was happening even if it wasn’t, contributing to the creep of “it’s hard to fire people even if they clearly deserve it”

          also:

          With the exceptions of positions where an employee has serious authority over members of the general public (cops and public school teachers) I will condemn anyone who gets anyone fired over politics.

          I respect this a lot and I wish more were like this.

      • JulieK says:

        not making websites dedicated to getting racists fired, or so forth.

        A particular employer firing an employee because the employer dislikes the employee’s politics is bad enough, but when third parties get involved and pressure the apolitical employer to fire the guy, it’s an order of magnitude worse.

        • Deiseach says:

          Just this minute saw some stupid self-congratulatory bullshit on Tumblr, a reblog of this tweet, and honestly – do they not realise the precedent this sets? Don’t they know anything about history? If they’re yowling about McCarthyism, don’t they recognise this is precisely what went on then: social disapproval marshalled against those deemed to be Enemies Of The State And People by association with Bad Ideas?

          If you’re going to clap back about “if you were spotted in the mob/and you lost your fucking job”, then you have no leg to stand on when it comes to “if you posted photos of yourself flipping off the president/and you lost your job”. Particularly when it’s coming from a self-declared socialist (see McCarthyism as referenced above – “the practice of making accusations of subversion or treason without proper regard for evidence”) who also has a tweet up inviting people to come along to an introductory Socialist meeting for an Introduction to Marxism. (This is the part where I go “How fucking stupid are you?”)

          Historically ignorant ignoramuses. I don’t know about teaching critical thinking in schools, but the skill of recognising something when you trip over it and fall flat on your face should be damn well inculcated into them.

          • herbert herberson says:

            Or, at least, if they don’t know their history–and I maintain that anti-left McCarthyism will be deployed when it is useful with absolutely no regard to whatever line various socialists/leftists took towards anti-right McCarthyism–then they should try harder to not forget who their enemies are.

          • Winter Shaker says:

            “if you were spotted in the mob/and you lost your fucking job”

            In neither case does it scan, but part of my brain is itching to continue with “…that’s amore”, or “…Yours is the Earth and everything that’s in it…”

          • Brad says:

            If you’re going to clap back about “if you were spotted in the mob/and you lost your fucking job”, then you have no leg to stand on when it comes to “if you posted photos of yourself flipping off the president/and you lost your job”.

            Tell that to the intellectual heirs of the actual McCarthyites — who by the way they are currently in the process of trying to rehabilitate. Because they are exactly the people that have no leg to stand on *right now*.

            There has never been a time in American history when the right was in ascendance and didn’t do this sort of thing to the left. So they have exactly zero credibility when they cry crocodile tears about the left defecting. There was never anything to defect from.

            If they want to establish that norm let them honor it unilaterally for a generation or two. Then they’ll at least have some sort of moral high ground to preach from. But the current crop of alt right “free speech norm” advocates hasn’t put in that time.

          • johansenindustries says:

            @Brad

            We’re a couple of generations after McCarthy. What groups did they try to get fired in the eighties?

            Plus, I think Deisach might actually have been aiming her words at the ‘you’re all Russian spies in cahoots with the evil empire’ people.

          • Brad says:

            @johansenindustries

            What groups did they try to get fired in the eighties?

            Gays, unmarried cohabitators, miscegenists, union organizers, atheists, and pot smokers among others.

            There was never any kind of truce. It’s the Gen-x version of baby boomers remembering the 50s as one long Leave it to Beaver episode.

          • Deiseach says:

            There has never been a time in American history when the right was in ascendance and didn’t do this sort of thing to the left.

            Brad, this is precisely my point. It is stupid, crappy, unhelpful and plain wrong behaviour, whoever indulges in it.

            When a self-proclaimed Marxist is the idiot thinking a re-run of McCarthyism is the way to deal with “political opposition” and moreover it cannot possibly have any downsides for him and his, that is when I go “I cannot believe I have to read this with my own two lyin’ eyes”.

            I’m not making any moral judgements here, I’m questioning how someone this dumb can even tie his shoelaces (let alone twitter).

          • Brad says:

            @Deiseach
            What herbert herberson is saying and you don’t seem to get is that this cannot possibly have any downsides for him and his because whether Marxists celebrate Nazis getting fired or condemn it harshly doesn’t make one bit of difference. If and when the shoe is on the other foot it will stomp on Marxists regardless of what the Marxists choose to do in 2017.

            There was never a truce, there isn’t a truce today, and a unilateral truce declared by the left now wouldn’t stick.

            If the Rationalists want to promulgate a brand new norm and get both sides to adopt it, first off good luck. And second off I’d like to see some reasonably detailed proposal. Because it isn’t at all clear to me exactly where the lines are supposed to be drawn.

          • John Schilling says:

            Gays, unmarried cohabitators, miscegenists, union organizers, atheists, and pot smokers among others.

            I don’t recall anyone trying to get miscegenists fired in the 1980s. Presumably there were people deep in the Bible Belt doing that sort of thing under the radar, but if they had to keep it secret then I don’t think it’s a fair comparison to people publicly shaming and firing Nazis (or whomever).

            Likewise unmarried cohabitors, and I think the anti-atheist stuff was mostly limited to political officeholders in practice. Being a union organizer is clearly job-relevant, and the War on Drugs had then and has now strong bipartisan support even if the Left now wants to put marijuana on the “good drug” side with alcohol (but tobacco on the “evil drug” side as soon as they can arrange it).

            Which leaves gays as your strongest example. And really, that’s a pretty strong example. Next time this comes up, I’d recommend just that one word as your response; the rest only weaken it.

          • Matt M says:

            When a self-proclaimed Marxist is the idiot thinking a re-run of McCarthyism is the way to deal with “political opposition” and moreover it cannot possibly have any downsides for him and his, that is when I go “I cannot believe I have to read this with my own two lyin’ eyes”.

            They don’t see it as an equivalence though. They aren’t thinking in terms of “Oh boy we have power now so now WE can use McCarthyism to get the right!” Because McCarthyism is defined as “that evil thing right wing people did to oppress left wing people for no good reason.”

            And they aren’t planning on doing that. They aren’t right wingers. They aren’t oppressing anybody. They are heroically protecting minorities from certain Nazis who would otherwise be killing them. They are defending the underprivileged from violent extremism. And they don’t understand how you could be so dumb as to not appreciate how completely and totally different that is!

          • MugaSofer says:

            Enraging, yet catchy.

            If you’re going to clap back about “if you were spotted in the mob/and you lost your fucking job”, then you have no leg to stand on when it comes to “if you posted photos of yourself flipping off the president/and you lost your job”.

            Well, on the meta level they have no leg to stand on. On the object level, they can quite accurately say “being a Nazi is bad and flipping off Trump is good”.

          • Toby Bartels says:

            It doesn’t matter if there was ever a truce or not. The situation is not symmetric, because a self-proclaimed socialist who wants to teach people about Marxism should understand and practice workers’ solidarity.

            If you side with the boss, you’re a scab! 👏👏
            If you side with the boss, you’re a scab! 👏👏
            If you side with the boss, then you well deserve your loss.
            If you side with the boss, you’re a scab! ✊✊

          • Bugmaster says:

            Ah, but you see Deiseach, the situation is completely different. McCarthy was a bad man who persecuted some good people. But we are good people who are ridding the world of evil, and because good always triumphs in the end, there’s no way this whole deal will ever turn around on us ! It’s a foolproof plan !

          • Deiseach says:

            On the object level, they can quite accurately say “being a Nazi is bad and flipping off Trump is good”.

            I hate that I have to say that yes, I agree, being a Nazi is bad (not because I think being a Nazi is good, but that I need to prove my right-thinkingness before making this argument since I’m not automatically enthusiastically on the side of “fire the Nazis! drive them out by doxxing!”). My point here is that:

            (1) certain parties have seized very quickly on “Nazi” as the be-all and end-all; whomever their political/philosophical opponents may be, as far as they’re concerned, they’re Nazis. I am concerned how swiftly a certain current has moved from “right-wing conservative” to “alt-right” to “neo-Nazi/white supremacist” to “Literal Actual Nazi”, particularly since for some persons my opposition to pet causes of their own makes me Literal Actual Nazi (e.g. I may not want to repeal the 8th Amendment to the Consitution in my own country, this does not mean I want to round up Jews/gays/Gypsies and send them to concentration camps, but as far as some people who believe there are such things as abortion rights are concerned, this makes me Literal Actual Nazi no different from Hitler)

            (2) I’ve had a very minor disagreement with someone online over this (luckily, we’re not fighting over this and we have enough mutual interests outside of “Tumblr political posts” not to be at one another’s throats) and I’d like to say that, if we’re going on “But Nazis are bad and Nazis do want to hurt me and people I know”, then firstly I agree about the badness! No quibble there!

            But Communism also is an ideology that has blood on its hands. If we know that, despite what “white nationalists” say today, when the Fascists were in power in several nations we have the evidence of what their regimes were really, then equally whatever “socialists” may say today about being all about economic justice etc. then we have the evidence of what Socialist/Communist regimes in several nations were really like when in power, and it’s not pretty. Back at the time of McCarthyism, Communism was a real threat (Stalin was in power). Some of the accused were real Communists who did want to overthrow the American state as constituted and change things and yes, there were “enemies of the people” who were on the list for “first up again the wall when the revolution comes”. People were genuinely worried and afraid. There was a Real Red Menace back then just as there is a Real White Supremacist/Neo-Nazi Menace today. And equally there were all kinds of left-wing groups and movements from old-style Labour activism to Socialism who got lumped in as “Communists”, when they disagreed with Communism, didn’t want to overthrow American government in a violent revolution, were pacifists and conservative on other social matters, and so on.

            Just as there is little discrimination about “I’m not a Nazi, I’m a conservative” “yeah right all you alt-right types are the same no matter what pretty labels you give yourselves”, then the same lack of discrimination about “I’m not a Communist, I’m a socialist!” applies, and I’m sure someone who describes themself as a Marxist would be pretty insistent they’re not a Maoist or whatever. So for someone self-describing as a socialist and inviting people along to Introduction to Marxism meetings, they maybe should be a little more cautious in their enthusiasm for public displays of sniffing out witches.

            And all that being said, I don’t want people like BostonDA (who tweeted that original dumb post) hounded out of their job for being a Known Socialist and Admitted Marxist and we know what Socialism/Communism leads to, which is totalitarian murder-states (because we’ve learned the hard way what driving Designated Villains out of their jobs and private lives leads to). Which is why I don’t want witch hunts about “we know this guy is a Nazi and his employer should fire him” (yes, even if he is an Actual Literal Nazi), unless he’s going out beating up Jews in the streets or burning down synagogues. McCarthyism was wrong even in the face of a real threat, and the version being promulgated just as enthusiastically today when it’s Fascist hunting not Communist hunting, is still wrong.

          • Viliam says:

            They aren’t thinking in terms of “Oh boy we have power now so now WE can use McCarthyism to get the right!”

            Well, some people are proudly saying “No bad tactics, only wrong targets”, which is quite close, only more general, IMHO.

        • slightlylesshairyape says:

          What about customers? On the one hand, I don’t particularly want to go to a sandwich shop that employs a white supremacist. On the other hand, can an apolitical employer really be a conduit for their customer’s discrimination?

          • I don’t particularly want to go to a sandwich shop that employs a white supremacist.

            ??? Really? Why is that? Why do you care at all? How do you have any idea at all? I might have patronized stores that employ homicidal maniacs, and I don’t care at all, as long they aren’t doing it at the time I’m there. Yes, I’m against their maniacally killing people, but I’m not against them selling me stuff.

        • just as there is a Real White Supremacist/Neo-Nazi Menace today

          So far as the U.S. is concerned, I don’t think so. There is a real populist menace. White Supremacists and Neo-Nazis seem, judged by the number that show up for confrontations, to be a tiny minority that gets attention wildly out of proportion to its numbers.

          I can’t speak to the situation in other countries.

      • Brad says:

        More like, “if more people on the left weren’t eager to tear down the 1st amendment”. It’s not even about being against racists fired, so much as not making websites dedicated to getting racists fired, or so forth.

        Unless the people being fired are public employees, that has nothing to do with the first amendment.

        I don’t see how Scott can blame people from harping on it only applies against the government when even proponents can’t seem to consistently remember that they are talking about an abstract social norm that has nothing at all to do with the First Amendment.

        • ThirteenthLetter says:

          You quite certainly are aware that the spirit of the first amendment is what’s being discussed here, not the letter of it. Why did you even bring up this pointless distraction?

          • Brad says:

            Any spirit of the First Amendment would have to do with state action that doesn’t contravene the letter of law but nonetheless chilled speech. Private reactions to speech have nothing to do with the First Amendment — letter or spirit.

            If you think otherwise, I wonder what the analogous spirit of the First Amendment as it applies to private conduct is regarding the Establishment Clause. Do Churches violate this spirit by their very existence? And what about the spirit of the sixth amendment? Does it require a private litigant to pay for its opponent’s lawyer if he can’t afford one? Does the spirit of the 4th amendment require wives to mirandize their husbands before inquiring what they were doing out so late? Or mothers before they ask their children if they took cookies from the cookie jar?

          • Nornagest says:

            If you think otherwise, I wonder what the analogous spirit of the First Amendment as it applies to private conduct is regarding the Establishment Clause.

            The obvious answer seems to be “don’t fire someone for belonging (or not belonging) to a church or for exercising their religion”. Which happens to coincide with one of the protected categories of the Civil Rights Act of 1964, but even if it didn’t I think you’d agree that it’d be a dick move.

          • Brad says:

            That would be analogous to the free exercise clause, not the establishment clause.

          • ThirteenthLetter says:

            Again, what does all this flyspecking accomplish? You know exactly what the original commenter was getting at.

    • Matt M says:

      Can I just say I think it’s unfair to characterize the lady flipping-off the motorcade as “getting fired for politics.”

      I’d say she’s getting fired for being very rude in a very public manner. People who get recorded screaming at Uber drivers get fired, and it ain’t because your random Uber driver is a powerful politician. It’s because being rude in a public way brings discredit upon yourself, and therefore indirectly, upon your employer.

      I suspect that if this lady had flipped off President Hillary, she’d get fired just the same.

      “Fired for politics” is James Damore. When you go out of your way to be respectful, and are told that your ideas are simply not allowed to be discussed. This isn’t that.

      • Harry Maurice Johnston says:

        An interesting perspective. Do you really think what she did was rude enough to be fired over? Seems pretty trivial to me, nowhere near the level of rudeness you compare it to.

        • Matt M says:

          It’s probably a mixture of the rudeness level AND the fame/notoriety/popularity of the target that determines whether you get fired or not.

          Screaming at an uber driver gets you fired because its REALLY rude, even though the victim is powerless.

          Flipping off a car Trump is in only minorly rude, but the victim is hugely powerful and has a legion of dedicated fans willing to agitate for you to be fired.

          So to extend the metaphor, flipping off an uber driver probably doesn’t get you fired, but maybe flipping off Oprah does? Especially if you do it in a public location and someone catches you doing it on camera.

          • John Schilling says:

            It’s probably a mixture of the rudeness level AND the fame/notoriety/popularity of the target that determines whether you get fired or not.

            And you don’t have a problem with that? Because I do. Lèse-majesté is a bad, stupid, wrong thing even if it is privately enforced and even if the majesty being lèsed is that of celebrity rather than monarchy.

          • Matt M says:

            I mean sure, in a perfect world, the punishment for some sort of offense would depend entirely upon the severity of the offense itself and not at all on the status of the victim.

            But come on, that ain’t happening anywhere anytime soon.

          • pdbarnlsey says:

            “Screaming at an uber driver gets you fired because its REALLY rude, even though the victim is powerless

            I think not treating those less powerful than you badly is an important and fairly widespread norm. Many of the accusations against, say, John Kerry, Hillary Clinton (troopergate), Sarah Palin, Donald Trump etc boil down to “lousy tipper/yelled at the staff”.

            Flipping off someone who you don’t have any power over feels really different to me. And also really, really mild.

          • Matt M says:

            Many of the accusations against, say, John Kerry, Hillary Clinton (troopergate), Sarah Palin, Donald Trump etc boil down to “lousy tipper/yelled at the staff”.

            Rumors and hearsay are one thing. Live video evidence is another.

  45. ymandel says:

    Yeshu is the numerical equivalent of drek, which is obviously Yiddish for sausage.

  46. INH5 says:

    After years of self-driving cars being five years away, there’s now a date for a self-driving car being available to ordinary people for a commercial purpose: next year. It’s pretty minimal – some cooperating ride-sharing passengers in Phoenix will get self-driving cars without human backup drivers – but it’s happening.

    From other articles that I’ve read, the popular reporting of this as “Google is taking the human backup driver out of self-driving cars” seems to be exaggerated. It would be more accurate to say that Google is moving the human backup driver to the backseat:

    Waymo, the self-driving car company created by Google, is pulling the human backup driver from behind the steering wheel and will test vehicles on public roads with only an employee in the back seat.
    […]
    Waymo has long stated its intent to skip driver-assist systems and go directly to fully autonomous driving. It said the Waymo employee in the back seat won’t be able to steer the minivan but, like all passengers, will be able to press a button to bring the van safely to a stop if necessary.

    While this does demonstrate more confidence in the safety of their systems (though I’d really like to know what roads, specifically, these cars will be driving on), it seems more like a publicity stunt than a practical step forward to me. If you have to pay someone to sit in a self-driving car and babysit it the entire time that it’s on the road, then what is the practical benefit, exactly, to putting them in the backseat instead of the driver’s seat?

    • ManyCookies says:

      They’re still shifting responsibility from the backup driver to the car, it shows confidence that the vehicle can ‘safely’ fail and doesn’t need someone to take the wheel at the first sign of trouble.

  47. benwave says:

    Not related to the links, but Net Neutrality is a big story right now. What do SSC readers think of this? Almost all of the commentary I see on this is against ending net neutrality, but I would guess I could find some dissenting voice here?

    • Bugmaster says:

      Personally, I don’t think it matters, because media consolidation can’t be stopped. The market forces that favor it are too strong.

    • shenanigans24 says:

      That would take us back to the dark ages of before 2015.

    • Urstoff says:

      Seems to me like it would be better to attack what enables telecom companies to become local monopolies than to try to ensure that those monopolies behave in some certain sort of way.

      • ManyCookies says:

        I thought telecom markets tended to form natural monopolies, in the same way home electricity markets did. Am I off base?

        • m.alex.matt says:

          No, they started that way. When they broke up The Phone Company into all the Baby Bells that formed a bunch of regional monopolies that have spent the last couple decades working hard to consolidate again off of that regional monopoly base.

    • toastengineer says:

      What I’ve heard from anyone other than the “this is good because everyone says it’s good” crowd has all been pretty negative. But it’s not like I’ve actually sat down and read the law.

      The idea seems to be that companies like Netflix and Amazon were asked to pay more for their Internet connection than normal people do considering they put way more strain on the system than anyone else does, Netflix traffic being the large majority of Internet traffic during the evenings. These companies then went crying to the government to make a law forcing ISPs to charge everyone the same amount, which the government successfully sold as “net neutrality” despite being pretty much the opposite of what the principle of network neutrality is about. So now everyone has to pick up part of Netflix’s bill.

      • orangecat says:

        These companies then went crying to the government to make a law forcing ISPs to charge everyone the same amount, which the government successfully sold as “net neutrality” despite being pretty much the opposite of what the principle of network neutrality is about. So now everyone has to pick up part of Netflix’s bill.

        Netflix is already paying for the network traffic that it generates, and net neutrality doesn’t prevent ISPs from charging heavy users more. It only prevents ISPs from treating traffic from different sites differently, so that they can’t intentionally block or throttle Netflix while letting their partner video site through.

        If we had effective competition, this wouldn’t be necessary because any ISP that deliberately crippled its service like this would lose customers. But we don’t, so it probably is.

        • Jaskologist says:

          T-Mobile is currently advertising free Netflix streaming on their data plans. Isn’t that treating traffic from different sites differently?

          • orangecat says:

            Yes, it’s called “zero-rating” and IMO is as much a violation of net neutrality as blocking traffic from sites unless they pay a ransom. See https://www.eff.org/deeplinks/2016/02/zero-rating-what-it-is-why-you-should-care

          • Standing in the Shadows says:

            Wireless carriers are in an interesting situation, and what T-Mobile is doing as contrasted with what AT&T and Sprint have tried to do is illuminating.

            When you are a wireless carrier, and you *know* that something is an AV stream with very well known characteristics (coming from the same source, known packet size, reasonably constant rate and constant interpacket gap, robust against packet loss, running for a longish length of time) then you can reconfigure some parameters of the connection between the terminal and the local tower and get some significant cost and efficiency gains, and you can also handle them somewhat specially in your wire infrastructure.

            So, what T-Mobile has done is published a general “if you are a streaming service and want to give a nice feature to your customers who are on T-Mobile, then do the following: go halvies with us on buying a sufficient number of one of the following brands of switch interconnect modules, for us to install in some of our big interconnect switches. Present your streams at those points. Tag your streams like so. We do not care what you are charging you customer for your service. If you do that, THEN we will not charge the megabytes of your streams against our customers monthly transfer limits. Oh, and you are allowed to use our branding and logos when you advertise this feature to your customers, and we get to use your logo when we tell our customers about it”. Notice that T-Mobile is not trying to dip into the streaming source’s revenue, nor are they charging their own subscribers extra for this. And this offer is open to any and all comers, just just “market strategic” ones, and T-Mobile will do this for you and for your competition, and you don’t get to demand that it not be offered to your competition on the same terms.

            I know people who have switched from other carriers to T-Mobile JUST FOR THIS, being able to listen to music streams without running out of transfer it in just a few days.

            Now, let’s look at what AT&T and Sprint have tried to “offer”, against fierce pushback. You are a streaming service. You pay ALL of the cost of the interconnect module, AND you pay a bit rate cost for the data you pump across it, AND you do “revenue sharing” with us for what you are charging your customer, AND we may intercept and retranscode your stream to lower quality AND we may still count the data against our customer’s monthly cap AND we may start charging our customers extra to access your service, OR ELSE we might accidentally de-prioritize your traffic or start cutting off our customers from accessing your service.

            See the difference?

            (I’m a T-Mobile subscriber).

          • Matt M says:

            Yep. It’s a lovely result. The doomsday scenario outlined by NN proponents is that carriers will charge you extra to watch Netflix.

            The actual result observed in the marketplace is that carriers are trying to give you Netflix for free to convince you to switch to them.

            Of course, the NN proponents aren’t celebrating this. They want the government to force T-Mobile to stop this “anti-competitive” behavior. Because upholding the “principles of net neutrality” is actually more important than consumers getting a good deal.

          • moonfirestorm says:

            Is the mobile situation a big driver of the concerns about NN though?

            Typically I hear about it in the context of home Internet, which has a single provider in a lot of places. So there’s no point in incentivizing customers, because they don’t have a choice anyway.

            I’m pretty convinced the monopoly is the real problem here, and NN wouldn’t matter if there was any sort of competition. Of course that’s not the issue at hand, we have to decide about net neutrality in the markets we have.

          • Jaskologist says:

            Shouldn’t mobile be a big consideration? That’s an internet provider the same as the physical lines to your house. It’s more limited than broadband, but a lot less limited than the dialup we had to use back in the day. Saying that ISPs have a monopoly if you ignore mobile is saying they have a monopoly if you ignore the competitors.

          • Standing in the Shadows says:

            It is also worth nothing that one of the significant reasons why Sprint wanted to buy T-Mobile, and why AT&T did not complain about it, was to try to get T-Mobile to stop doing this.

          • moonfirestorm says:

            Shouldn’t mobile be a big consideration? That’s an internet provider the same as the physical lines to your house. It’s more limited than broadband, but a lot less limited than the dialup we had to use back in the day. Saying that ISPs have a monopoly if you ignore mobile is saying they have a monopoly if you ignore the competitors.

            That’s a fair point, as is John Schilling’s mention of satellite internet either here or in another thread.

            If you end up having to rely on a mobile-level connection because all your broadband providers are doing crazy things though, I think I’d consider that a significant negative consequence. I’ve seen what playing WoW through a phone hotspot looks like. It’s not pretty.

            Maybe the threat of “oh well we’ll just cut the cord and do all our browsing on our phones” is real enough to keep the local broadband supplier honest, though.

          • Standing in the Shadows says:

            There are specific technical aspects for wireless mobile IP service that make tagging and discounting AV streams a much much much bigger win than are possible for a fixed line IP service.

          • Standing in the Shadows says:

            If NN falls, then suddenly there is going to be a demand driven market for wildcat, pirate, black market, and co-op WISPs and neighborhood and condo ISPs. The incumbents are going to double down and double down again on getting local munis to ban them, and the voter opposition to those bans is going to rise. It’s going to be chaos.

            Huh. Black market internet service… Maybe I want NN to fail after all…

      • Standing in the Shadows says:

        Amazon and Netflix already pay for their own internet connections.

        You think that Amazon gets to force multiple carriers to connect multiple redundant connection to all their many MANY data centers… for free? Or at a government regulated tariff? Or at the same rate that you do for your tiny little Comcast DOCSIS connection?

        Doing so for something the size of Netflix or of Amazon is beyond my paygrade, but in jobs past I have signed off on and then done the recurring audit for the multicarrier internet connection bills for a “small” datacenter, “only” a few dozens of dozens of full racks.

        You pay for the switches, and hardware interconnects, and the transceivers, and exterior cable and glass pulls to your location, and the power bill to run it all. You pay a monthly bill, to each carrier. And you do in fact also pay by the bit. And the per bit rate goes up during peak and surges, in both 6 second and in 60 minute buckets.

  48. Matt M says:

    Good news for those of you who majored in “good calibration, wide knowledge and ability to work out what’s important”.

    Aww yeah where my fellow management consultants at!

  49. Well... says:

    A thought on the trans hair dye thing.

    In previous OTs I argued that seeing one’s body as a canvas (a common theme you hear from people with tattoos, lots of piercings, etc.) might be a type of disorder (maybe similar to facial blindness?) in which people cannot see that their bodies are in fact “finished works”. If I’m right, then those people have some kind of dysmorphia or dysphoria.

    As I understand it, transsexualism is often caused by gender dysphoria. So…maybe dysphorias fly together?

  50. ArsGenetica says:

    One reason to be averse to various genetic testing startups is that under currently prevailing business models the company keeps the data and reserves the right to sell it to anyone – so you are potentially setting up yourself, your progeny and other relatives to be targeted or discriminated against based on criteria which one may not be even able to conceptualize today.

  51. Urstoff says:

    What exactly is patented with respect to insulin? The production process? The particular chemical variant of insulin that a company makes? Why isn’t there a low-cost ‘generic’ insulin?

    • US says:

      I know next to nothing about patenting so this might not be the right place to add this particular comment, but I know a lot of other stuff about diabetes, insulin and related stuff. Some problems/observations, just off the top of my head (I don’t have much time to comment and won’t be able to follow up for the next couple of days, so I was considering only adding a comment in the next open thread – however I might as well start here):

      i. There are lots and lots of different formulations, with different effects. You have multiple different types, which differ in terms of how fast and how long they act. Some action patterns are desirable in some patient settings but not in others; diabetics are a heterogenous group of patients, and different patients have different needs. In a classic type 1 intensive treatment regime you’re actually not only taking one type of insulin, but rather two; one long-acting agent that works between meals (basically all the time, but it’s important that you have insulin in your body also between meals, which is what this type secures), and another type that work for meals to handle glucose spikes associated with meals. A type 2 setting might be a setting where (at the beginning, at least) only a long-acting insulin is needed, because short-acting spikes are handled by oral antidiabetica. Some patients are on formulations with different fractions of the two types; e.g. a 70/30 formulation, where one part of the desired effect is short acting and the other is long-acting, but it’s packed into the same pen to minimize the number of injections (which may increase compliance). Then you have pump treatment, where only fast-acting insulin is used but is continuously being released into the body, to avoid the need for taking separate long-acting insulins; this has a number of therapeutic benefits, but also some associated risks. There are different types of fast-acting and slow-acting insulins, which have different action patterns. I assume some aspect of the chemical formulations that secures these different action patterns is what is patented, but I don’t really know. For example some slow-acting insulins have been developed that make their effect less ‘spiky’ and more reliable, to avoid things like nocturnal hypoglycemia (low blood sugar during the night – which can be fatal). Commercially available fast-acting drugs are reasonably similar now, but used to be somewhat more different from one another; I’m old enough to remember a time where the ‘fast-acting’ insulins where not nearly as fast-acting as they are now. The action pattern’s also here important e.g. because if a fast-acting insulin is working too slowly, the effect you get from one meal dose might not adequately address the need at the time, and may ‘carry over’ to work when it’s not intended to – both these things are undesirable. Insulins that work too fast may also have issues.

      ii. The therapeutic index for insulin is low. You don’t need to get the dose very wrong to have potentially fatal side effects. This makes the precise effects of the specific drug you’re taking very important. Again, not all patients will need the same type, and the type of insulin treatment you need will often change over time in the type 2 setting, which is what constitutes the great majority of diabetes patients. It will also change in the type 1 setting, but that’s mostly in the first few years of disease.

      iii. Keeping diabetics alive in the short run with insulin is a lot easier than keeping them alive and relatively healthy. There’s a very strong relationship between glycemic control and complication risk, and complications make up a much larger proportion of the diabetes costs than do the cost of drugs – I can’t recall the exact figures, but 10-15% of total diabetes costs being insulin related is probably not far off. Many type 1 patients who were treated with insulin in the 1920s survived a decade or two and then died from kidney failure. If I’d been born then, I probably would not have been alive to write this comment; I have had type 1 diabetes for roughly 3 decades and got type 1 when I was 2. A lot of diabetes treatment is focused on avoiding the development of complications, and you need more than one ‘generic insulin, isle 3’ type product to accomplish this goal. In theory the closest thing we have that might be the ‘generic type insulin’ you’re looking for that could actually address all patient problems is pump treatment, but that one is combined with a highly sophisticated delivery system which is quite expensive. It also happens to be a treatment modality which is well beyond what many patients will be able to handle, as they are not able to follow the regimen; it requires much, much stricter compliance than that displayed by many type 2 patients in the community setting in particular, and non-compliance leads in that setting to rapid and potentially fatal complications.

      iv. Even non-fatal insulin-related complications can be very important from a quality of life point of view. Many years ago I switched from an old type of slow-acting insulin to another which was by then a very new drug just developed to minimize the risk of nocturnal hypoglycemia in my teenage years. After I made that switch I did not experience a single case of nocturnal hypoglycemia for more than five years. Before the switch, I’d had monthly hospitalizations (occasionally weekly, when things were not going well; some parts of my teenage years were not much fun in that respect). No, I didn’t die in either setting, both drugs kept me alive, but from my perspective those two drugs were not equivalent. I also know now that the difference between pump treatment and standard intensive treatment is likewise huge, though the changes I experienced when making that transition were smaller as I was much better regulated when I started (specific life phases may also change insulin needs, it’s not just timing from diagnosis or disease duration that matters; teenagers are often very difficult to treat satisfactorily, for multiple reasons including hormonal changes that make the effects of insulin harder to predict).

      v. From a cost perspective, insulin can not be viewed in isolation. If you’re saving $X/patient/year on insulin but patients are experiencing X more hospitalizations or X more cases of diabetic nephropathy or X more cases of foot ulcer development, you’ve not saved any money. Well, maybe you have, but someone else has to pay for that. Lack of care integration and related cost-switching tends to increase costs. The proportion of the cost paid out of pocket is of course highly relevant to patients, but if you’re interested in the total cost of diabetes (which you should be, whether or not you have the disease – it’s a really expensive condition that is taking up a larger and larger share of the health care budget over time) it’s far from the only variable of interest. It’s perhaps worth noting that although type 1 patients die if they don’t take any insulin, if they take too little they can actually muddle through for some time – years – before they croak, which means that compliance is a variable of interest in the optimal pricing setting. This is another way of saying that although the price elasticity of demand for insulin is not high, it is also not zero (one of the econometricians at my university incidentally showed this in his thesis, using Danish pharmacy data). Aside from diabetes-related stuff, high insulin prices may also lead to non-diabetes related health risks of patients, such as increased risk for infectious diseases transmitted through blood in a pen/needle sharing context.

      I’ll probably add a lot more later on next week, if people are interested to know more. The pricing of insulin is not something I’d read a lot about, but I know a lot of stuff about related things and might be able to add some details people are unaware of.

    • Douglas Knight says:

      These drugs are off patent, but there probably aren’t generic versions available, because they are hard to make. They are proteins, made by genetically engineered bacteria. The typical generic company that makes small-molecule drugs isn’t trying to mimic them. I think that the FDA does now have a process for qualifying generic engineered proteins, but it took years to develop because it’s a lot harder to check the purity of a protein than of a small molecule. (The situation for monoclonal antibodies is much worse, where it’s hard to pin down what the drug even is.) Also, as US implies, you should worry about the purity of your insulin more than you worry about the purity of other drugs.

      The article is about modified versions of insulin that are very close to regular insulin, but slightly better. (In particular, still fast-acting.) One entered the market in 1996 and the other in 2000. They have always had exactly the same price, rising ~10% per year for the 20 years since 1996, now up 10x. It would be interesting to know what the price of regular insulin has been doing, since that is a pretty direct competitor. Insulin has been around for a century, and it wasn’t made by genetic engineering back then, but it is still pretty hard to make, so maybe there aren’t many companies that make it. The two drugs in the article are made by companies that already made regular insulin, so these drugs didn’t represent an expansion of competition.

      But that all is about why prices aren’t falling, not why they are rising fairly rapidly.

      • Ratte says:

        ‘Regular’ insulins like SPH are available for less than $25 at Wal-Mart, without a prescription. Poor patients also have access to extended supplies of the better stuff at negligible cost, as well.

        Also, as a diabetic, the price I’ve personally paid for insulin over the last 20 years is essentially unchanged; most diabetics just have a co-pay with the bulk of the rapidly increasing prices being passed on to the insurance company. As far as I can tell, the price increases are simply a scam to extract as much as possible from health insurers.

        • Brad says:

          As far as I can tell, the price increases are simply a scam to extract as much as possible from health insurers.

          It isn’t like it is coming out of health insurance company profits. They are scamming all of us.

      • US says:

        “But that all is about why prices aren’t falling, not why they are rising fairly rapidly”

        Even so, it’s a good comment, from which I learned something new (thanks!).

        I hope to be able to find the time to post another comment on these topics tomorrow.

      • US says:

        Other products besides insulin may also be relevant for the price development of insulin-related products, right? Several new oral antidiabetic drugs have been developed over the last decade or two, and the prices of these drugs are high compared to the older alternatives. This may matter in terms of how companies set the prices of their insulin-related drugs. A quote:

        “From 1999 to 2011, national data suggest that growth in the use and price of prescription medications in the general population is 2.6% and 3.6% per year, respectively; and the growth has decelerated in recent years (22). Our analysis suggests that the growth rates in the use and prices of prescription medications for diabetes patients are considerably higher. The higher rate of growth is likely, in part, due to the growing emphasis on achieving glycemic targets, the use of newer medications, and the use of multidrug treatment strategies in modern diabetes care practice (23,24). In addition, the growth of medication spending is fueled by the rising prices per drug, particularly the drugs that are newly introduced in the market. For example, the prices for newer drug classes such as glitazones, dipeptidyl peptidase-4 inhibitors, and incretins have been 8 to 10 times those of sulfonylureas and 5 to 7 times those of metformin (9).”

        “Between 1987 and 2011, medical spending increased both in persons with and in persons without diabetes; and the increase was substantially greater among persons with diabetes. As a result, the medical spending associated with diabetes nearly doubled. The growth was primarily driven by the spending in prescription medications.”

        More widespread use of oral antidiabetic drugs should in theory be expected to affect both the timing of initiation of insulin therapy in type 2 patients and the dosage of insulin in those patients (these patients should be expected to need less insulin, and they should be expected to be able to wait longer until they eventually start on insulin, for a given glycemic goal, due to a delay of treatment failure). But also related, if the price difference between ‘insulin-related products’ and oral antidiabetic drugs is smaller, all else equal that should make the use of the oral drugs more attractive to patients and care providers. I’m not sure how this stuff works, but I’d be surprised if these companies are not considering such aspects as well when making their pricing decisions.

        • Douglas Knight says:

          This article, linked from the original, claims that regular insulin can’t be used in an insulin pump, only these super-fast analogues. Is that true? That seems weird to me, but if it’s true, then they really are a lot better than regular insulin.

          I had been wondering if the drug was cheap to get people to try it and was slowly coming up in price as they convince people it’s better than the alternative. I haven’t thought through the economics, but it seems more intuitively plausible in the case of big up-front cost of the pump, than just an ongoing cost.

          ━━━━━━━━━

          No, I don’t see how the price of glitazones should affect the price of insulin. The insurance company will do a cold-eyed cost-benefit calculation. The patient might think in terms of a diabetes budget, including glitazones, but not hospitalization. But is the patient making the purchase decision? Most decisions are made by doctors. At some point we should step back and ask how other drugs are priced. This is pretty opaque.

          Even if these analogues are not really competing against generic insulin, there are other drugs rising prices against quasi-generic competition, such as Epi-pen. How did they do it? I don’t really know. I think a lot of it comes down to the pharmacy not having the option to substitute generics and the insurance company unable to communicate to the doctors how to write the right prescription. They could just refuse to pay for the brand name, but if the doctor doesn’t remember, that’s a hassle for the patient, who has to go back and get another prescription.

          • US says:

            “regular insulin can’t be used in an insulin pump, only these super-fast analogues. Is that true? That seems weird to me, but if it’s true, then they really are a lot better than regular insulin.”

            There are a few ways to approach this question, so I’ll split up my answer:
            a) I don’t think there are any strong technical barriers to patients filling their insulin pumps with other types of insulin. That is, in theory you probably could do that.
            b) But this would be a really weird thing to do, as all the therapeutic benefits of pump treatment would go right out the window and you would no longer be able to use the sophisticated software developed for the pump.

            Some major benefits of pump treatment includes: i. Faster and more efficient drug absorption (in a standard setting, new pump users will reduce their insulin requirements by ~20%, and the insulin is also working a bit faster when infused). ii. The ability to regulate insulin levels to metabolic demands continuously throughout the day (and night).

            In a standard pump setting, if you’re about to go for a run (this will increase insulin sensitivity acutely and reduce insulin requirements) you can adjust your pump settings so that less insulin will be infused for the duration of the exercise (…if you’re a bit more sophisticated, you could also adjust the dosage after the exercise, as insulin sensitivity might be increased post-exercise for a significant amount of time – as in, hours or even days). This means that you’re not required to compensate for the increased insulin absorption by the intake of carbohydrates as in the ‘standard diabetes setting’, or at least not nearly to the same extent as you would need to if you did not use an insulin pump. Any diabetic who also take long-acting insulin will not have this option; they can adjust their meal insulin bolus pre-exercise, if they take those, but they can’t shut off the long-acting insulins already in their system because those molecules were injected many hours ago. If you use long-acting insulin in a pump, you likewise won’t be able to ‘shut off’ (or reduce significantly) your insulin. (You might think: But if you’re a diabetic not on pump treatment who is planning on going for a run, why not just adjust the long-acting insulin the day before? You can do that, yep, and then you’ll end up struggling with high blood glucose levels at other times during the day of your run, when you’re not exercising. Exercise adds a lot of complications to diabetes treatment). Consider exercise an example of a setting where the pump has useful properties; it’s an important one, but there are many other contexts where the increased ability to adjust dosage rapidly and accurately in the pump setting will be helpful to a diabetic.

            Insulin pumps have software that calculates how much active insulin you have in your body, based on half-life calculations of the insulin boluses that are introduced into your body throughout the day. This stuff’s very useful in a fast-acting insulin setting, because it makes it easy to derive benefits from e.g. taking blood glucose levels an hour or two after a meal to figure out if the dosage was sufficient/excessive. The value of such estimates would be near-worthless if you’re combining a pump with longer-acting agents, not only- but also because i) the estimates would be much less reliable because of increased measurement error (a lot of things people do throughout the day will in some way affect glucose metabolism and insulin sensitivity, so the longer the duration of action of the drug, the more uncertainty is introduced), and ii) they wouldn’t have much therapeutic value as e.g. correction of a high blood glucose would take a lot longer if the insulin your pump is loaded up with have a long-ish duration of action.

            I think the major takeaway from all this stuff is that although from a technical point of view it might in theory be possible to load up your pump with ‘non-pump insulin’ (i.e. slower acting agents than the ones currently being used), there really wouldn’t be any point of doing it. The pumps have been developed to be used with fast-acting insulin, and most of the benefits derived from pump treatment hinges crucially on that fact; and actually, if you change the insulin type, that treatment modality looks a lot more unattractive. Why would I walk around with a tube in my stomach and a big mobile-phone like thingy fixed to my body 24/7 when I could just take a few injections throughout the day and get roughly the same benefits?

            “No, I don’t see how the price of glitazones should affect the price of insulin. The insurance company will do a cold-eyed cost-benefit calculation. The patient might think in terms of a diabetes budget, including glitazones, but not hospitalization. But is the patient making the purchase decision? Most decisions are made by doctors.”

            I should perhaps interpose at the start in case you don’t know that I do not live in the US but in Denmark. There’s definitely some differences in terms of how purchase decisions are made across countries, with varying bargaining power and influence of the various agents involved. Some aspects that springs to mind that might be relevant:

            i. Out of pocket costs are higher in the US than in Europe. Don’t know what to make of this in terms of the discussion here right now, but it’s a relevant fact to keep in mind. Relatedly, the price differences between generics and drugs on patent are higher in the US than e.g. in Europe.

            ii. Type 1 and type 2 patients are likely to be very different. Type 1s tend to be highly motivated, compared to type 2s, and they’ll often be telling the doctor that they want to start out on this new drug they’ve been reading about. Many of them have been ‘their own doctors’ from the moment they got the diagnosis. Type 2s are in general much more passive, and here the role of the GP is much more significant.

            iii. A clarifying question, just to make sure you understood where I was going: You do follow that if the use of an oral antidiabetic drug affects the timing of- or the level of consumption of insulin treatment (including non-generic insulin produced by the same company producing the oral drug), these drugs can be considered imperfect substitutes in implicit competition with each other? I think this is the case almost regardless of what coverage decisions are made by insurers because of the comments included in the point below (I don’t disagree with the notion that insurers implicitly affect the level of competition).

            iv. Insurance providers do cost-benefit calculations, but the prices of the various drugs will always to some extent be a result of a bargaining process between insurance providers and producers (…and patients, but let’s leave them out of the picture for now – the fundamental effect is the same, we’re dealing with a change in the relative prices of the drugs), no? If producers have a strong incentive to increase the sales of non-insulin antidiabetica (long-term, ideally post-patent) and they already earn monopoly profits from the patent, adjusting the prices of at least peripherally competing products (which are off-patent and so no longer earn monopoly profits anyway) might make sense, and insurance providers would not in most contexts, I believe, have enough bargaining power to shut that sort of behaviour down completely. Rising insulin prices is what we’re trying to explain here and this is at least a mechanism.

            v. What insurance providers include in the cost-benefit calculations will vary a lot geographically because coverage profiles vary, for many reasons. It’s important to note that insulin is a small component in the diabetes cost function, as already mentioned, so to the extent that they are required to deal with all diabetes-related costs it might not be optimal to minimize insulin-related costs. On the other hand you could also go in a different direction and claim that rising insulin prices is a signal that insurance providers are not optimizing all that well (a ‘lower prices would increase compliance and lower the complication rate, which would save them money’-argument). One reason why they might not be optimizing all that well might be that it’s really difficult to optimize here because of the complexity involved (and the associated in most settings pervasively disappointing level of care integration). The US market is probably more complicated to handle than some European markets because of the competitive structure.
            If you don’t like an ‘it’s hard, even for private firms, to optimize’ argument, another way to look at it is that high insulin prices might be a key mechanism some private insurance providers employ to avoid adverse selection (…get rid of diabetics). That would be a reason for observing high insulin prices, not rising insulin prices, but on the other hand diabetes incidence and prevalence have been increasing rapidly over the last couple of decades, perhaps making the problem more severe, necessitating such measures.

            vi. Recent changes to Medicare/Medicaid prescription drug coverage and related stuff? It’s been a while since I read about this kind of thing, but it’s probably relevant.

            vii. “I think a lot of it comes down to the pharmacy not having the option to substitute generics and the insurance company unable to communicate to the doctors how to write the right prescription.”

            In Denmark I believe a GP can write on a prescription that it’s okay for the pharmacy to switch to the cheapest drug if multiple drugs of the same type is available – the pharmacist will then if confronted with a prescription of that type ask the patient if it’s okay that he gets drug X instead of drug Y, because it’s the same stuff, really, and the patients will often walk away happy to have saved a few bucks. I don’t know if it’s ‘a standard’ or not that prescriptions are written this way; I believe doctors ask patients before writing the prescription if this option is okay with them (I’ve been asked that question myself and used a prescription like that). As mentioned, there are a lots of cross-country differences of relevance.

          • Douglas Knight says:

            (iv)…If producers have a strong incentive to increase the sales of non-insulin antidiabetica (long-term, ideally post-patent) and they already earn monopoly profits from the patent, adjusting the prices of at least peripherally competing products (which are off-patent and so no longer earn monopoly profits anyway) might make sense

            First of all, Novo Nordisk makes very few oral antidiabetics.

            Do they make “monopoly profits” which happen to change every year? It’s theoretically possible that it changes every year because of the introduction of new drugs, but, nope. More plausible is that is the result of marketing increasing the market or exploring the space of prices. But what’s very strange is that the curve was unchanged by the introduction of competitors in 2001 and 2004.

            (vii) It depends on what is considered equivalent. The pharmacist can always make generic substitutions (unless specifically disallowed), but the doctor can write a broad prescription allowing more substitution (eg “epinephrine autoinjector” rather than “Epipen”). But only if they know to do so.

          • US says:

            Right, it’s probably a bad model (and I probably should have added those “monopoly” quotes myself). But I find the increasing prices weird and it seemed to me to at least in theory be a potential mechanism. While writing my long reply I was almost certainly much too focused on trying to explain a large ‘actual’ increase in insulin prices; but the truth of the matter is probably that the extent to which the insulin prices that decision makers pay for the drugs have actually changed is very hard to even evaluate.

            The idea that drug companies have been moving towards an ‘Ivy pricing model’ (i.e. have an insanely high ‘standard price’ (list price) and then use a lot of different rebate systems and so on to price discriminate) may mean that the ‘true price development’ may have been quite different from what you observe from the list price graph, as also mentioned in the article.

            This stuff’s really complicated – many decision-making entities with conflicting interests, highly dynamic and heterogeneous disease patterns, huge differences in how markets are structured and regulated, big differences in which decision problems the various decision-making entities involved may even be expected to optimize, etc., etc. – so it’s probably highly unlikely there are any simple explanations for any specific price developments – ‘real’ or not – observed.

            But this was an interesting exchange, thanks.

  52. Yosarian2 says:

    There already is (and has been for decades) one specific case where married men can become Catholic priests. If an Anglican minister or Eastern Orthodox priest is already married (as is allowed in both of those sects) and then converts to Catholicism, he is allowed to remain married and become a Catholic priest. It is estimated there are about 120 married Catholic priests in the US right now.

    http://www.latimes.com/nation/la-na-pope-married-priests-2017-story.html

    That’s a pretty limited exception but if the Pope wants to expand that exception, he has some precedent.

    • Anonymous says:

      Non-Roman Catholics often allow married clergy (but not bishops).

      Roman Catholic clergy used not to be banned from marriage in the past. It’s just a canon law matter, too, so the Pope can lift the ban at-will. Whether he actually will is wholly another matter.

  53. JohnBuridan says:

    When I was in High School, my Christmas project team made a ginger bread manger scene in which Jesus was a large marshmallow with a creme-colored marshmallow-head.

    One from the enemy teams ate our baby-Jesus.

    That is all.

  54. habu71 says:

    Regarding the grad student tuition waiver, it should be known that not all universities currently use this. For example, the 30,000 student university that I am currently a PhD candidate at does not use tuition wavers. Or, at least, does not use them in all colleges. Instead, the tuition amount gets recorded as a “fringe benefit” that grad students must pay taxes on.
    Don’t get me wrong, paying taxes on research credit hour tuition is both ridiculous and infuriating (although, truthfully, the same things could be said about most aspects of academic research funding…). But, as proven by the continuing existence of our grad program at my university, taxing grad tuition isn’t some new, previously unheard of practice and its spread is highly unlikely to cause the immediate death of all graduate schools in the country.

  55. Erusian says:

    This is not what I wanted my first post in SSC to be about, but here we go.

    I know SSC is a generally blue/grey space but part about the GOP tax plan is exaggerated. Not because the GOP tax plan is the greatest thing ever, but because you’re Still Crying Wolf. “Complete destruction” is “only a slight exaggeration”? It’s complete hysteria. It is exactly, down to the rhetorical device, what Republicans say when a new tax on business comes up.

    My prediction is that you’ll see some marginal drop outs (those who can’t get educational loans, find jobs, or who have better options that become more attractive.) Mostly you’ll see students begin to take on debt in the style of MBAs, MDs, or JDs. (Professions that are debt fueled and, you might notice, still very much exist.) In other words, they’ll start their professional careers debt ridden and pay it off with their new, higher earning potential.

    I’ve not heard a good counterargument for why mass drop outs will be the result, rather than something like this. Does anyone have one? (Completely sincere request here.)

    Also, this is how we treat everything else that looks mildly like a waiver. Loan forgiveness, for example. Do you really think students are more deserving of a special tax status than a poor worker struggling to get out of debt? Also, once employee discounts pass a certain threshold they can get taxed too. Student loans are an exception.

    And the bill is “next-level” targeting of political opponents? Do you think things like state and local tax exemptions just appeared out of nowhere? The tax code has been written to favor wealthy, urban areas. Now those areas are ideologically uniform, and their ideology is hostile to the ruling party. Unsurprisingly, they’re seeing policies favoring people who aren’t them. The case with the universities is similar. But this is the first time it’s happened in decades so they’re shocked.

    Their shock alone is probably sweet music to rural Republicans, who are accustomed to politicians rhetorically attacking their livelihood and ignoring their special interests. See how farmers were treated with the ACA, for example. They’ve suffered this type of targeting since at least the Nixon administration.

    This accusation also ignores where the Republicans go after ‘their own’. For example, modifications to how LLCs are taxed to make it harder for millionaires to get around taxes by incorporating themselves. They’re doing this less, but “Republicans reform tax code, Democrats get somewhat shorter end of stick.” is not as good a partisan rallying cry.

    In general: If your taxes are going up, it’s because you got an exemption. Instead of generally lowering taxes, they were lowered for your group specifically. That means, at some point in the past, your group was powerful enough to shift the burden of taxes onto everyone else.

    Now, are all exemptions bad? No, of course not. The government wants to incentivize certain things. But is the rhetoric you’re hearing, “We understand we have been given a privilege by the government and the rest of society. We are eternally grateful for our privilege and understand this is not a right, but here is why taking it away would damage not just us but everyone.” or “HOW DARE THEY!?”

    Lastly, vesting is something used by mature companies. It hurts companies that have raised money, have a lot of assets without liabilities, or are mature enough to have market valuations. Otherwise, the company can claim the company is worth the minimum possible, which will produce a very small tax burden. So it’s worse for companies like Google than startups. It does hurt them, though. I imagine the best way to do this would to apply it only to companies who cross certain thresholds, sparing small startups while taxing Google employees.

    PS: For the record, I think there’s a good argument the education tax credit should stay. So does the Republican led Senate. Let’s see whose version wins.

    PPS: If I might indulge in my little conspiracy theory, I suspect what is happening is that institutions are telling graduate students their futures are in danger. This is not because universities actually could not afford to continue or it would be legally impossible to continue their current practices. It’s because the universities would lose some of their favorite tax tricks. But no one shows up for ‘let the universities write off more on their taxes.’ So the universities claim that this will prevent them from serving lower income students. (Which, on the margin, it will. But say something vague about ‘as much as ten thousand’ and the entire body will be in uproar.)

    Of course, the university could choose not to modify its practices, thereby harming lower income students. But that casts the university as the villain, not the GOP. After all, what’s stopping them from discounting tuition for everyone with a certain income bracket? That would still be legal and impose no tax burden on the students. (Answer: They’d have to give it to everyone who met the qualification, they couldn’t force grad students to work, and it’s less advantageous for them at tax time.)

    PPPS: I also have another conspiracy theory it was included as a major concession to bargain away. The Trump-style Republicans score points for making educated, relatively well off professionals yowl and then get to bargain it away for the rest of the bill to be a compromise.

    • Matt M says:

      PPPS: I also have another conspiracy theory it was included as a major concession to bargain away.

      This is one of the few Scott Adams “master persuader” theories about Trump that seems to still stand up. Seems to keep popping up time and time again, and the left always takes the bait…

      • Jaskologist says:

        The “master persuader” theory always seemed iffy to me, but then, within the same week, Trump opponents declared that
        a) Patriotism is basically equivalent to support Trump.
        b) Kneeling before Trump is the new sign of #TheResistance

        So now I’m thinking there might be something there after all.

        • Deiseach says:

          I don’t think Trump is a master anything, but the extremes of pop-eyed frothing rage against him are so tying people up in knots that they’re contradicting themselves six different ways at the same time, making themselves look and sound ridiculous, and by comparison he and/or his administration come off looking “Well that was clever manipulation, who would have guessed?”

          I mean, now I’m seeing patriotism is good, nationalism is what is bad from the likes of those who up to now had been denouncing patriotism as dangerous demagoguery from right-wing Republican rich white old cis het dudes. Who knows, maybe saluting the flag will become the emblem of La Résistance!

      • Iain says:

        “Bargaining away” implies that Trump actually gets something in return. Do you have any examples? Otherwise it’s just a bunch of cases where Trump proposes something, people yell at him, and he backs down.

        • Matt M says:

          Not necessarily. What he “gets in return” is the parts of the bill that he doesn’t bargain away, but which would, themselves, have been seen as unacceptable extremism were it not for the fact that he originally proposed something worse, and then backed down from it.

          • Iain says:

            Okay, but what bills has he passed?

          • Erusian says:

            I actually see this more as an establishment Republican bill than Trump’s thing. Trump has repeatedly said tax reform is not a huge priority for him. He’s for it but I suspect he’s stumbled over his various initiatives and so he’s playing ball with the Republican leadership on this to get a legislative win. And in general, it’s very common to include something outrageous to negotiate it away.

          • AnonYEmous says:

            Okay, but what bills has he passed?

            this is a fair point for any other bills, but the tax bill seems like it might pass. If I recall correctly, the House’s bill has passed and the Senate’s bill has passed; unlike normal, I don’t think the latter is the continuation of the former, but I do think that they’re quite similar. So it’s pretty likely to pass mostly unchanged.

          • pdbarnlsey says:

            As others have noted, this might well happen, but there doesn’t seem to be evidence of it having happened in the past, which was the tone of your original comment.

    • swarmofbeasts says:

      Mostly you’ll see students begin to take on debt in the style of MBAs, MDs, or JDs. (Professions that are debt fueled and, you might notice, still very much exist.) In other words, they’ll start their professional careers debt ridden and pay it off with their new, higher earning potential.

      Students getting MBAs, MDs, or JDs can expect enough higher earning potential to make it worth it to go into debt for a graduate degree. Students getting humanities MAs and PhDs really can’t – most of my classmates are probably going to go into lower-middle-class nonprofit admin jobs or college adjunct teaching. Humanities graduate school is a really dicey proposition in terms of whether it’s going to improve your career prospects or your earning potential. Maybe it’s a good thing if tax changes discourage people from going to graduate school in the humanities – although many colleges will be in trouble if they no longer have a small army of grad students to teach freshman composition – but it will definitely change what decisions students make about grad school. (I guess this won’t go into effect before I graduate, if it passes, but I’m already thinking about if I’m going to have to drop out).

      I know less about graduate school in STEM fields – I know that students in some science disciplines have a high likelihood of getting well-paying private-sector jobs, but for students in other disciplines, or students who don’t want to go into the private sector, it’s likely to be a fairly strong disincentive.

      • Nabil ad Dajjal says:

        I know less about graduate school in STEM fields – I know that students in some science disciplines have a high likelihood of getting well-paying private-sector jobs, but for students in other disciplines, or students who don’t want to go into the private sector, it’s likely to be a fairly strong disincentive.

        In the biomedical sciences, the vast majority of PhD seekers plan on going into academia despite the fact that there are already too few professorships available and the pay for postdocs is lousy.

        Those of us who are looking forward to industry jobs will stand to make a lot more money. ‘Scientist I’ pays between double and triple a Postdoctoral stipend and there’s a clear career path. That said, if you’re dead set on bench science your salary will plateau very quickly. You’re far better off getting an MD.

        The best ways to make money in my field are the least appealing to most graduate students. Management in industry, consulting, patent law and (if you’re willing to take some big risks) entrepreneurship. I wouldn’t mind seeing more scientists who think like me but at the same time I’m not sure how that will affect the culture of science.

        • Erusian says:

          I’m a bit confused. The argument here seems to be that getting a humanities degree is a dubious financial proposition, but once it becomes a bad financial proposition people will drop out en masse? If someone offered you a loan to finish your degree, would you not stay? Especially knowing the median PhD makes (googles) 30k more than the median bachelors holder?

          Let’s say your tuition waiver is 50k, plus a stipend of 30k. Your PhD takes… eight years, say. Right now, post-tax, that’s about 25k. Let’s say the new bill passes but your state keeps up the waiver exemption. You’d owe $13,000, leaving you with $17,000 to live off of. If you need more and can’t work a second job, you could restore your previous income by taking on $8,000 in debt a year. This is $64,000 over the course of your degree. With a student loan at 7% over ten years, that’s less than $750 a month to pay it off.

          Or, putting it another way, you’re gaining a 30k median boost in income in exchange for 9k a year in extra expenses for ten years. That actually seems like a positive financial outcome. Though I might be missing something.

          • pontifex says:

            I’m a bit confused. The argument here seems to be that getting a humanities degree is a dubious financial proposition, but once it becomes a bad financial proposition people will drop out en masse? If someone offered you a loan to finish your degree, would you not stay? Especially knowing the median PhD makes (googles) 30k more than the median bachelors holder?

            “The median PhD salary” is a useless number for making decisions, though. You’re lumping together people from extremely different fields which have extremely different salary outcomes.

            Also, being financially successful and earning a PhD may both be separately caused by some third factor, like being smart. If I may use an analogy: people who can survive being shot at close range by a shotgun are probably healthier than people who cannot. But that does not imply that you should call up Dick Cheney and ask to go on his next hunting trip, for your health.

          • Nabil ad Dajjal says:

            Though I might be missing something.

            Yes you are. What you’re missing is that this PhD program already comes with substantial opportunity cost and isn’t going to do anywhere near as much for my future earnings as, say, an MBA would.

            My little brother is in an entry level corporate job and makes more than double my stipend with only a bachelor’s degree. The new tax plan would reduce that stipend to less than 1/10th of it’s current amount. I’m already taking a huge loss in earnings just being here but I refuse to go into debt on top of that!

            Luckily, I checked with my university and the way they handle their tuition waver means that it will still count as a scholarship under the new tax law. But if it hadn’t I would probably just leave and go into consulting. You make a lot less without a PhD but at least it’s a positive amount of money.

          • Matt M says:

            “The median PhD salary” is a useless number for making decisions, though. You’re lumping together people from extremely different fields which have extremely different salary outcomes.

            Are the costs of the program not correlated in any meaningful way to the expected salary outcomes?

            Furthermore, don’t you think that, on average, most people probably enter a PHD program expecting that they will end up doing better than average, not worse, right?

            It seems to me that people are howling about this, but the end effect, worst case scenario, is that various PHD programs become X% more expensive, with X being the current tax rate of someone’s current tax bracket while they are earning their PHD (likely to be in the lower brackets, even if you count tuition waivers).

            Obviously that will chase some people away, on the margins. But if we look at how college tuition costs have risen by like 5000% (citation needed) over the last few decades, it seems unlikely that a 20-30% bump (once again, that’s worst case scenario) is going to complete devastate and destroy the entire system.

          • Erusian says:

            This is a fundamentally financial argument, isn’t it? The claim is ‘when the bill causes higher education to cost ~10k more a year, it will virtually disappear’. It’s quite relevant to say, on median, most will still have a net financial gain out of what they spend on their PhD.

            It’s definitely true they could make more in, say, consulting. Opportunity cost and all that. But that’s true now. If you’re positing wide change (that a lot of people will drop out) you can’t explain it with a constant. You’re right it changes the economic calculus so that consulting is relatively more attractive. More people will then become consultants.

            But I doubt it’s everybody, or even the majority people. The demand for education is pretty inelastic, or else we’d see the vast increase in tuition destroying all higher ed.

          • pontifex says:

            Are the costs of the program not correlated in any meaningful way to the expected salary outcomes?

            As far as I know, they are not.

            Furthermore, don’t you think that, on average, most people probably enter a PHD program expecting that they will end up doing better than average, not worse, right?

            This is a weird comment. I knew a few people who got PhDs, and none of them ever mentioned financial reasons for doing so.

            It seems to me that people are howling about this, but the end effect, worst case scenario, is that various PHD programs become X% more expensive, with X being the current tax rate of someone’s current tax bracket while they are earning their PHD (likely to be in the lower brackets, even if you count tuition waivers).
            Obviously that will chase some people away, on the margins. But if we look at how college tuition costs have risen by like 5000% (citation needed) over the last few decades, it seems unlikely that a 20-30% bump (once again, that’s worst case scenario) is going to complete devastate and destroy the entire system.

            I think what you are missing is that this tax hits people’s current cash flow, whereas an increase in debt hits their future cash flows. Image the grim reality of “my take-home pay is 35k a year, and now the government says I have to pay an extra 10k a year in taxes.” People may literally have difficulty surviving on this amount of money, especially in expensive metros.

    • Brad says:

      Their shock alone is probably sweet music to rural Republicans, who are accustomed to politicians rhetorically attacking their livelihood and ignoring their special interests. See how farmers were treated with the ACA, for example. They’ve suffered this type of targeting since at least the Nixon administration.

      Err … are you completely unfamiliar with agricultural subsidies?

      • Erusian says:

        I am, in fact. Probably more familiar than them with you, unless you spent a lot of time in an agricultural region and had many friends/family members who owned farms.

        Agricultural price supports are a legacy of the New Deal. And like much of the New Deal, they were changed in the 60’s and 70’s. Nixon suffered a lot of protests over poverty and high food prices. So he purposefully pursued a policy of balancing agricultural price supports with depressing the actual cost as much as possible, largely by politically squeezing out small farmers in exchange for large ones. The policy was literally called ‘get big or get out.’ These large farms were increasingly corporate and fed into food giants. They reduced many farms to basically a state of debt peonage in order to reduce prices.

        Don’t mistake agribusiness for the average rural farmer. Perdue, for example, is a multi-billion dollar company in Maryland. They indirectly manage thousands of chicken farms (at least). Does keeping the price low benefit the person selling food or the person buying food? Does encouraging large, coastal corporations to run the distribution in a quasi-monopolistic manner to depress prices favor the consumers or the farmers?

        • Brad says:

          What’s your explanation for voters in Iowa and Nebraska and so on sending a continuing stream of politicians to Washington to defend agricultural subsidies including tariffs? If it is all to the benefit of large coastal corporations why would they care?

          And before you say corruption, were that the answer we’d expect the congressional supporters of, say ethanol mandates, to be equally likely to come from New Mexico or Maine as Iowa or Nebraska. Unless you are arguing that politicians from those states are uniquely corruptible?

          • Erusian says:

            Er, my first explanation would be that they don’t. The last two major Ag bills came out of Texas and Minnesota, both of which have major ag corporations and not particularly agriculture dominated economies. (Though they both do have large agricultural production.)

            If you’re looking for some system level theory, I would say that urban politicians want to maintain low food prices and agribusiness wants to maintain their subsidies. They form a coalition combining agribusiness’s quasi-rural influence (which includes all those fun things like voter suppression) with the genuine popularity of cheap food in the cities.

            Also, farmers are not a huge constituency. There’s a lot of closed farms, or bought out farms, or economically exploited farmers that impoverished communities along with factories.

          • If you’re looking for some system level theory, I would say that urban politicians want to maintain low food prices and agribusiness wants to maintain their subsidies.

            I’m not an expert on the current status of agricultural subsidies, but as a rule they have been designed to make food more expensive, not less expensive–buying agricultural output to hold the price up, paying farmers to hold land out of production. I think the biggest one at present is the ethanol requirement, whose effect is to bid up the price of corn and, indirectly, other crops grown on land that could be used to grow corn.

            What parts of current regulation are there with a similar or larger impact in the other direction?

          • Erusian says:

            Then why is food so cheap? Americans have the cheapest food in the world. Here’s an article, there’s a lot more on Google. Any theory that the government is keeping food prices high has to explain why, despite their efforts, food prices are very low.

            (Also, you know, politicians said they introduced the subsidies to make food cheaper.)

            Thinking of it another way, were the subsidies introduced when food prices were high or low? The answer is high. In the past hundred years bread prices peaked in the ’30s and ’70s, and this is precisely when we see subsidies receiving the most attention. The prices then decreased.

            A somewhat wordy example: Let’s say there are three farms: Farm A, Farm B, and Farm C. Farm A can make wheat for $3.00 a bushel. B $2.00 a bushel. C $1.00 a bushel. The federal government pays a $.25 bounty on each wheat bushel and guarantees a minimum price of $3.50. The market price for wheat is $2.50.

            In no-subsidy world 2017, Farm A goes out of business. Farm B makes $.50 a bushel and Farm C makes $1.50 a bushel. In 2018, the nation needs more wheat, so they’re willing to pay $4.00 now. Farm A reopens and makes $1.00 per bushel. Farm B now makes $2.00 per bushel. Farm C makes $3.00 per bushel.

            In subsidy world 2017, Farm A makes -$.50, plus $1.25 from the government, so they stay open with $.75. Farm B makes $.50 a bushel, plus another $1.25 from the government, so $1.75. Farm C makes $1.50, plus $1.15 from the government, so $2.75. In subsidy world 2018, the price never rises because production already meets the need.

            So Farm B and C make less than they would without the subsidies, plus the price remains stably low. Farm A still produces but probably goes into debt until they get their subsidy money (and debt is always a problem for a cyclical business like farming anyway…) They also have to deal with regulatory burdens and pro-corporate, pro-send it to the cities policies in regards to those subsidies.

            From the perspective of the bakeries, and their customers, the only effect is they have to spend less money on their bread. It’s effectively a wealth transfer, though I can’t speak as to whether agricultural subsidies are a net economic benefit. I can only say it’s not good for individual farmholders.

          • @Erusian:

            The article you linked to claims food is cheap in America, but puts all of its figures not in terms of cost but of percentage of income spent on food. Percentage of income spent on food decreases as income goes up, and America is a rich country. Western Europe is also rich, but it has trade barriers designed to drive up the price of agricultural products in order to subsidize its farm sector. Most of the rest of the world is considerably poorer than the U.S.

            And the comparison to earlier times in the U.S. is again put as percentage–Americans were a lot poorer in the 1930’s than they are now. Further, there has been progress in agricultural technology, which makes food cheaper.

            I again ask–what government agricultural programs are there that push the price of food products down by as much as the programs I describe push it up? If the purpose was to push prices down, why did the original New Deal farm program restrict the amount farmers could grow?

            You might try googling for information on the history of the farm program and see if it supports your view, which is the opposite of my understanding.

          • Erusian says:

            The article you linked to claims food is cheap in America, but puts all of its figures not in terms of cost but of percentage of income spent on food. Percentage of income spent on food decreases as income goes up, and America is a rich country. Western Europe is also rich, but it has trade barriers designed to drive up the price of agricultural products in order to subsidize its farm sector. Most of the rest of the world is considerably poorer than the U.S.

            And the comparison to earlier times in the U.S. is again put as percentage–Americans were a lot poorer in the 1930’s than they are now. Further, there has been progress in agricultural technology, which makes food cheaper.

            The price of bread, adjusted for inflation, is not a percentage of Americans income. So this critique is based on a false premise, or at least has further evidence to refute. It spikes in the Depression and again in the 70’s. These are the two periods of the biggest shifts in agricultural policy. As for agricultural technology, absolutely true. But if that were the only effect, you would expect to see us more on par with Europe or New Zealand. We’re far cheaper instead.

            I suggest you google around. Food in America is extremely cheap.

            I again ask–what government agricultural programs are there that push the price of food products down by as much as the programs I describe push it up? If the purpose was to push prices down, why did the original New Deal farm program restrict the amount farmers could grow?

            You might try googling for information on the history of the farm program and see if it supports your view, which is the opposite of my understanding.

            Respectfully, I know exactly how the USDA programs work. I’ve had family members who received money out of it. The USDA (sometimes) pays a flat bounty for each unit produced plus (almost always) they guarantee a minimum price for anything you sell. If you have to sell for below that price, the government makes up the difference. If you don’t understand how a subsidy drives up production and lowers prices, I suggest you see my example above. It’s supply-and-demand economics.

            If you’re talking about the Federal government’s policy of paying farms to dump produce, firstly that was repealed more than twenty years ago. And both periods where it was used (there were only two), food prices fell steadily. Secondly, keeping prices above a certain point encourages production, which prevents a sharp rise in food prices when production falls too low.

            Also, keep in mind the government was paying them. When you pay someone a price for dumping something, you are still establishing a subsidy for producing it. A subsidy still encourages overproduction and lower prices. In this case, the government tried to deal with the overproduction while keeping lower prices by dumping the excess while still paying for it.

            So in short, my answer is: Exactly the ones you think are raising the price. Except they aren’t. The government is lowering prices in net, but takes some actions to keep them from going too low. You’re paying attention to the actions that they take to keep them from collapsing completely and ignoring the ones they use to keep them low. Probably because ‘government pays millions for farmers to destroy crops’ has better optics than ‘government pays billions to overproduce food, driving prices down’.

            And again, if your contention is that government intervention was to keep prices high, you have to explain the empirical fact that food prices fell with government intervention. Even if you say none of that was due to Federal policy, why did they let it fall?

          • The USDA (sometimes) pays a flat bounty for each unit produced plus (almost always) they guarantee a minimum price for anything you sell. If you have to sell for below that price, the government makes up the difference.

            I agree that that would tend to push prices down–I was arguing in terms of the program as I was familiar with it quite a long time ago.

            The price of bread, adjusted for inflation, is not a percentage of Americans income. So this critique is based on a false premise, or at least has further evidence to refute. It spikes in the Depression and again in the 70’s. These are the two periods of the biggest shifts in agricultural policy.

            You don’t give your source for the price of bread. Checking in Historical Statistics of the United States, I observe that farm products prices drop sharply from 1929 to 1932, rise from 1933-1937. The Agricultural Adjustment Act, the original farm program, was passed in 1933. Quoting Wikipedia:

            The law offered farmers subsidies in exchange for limiting their production of certain crops. The subsidies were meant to limit overproduction so that crop prices could increase.

            I don’t have data for the 1970’s, since my edition of Historical Statistics only goes to 1970. But when I spent a summer as a congressional intern about 1967 and my Congressman proposed abolishing a large chunk of the farm program, the Agriculture Department’s response projected a substantial drop in agricultural prices and farm income if his bill passed.

            I don’t think you have responded to my point about the effect of biofuels requirements. As of 2013, 40% of the U.S. corn crop went to produce ethanol. That’s a huge increase in demand. The legislation was the Energy Policy Act of 2005 (and subsequent ones). If you look at a graph of corn prices, you can see the effect. It’s pretty striking.

            In summary: You are probably correct about the effect of current policies other than the biofuels mandate. You are mistaken as to the effect of policies in the past. The effect of the biofuels mandate is very large, so I don’t know if abolishing both that and the programs you cite would result in raising or lowering food prices.

          • baconbacon says:

            The price of bread, adjusted for inflation, is not a percentage of Americans income.

            Ctl F finds no instances of the words ‘bread’ or ‘inflation’ in the article you linked.

          • baconbacon says:

            @ Eursian

            If you’re talking about the Federal government’s policy of paying farms to dump produce, firstly that was repealed more than twenty years ago. And both periods where it was used (there were only two), food prices fell steadily. Secondly, keeping prices above a certain point encourages production, which prevents a sharp rise in food prices when production falls too low.

            This ignores half of the effect, keeping prices above a certain point prevents prices from falling when production is high. Actually that is likely to be more than half of the effect because production fairly rarely (in modern times) actually falls to low.

            All of your points miss half of the effects. When the government pays a bounty they encourage production of the lowest quality goods that meet the bounties standards. This pushes land that would be better off used for other production into specific categories. The price of bleached white flour falls, but it pushes up the goods that could/would have been produced with that land. For years the existence of small breweries was made very difficult by a variety of circumstances, during this time large breweries put out masses of extremely cheap beer. If you looked just at the prices of US produced beer it looked extremely low, non bottom of the barrel beer was fairly expensive (almost always imports) though. Citing the price of a loaf of bread as low leads to an inaccurate view, bottom rung bread is extremely cheap compared to other foods, but food as a whole can still be expensive. The article you linked brushes against this idea, its not that fast food is cheap, its often that fast food is really cheap vs the costs of buying and preparing fresh foods.

          • Erusian says:

            @DavidFriedman
            You don’t give your source for the price of bread. Checking in Historical Statistics of the United States, I observe that farm products prices drop sharply from 1929 to 1932, rise from 1933-1937. The Agricultural Adjustment Act, the original farm program, was passed in 1933.
            The data is from the USDA. I’d imagine farm products actually aren’t a great metric. I’m not aware of the specific statistic, but there’s an issue if it’s the price farmers get at sale. Bread is a consumer product and so determines what the end consumer actually pays. I’d fear they were somewhat decoupled by a strategy of government subsidies. Thus why I grabbed bread when I couldn’t find a general food price index.

            According to wikipedia:
            “Congress declared its intent, at the same time, to protect the consumers interest. This was to be done by readjusting farm production at a level that would not increase the percentage of consumers’ retail expenditures above the percentage returned to the farmer in the prewar base period.”
            In other words, they intended both to maintain a high enough price for farms to prosper and to lower food prices to prewar levels simultaneously. If I understand correctly. They did this through a combination of subsidies, encouraged overproduction, and destruction of excess. Or at least that’s what I’ve seen evidence for.

            I don’t have data for the 1970’s, since my edition of Historical Statistics only goes to 1970. But when I spent a summer as a congressional intern about 1967 and my Congressman proposed abolishing a large chunk of the farm program, the Agriculture Department’s response projected a substantial drop in agricultural prices and farm income if his bill passed.
            You’d expect prices and income to drop immediately. The farms would have less money because the government was giving them less. They’d lower prices to compete with each other. Unprofitable firms would then close, lowering production and driving prices up until they reached a high enough point profit returned to the remaining firms. Then the market would stabilize.

            I don’t think you have responded to my point about the effect of biofuels requirements. As of 2013, 40% of the U.S. corn crop went to produce ethanol. That’s a huge increase in demand. The legislation was the Energy Policy Act of 2005 (and subsequent ones). If you look at a graph of corn prices, you can see the effect. It’s pretty striking.
            Do you have a citation for that 40% number? I’ve seen as high as a fifth but never anywhere near half. Regardless, I think we can agree that bill had more to do with environmentalism than helping farmers. And the price has now dropped significantly, down to a much less dramatic increase (per your own graph, and the USDA has it even lower). That is arguably attributable to the fact food prices are currently rising in general. But a ten year period that’s now over is not really a strong argument for a sustained benefit to farmers.

          • Regardless, I think we can agree that bill had more to do with environmentalism than helping farmers.

            (your comment was a little confusing because you forgot to mark the quotes from my comment as quotes)

            At this point, most of the environmentalist movement has conceded that turning corn to ethanol does not have desirable environmental effects. To quote Gore on his support for the program:

            “One of the reasons I made that mistake is that I paid particular attention to the farmers in my home state of Tennessee, and I had a certain fondness for the farmers in the state of Iowa because I was about to run for president,”

            If you google on [ corn ethanol environmental impact] you can find lots more stuff on the lack of net environmental benefits. The program continues entirely because of the support of the farm lobby.

            It’s true, of course, that environmental benefits were the argument under which the program was started, but that says nothing about either the real effects or the real politics of it.

            Do you have a citation for that 40% number?

            I got if off a Forbes article.

            What was your source for bread prices?

            You quote from Wikipedia:

            Congress declared its intent, at the same time, to protect the consumers interest. This was to be done by readjusting farm production at a level that would not increase the percentage of consumers’ retail expenditures above the percentage returned to the farmer in the prewar base period.

            The announced purpose and effect of the program was to hold food prices up, not down. That isn’t changed by a statement claiming a limit to how high they were trying to hold them.

    • pontifex says:

      My prediction is that you’ll see some marginal drop outs (those who can’t get educational loans, find jobs, or who have better options that become more attractive.) Mostly you’ll see students begin to take on debt in the style of MBAs, MDs, or JDs. (Professions that are debt fueled and, you might notice, still very much exist.) In other words, they’ll start their professional careers debt ridden and pay it off with their new, higher earning potential.

      I’ve not heard a good counterargument for why mass drop outs will be the result, rather than something like this. Does anyone have one? (Completely sincere request here.)

      Well, sure, an MBA, JD, or especially an MD is something that you can probably reasonably use to pay down a few hundred thousand dollars in debt, over the course of a decade or two. Those are high-paying careers. A degree in literature, or music, or african-american studies is for the most part not.

      • Futhington says:

        Fewer degrees in music and african-american studies. What a travesty.

      • David Speyer says:

        An MBA or an MD is basically a guarantee of high earnings if you grind enough, or at least that’s how they are advertised. This was also true for JD’s 15 years ago, although I understand it isn’t any more.

        A high fraction of math PhD’s, and I imagine the same for all other STEM courses, go on to adjunct or to teach at low paying community and regional colleges. It’s one thing to ask someone to gamble 4-7 years of their life, it is another to stick them with a massive date at the end. If you’ve ever complained that all the mathematical talent of the country is getting sucked up by Wall Street, here is a proposal which will guarantee it.

        • pontifex says:

          Yes, exactly. Mathematics and science will be collateral damage in this battle between Red Tribe and Blue Tribe.

    • ugly hedgepiglet says:

      Is it not the case that this proposal was dropped two days after the original article was posted, and a week before this blog post we’re commenting on?
      https://www.reuters.com/article/us-usa-tax-options/senate-drops-proposal-to-change-taxes-on-startup-stock-options-idUSKBN1DF2O6

      I ran across this while trying to figure out how the options are valued for tax purposes, which I didn’t find but seems less pressing now.

  56. pontifex says:

    Re: empirical software engineering. The field has been around for a long time and never really contributed much useful knowledge to practical engineers (in my opinion at least). There was stuff like the CMM (capability maturity model), and ISO9000, which nobody thinks about much any more except as the punchline to Dilbert strips from the 1990s.

    And somehow, academics always managed to find that academic experience was universally useful and valuable, whereas practical experience was of no consequence. But practicing software engineers almost always found the opposite. It’s a very old discussion that was going on even before I graduated, last millennium.

    A cynic would say, academics who didn’t look down on practical engineers would simply join the ranks of practical engineers and collect a higher salary. So the academics who do not must have a very strong bias. (And you will observe this bias if you ever want to go back to school in CS…. academics count experience outside academia as a zero)

    P.S. I have also seen the kind of software engineering academics do. And it’s not good. See for yourself if you doubt 🙂

    • rlms says:

      “academics count experience outside academia as a zero”
      Only bad academics do that.

      • pontifex says:

        I agree that some individual academics might respect and understand industry experience, but “academia” as an institution does not. For example, you won’t get any closer to getting your PhD, let alone tenure, by pointing out that you worked 10 years at Microsoft and built big systems X, Y, and Z for them.

        Some day, if I manage to make “fuck you” money, I might consider going back to get a masters or PhD. But I definitely won’t consider it unless I am independently wealthy.

        • darmanithird says:

          For example, you won’t get any closer to getting your PhD, let alone tenure, by pointing out that you worked 10 years at Microsoft and built big systems X, Y, and Z for them.

          That’s kinda like saying that building a bridge won’t get you much towards a Ph. D. in theoretical physics. Some overlap in skillset, but being a researcher is quite different from being an engineer, and the subject matter is different.

        • rlms says:

          “For example, you won’t get any closer to getting your PhD, let alone tenure, by pointing out that you worked 10 years at Microsoft and built big systems X, Y, and Z for them.”
          Sure, and that’s how it should be. Engineering projects don’t necessarily have academic value. But they certainly *can* do; there’s a lot of corporate-funded research.

          • pontifex says:

            I agree that in many sub-disciplines of computer science, software engineering experience is irrelevant. For example, it would contribute little or nothing to complexity theory research or queuing theory research. But this is not true of all sub-disciplines of computer science. For example, systems software research, programming languages research, and empirical software engineering research could benefit from practical engineering expertise. But the incentives are not set up in such a way that this benefit can be achieved. So in practice, you have to take what academia has to say about these subjects with a rather large grain of salt.

  57. toBoot says:

    So, I’ve been prescribed xanax at various times in the last ten years for anxiety that was very obviously related to circumstances going on in my life (working really long hours on projects that I found morally questionable, finding out my partner was leading a double life, etc). I took it as needed, and generally never had my refills filled because I didn’t need it that often. But it was really, really helpful to manage what felt like unbreakable mental loops and occasional panic attacks.
    Then, recently, as I was undergoing a huge transition in my life, I was starting to notice some of the signs of anxiety returning. I asked a doctor about re-starting a xanax prescription and he said that he thought I could manage without (!). Another doctor told me the same thing(!!). A third doctor prescribed me hydroxyzine, but it was to treat my sleeplessness. It didn’t help me sleep – it made me tired, but I was already tired, so it just made me feel worse.
    I went to a fourth doctor to discuss my anxiety. (Note: I wasn’t doctor shopping, I was assigned these doctors by the clinic I went to). First he pressed me to consider an SSRI. I told him that my bouts with anxiety were so intermittent that that seemed like overkill, especially considering that one of my main issues is sleeplessness, and SSRIs might make that even worse. So he recommended hydroxyzine instead of xanax as it was now the “preferred method” for treating anxiety, and does not have the same addictive qualities. I told him it hadn’t helped with my sleeplessness, but I’d give it a try during the day if I was having trouble with anxiety.
    I have not found it helpful. It doesn’t seem like any of these doctors are able to look at my prescription history to see that I demonstrate basically the opposite pattern of someone who is at risk for addiction (ie. I don’t tend to get prescription refills). And while it does sound like there’s some evidence that antihistamines, and especially hydroxyzine work better for treating anxiety than benzodiazapines, that doesn’t seem to be the case for me. It’s incredibly frustrating to not have access to a treatment that works for me, and to have doctors repeatedly substitute their judgment for mine. I understand that doctors probably have to do this with patients who don’t have the education or ability to understand the risks associated with their treatment. But that’s never even been a conversation that any of these four doctors have had with me. And, in case it’s not obvious, I am quite capable of understanding and contributing such a conversation.
    I’m curious if I’ve just had bad luck, or if beznodiazapines are just medicine non grata now as a result of reported rising rates of prescription drug abuse?

  58. MugaSofer says:

    Also, some neural net is very convinced that “either an animal face or a car body” is a fundamental concept that cleaves reality at its joints, and now I’m questioning how I know for sure that it isn’t.

    I side with the neural network here; the front-ends of cars totally look like weird animal faces.

  59. James Koppel says:

    I’m a researcher in software engineering. I don’t work in empirical SE, but I know people who do. I’ll take questions about research.

    • pontifex says:

      Hi James,

      What’s your opinion on model checking? How about the Midori experiment?

      • James Koppel says:

        Model-checking has a pretty nice promise: write your specs, take your program, get it checked. It’s a 100% automated approach.

        It’s been very successful, but, overall, I’m pretty skeptical that the approach will scale to programs much more sophisticated than device drivers. Software model checking relies on automatically generating abstractions of your program state so that your infinite program state becomes finite. However, these abstractions are not influenced by the design of your program, and so they’re limited in ability to help reason about your program. So I still hear model checking people talk about trying to check programs with 10^100 states, which is absolutely tiny. I’ve played with two model checking tools: SMV doesn’t check actual code, while Java PathFinder suggests that you “model” your program in a restricted subset of Java.

        If you’re considering using model checking at work to check your software, then, unless you’re doing hardware or device drivers, consider it off the table. However, you may have some protocol (e.g.: replicated storage) that can be described essentially as a state machine. Model checking works great for that and has been successfully used at Amazon and (kinda) Dropbox.

        I haven’t heard of the Midori experiment, and didn’t find anything from a quick Googling.

        • pontifex says:

          Is there a paper about the Amazon / Dropbox experience with model checking? Sounds interesting.

          Midori was a Microsoft project to rewrite the entire operating system in a managed language. Another code name for the project was Singularity. There is not a lot of information available about it online but there is some here and there. The project appears to have failed. There is a rumor that Microsoft put a lot of resources into this as the successor to Windows, at one point.

  60. Joyously says:

    You can sometimes get grants to pay grad tuition. Of course, without that you just get a waiver anyway, so it’s more or less a scam to direct more money to the university while pretending it’s going to the student.

    I have mixed feelings about taxing tuition wavers. They’re a weird fiction which means they’re not really income going to the student, so why should they be taxed like income? But on the other hand, they’re a weird fiction that is propped up by the university system causing weird distortions, so why shouldn’t the tax code call their bluff?

    It’s similar to student loans–should the government help students with student loans since they really are a raw deal, but does that just help propagate the raw deal?

    • Joyously says:

      I’m not sure whether it would be especially bad for my university. On the one hand we give relatively small grad stipends (we’ve been trying to make them bigger) so adding another financial burden to grad students could hit us particularly hard. On the other hand our nominal tuition is smaller than other universities, which means our waivers are smaller.

    • should the government help students with student loans since they really are a raw deal

      In what sense are the loans a raw deal? You can argue that what they are used to buy is sometimes a raw deal, but that’s not the same thing.

      • Joyously says:

        You can argue that what they are used to buy is sometimes a raw deal, but that’s not the same thing.

        I think they are though. Or more accurately that they’re both part of the same raw deal. We put more emphasis on degrees than we should, prices rise, students pay more than they probably “should”, student loans keep the price from breaking them but probably also inflate the price…

  61. danjelski says:

    I certainly understand the political motives for dissing higher education, but public policy is also well served by the new policy. We are grossly over-invested in education generally, in higher education specifically, and even more so in graduate education. Just look at all the humanities PhDs who can’t find jobs in their discipline. And that rot is even spreading to STEM disciplines, esp. in biomedical fields.

    When you subsidize something, eventually you get too much of it. The new tax plan at least goes some way toward reducing the subsidy.

    • pontifex says:

      I don’t think we’re over-invested. I think we’re over-leveraged. We have too many people taking out huge loans for education and too many people giving out huge loans. And the underlying asset, the degree, is starting to look somewhat subprime.

      The problem is, young people are stuck in a metaphorical Red Queen’s race where they must run in place to stay where they are. If you don’t take out the big loan to get the expensive education, you fall out of the middle class and into the lower class.

      I think firstly, we should allow student loans to be dischargable by bankruptcy. There is absolutely no reason why student loans should be treated differently than other loans in this regard. If lenders had to consider the performance of the underlying asset (the degree) when making their loans, we would start seeing some academic birth control– especially in the humanities. And everyone would be better off.

      Secondly, we should limit the amount that students can borrow. Or perhaps stipulate that at least 1% of the loan has to be paid every year. This would limit the damage from the degree arms race that we’re seeing.

      • I think firstly, we should allow student loans to be dischargable by bankruptcy. There is absolutely no reason why student loans should be treated differently than other loans in this regard.

        There is an obvious reason–you can foreclose on a house or a factory, but not on an education. If student loans are dischargeable, lenders will be reluctant to lend to anyone who doesn’t have substantial assets because the borrower can spend a hundred thousand dollars of borrowed money getting an education, then wipe out the debt by bankruptcy while retaining the education and the increased income it gives.

        • pontifex says:

          Defaulting hurts your credit rating, though. So it’s not “free.” Also, if lenders carefully consider whether a student loan is really worth giving, that’s a feature, not a bug, in my opinion. And there will still be need-based scholarships.

      • Brad says:

        Better than 90% of new educational loans (on a dollar weighted basis) are made by the government. Trying to effect a change in lending behavior indirectly via a change in the bankruptcy code doesn’t make much sense. First because if you are writing legislation anyway you could just change the lending behavior directly and second because the you can’t change the behavior indirectly because the lending criteria are written into the law.

      • The Nybbler says:

        The reason student loans are non-dischargeable in bankruptcy is so they aren’t grants. If they were dischargeable, most students would strategically default and declare bankruptcy immediately after getting their degree (income: 0; assets: near 0; debt: $$$$$) and get a free education as a result. With non-dischargable loans if you screw up the government can force you to live in poverty for the rest of your life, so there’s at least a little incentive to be responsible.

        As Brad points out, Obama federalized the student loan system, so they’re made directly by the government instead of guarantees like they used to be. I’d suggest just cutting off government student loans entirely, but the result would be utter chaos.

        • Protagoras says:

          Starting your life with a thoroughly screwed up credit rating isn’t an obviously wonderful scenario; a lot of employers look at credit score as a measure of trustworthiness, it screws up your car insurance rate, makes it difficult to get an apartment, etc. Certainly there are people for whom it would work out as the best bet, but all the evidence I’ve encountered suggests that student loans aren’t particularly subject to greater strategic abuse than typical loans of other kinds.

        • The Nybbler says:

          Certainly there are people for whom it would work out as the best bet, but all the evidence I’ve encountered suggests that student loans aren’t particularly subject to greater strategic abuse than typical loans of other kinds.

          That’s because they are currently non-dischargeable. As for the bad credit rating, it just moves you from “no credit” (which is bad) to “bad credit”. Is that worth an average of $37,000? I’d say “you bet”. And I suggest so many people would do it if they could that for new grads, it would not be useful as a measure of trustworthiness.

          • Protagoras says:

            There has been some research on what happens when the law changes, which is what I was referring to. Your suggestion would be of greater interest if there were actual evidence for it.

          • quanta413 says:

            @Protagoras

            Can you give a citation (or two) to this evidence you refer to?

          • Protagoras says:

            Here you go. TLDR; changes in bankruptcy law to making private loans non-dischargable in 2005 seemed to have no discernible effect on default rates, suggesting that there had not been widespread opportunistic defaults exploiting the previous legal regime. Of course, the private and federal markets are importantly different, but since the same argument was made for why private loans should become non-dischargeable in 2005, this should cast some doubt on the wisdom of accepting the argument without empirical evidence.

          • The Nybbler says:

            It may not have affected default rates, but it _did_ affect the size of the market.

            The fact that the private student loan industry saw a 50% increase in the years immediately following a change in the Code pertaining to private student loans is likely no coincidence.

            So, private student loans which were too risky to make before were now safe to make. I don’t know how private student loans were guarded against strategic default before 2005, but it seems they were, as without that restriction, more loans were made.

          • quanta413 says:

            @Protagoras

            Thanks! When I think about it, I find the combination of outcomes not easy to reason about. Presumably before the loans became non-dischargeable, lenders had a strong incentive to avoid loaning out to anyone for whom bankruptcy would later likely be a better option than paying off the loan. It’s not obvious to me a priori that I should expect a large change in the number of people trying to cheat when the law changes in this direction. That also brings up the question of why the law needed to change exactly. It obviously optimizes one part of the bankruptcy system in a totally different direction than the rest of it (more credit available at lower rates, but debt is more punishing), and I don’t see why that should be done. It seems wrong to me. I understand why government loans work differently though (because otherwise taxpayers would be on the hook forevermore because the government has much less incentive to not lose money).

            Anyways, about the effects. After the law changed, loans became less risky to lenders so presumably they’d make more loans than before if the real rate of interest was unchanged. They did make more loans so that makes sense, but that could also be other changes in the market.

            I feel like to answer my own question about what I even empirically expect at this point, I’d have to set up a game theoretic model.

          • moonfirestorm says:

            Couldn’t you just get a bankruptcy-dischargeable loan anyway and use that to pay for college? Banks do make loans outside of student loan deals.

            I mean granted, you have no collateral and low chance of repayment, and there’s basically no way any intelligent bank would take that risk, and if they did you’d end up with a ridiculous interest rate, but isn’t that basically the situation we’re describing anyway if we made student loans dischargeable in bankruptcy? No need to remove the option of non-dischargeable loans.

          • Protagoras says:

            @moonfirestorm, Indeed, banks love giving credit cards to college students, which is yet another of the reasons I think all this concern over strategic defaults must be greatly overestimating the appeal of bankruptcy to recent college graduates. Or perhaps overestimating the willingness of bankruptcy judges to be convinced that someone who just got their degree is absolutely unable to get a job and actually repay their creditors?

          • moonfirestorm says:

            @Protagoras:

            Do banks give college kids credit cards with significant credit limits though? I remember only being able to get a $300 limit coming out of high school. Even if that gets maxed out and the interest rate is nasty, that’s the sort of thing you can pay off with a fairly low-wage job, and is probably still cheaper than figuring out bankruptcy proceedings or the credit complications.

            It doesn’t seem like that would scale up to college-tuition amounts, which take a while to pay off even if you do get a good job.

          • Protagoras says:

            It was some time ago, but I’m pretty sure my first credit card had a much higher limit than $300. And anyway there are devious things you can do to build up your credit rating over the course of a few years, at which point with multiple cards you could have a lot of available credit on credit cards. And people sometimes do that and strategically default, but it’s not incredibly common (presumably because the profits, while not trivial, are not enormous, and having a disastrous credit rating is a major pain). Similarly, of course the ultimate exploit of bankruptcy is to form a corporation, fund it through debt, pay yourself a big salary, and let the corporation go bankrupt; again, it is something that happens, but in order to do it successfully you need to be pretty good at manipulating the system, so it’s not an easy money strategy that just anyone can employ (though that one doesn’t even screw your credit rating). I continue to suspect that the people fretting about strategic defaults on student loans are underestimating the downsides to the strategy (and perhaps also to some degree complaining about problems for which making the loans non-dischargeable was the wrong fix, but overall my impression is that lawyers wanting to make bankruptcy harder to give themselves more work are a more powerful lobby than people going through bankruptcy).

          • Edward Scizorhands says:

            The fear that students would load up on debt and strategically default was never based on reality.

            I would like to see just one example of someone strategically defaulting on medical school debt and then becoming a doctor, back before the rules changed. (1979?) Bankruptcy court proceedings are public so it should be possible to find these. Especially if there were a lot of them.

            The fear was always overblown because you would still need a bankruptcy judge to sign off on it.

          • Matt M says:

            I don’t have data on this, but I’ve heard anecdotes that the military used to struggle with ROTC students completing valuable/expensive training (like pilot school), and then, upon completion, declaring themselves to be conscientious objectors.

            This is almost a “strategic default”, in a way.

      • Matt M says:

        If you don’t take out the big loan to get the expensive education, you fall out of the middle class and into the lower class.

        I don’t think this is true. There are a lot of things you can do to optimize for not ending up with massive loans. You can select in-state public schools over out of state or private. You can endeavor to find some sort of scholarship, and base your selection of school on whoever offers you the best deal. You don’t have to go to Harvard just to stay in the middle class. And once the choice of school is taken care of, you can carefully select a major that leads to a high paying career.

        Running the race isn’t optional, but there are definitely smarter ways to run it than others. The problem we have isn’t student loans. It’s that we tell students to go to whatever college they like the best and to study whatever they’re most interested in and to ignore all costs. Well, that works about as well as it sounds like it would work.

        Also, there are also even more out-of-the-box solutions out there. I enlisted in the military and you suckers taxpayers ended up paying for two of my degrees. In fact, for my MBA, I got a full ride from a 20-ish ranked state school rather than paying full price at top 10 private school, which ended up being additive to my MGIB, so actually, I got paid to go to school.

        • pontifex says:

          @MattM, I also got a scholarship which paid for much of college. But college prices have gone up since then. It’s not uncommon for students to pay $30,000 per year now, which is much more than it was when we went to school.

          “Major in something safe and lucrative and get a job at Google” might be good advice to give to an individual person. It’s poor advice to give to a society, because there are only so many jobs at Google. And majors are only safe and lucrative because not everyone is doing them. If everyone gets an MBA, then the worth of MBAs plummets. Unless… your MBA comes from someplace exclusive, like Harvard. And I think this is about where we came in?

          • Matt M says:

            “Major in something safe and lucrative and get a job at Google” might be good advice to give to an individual person. It’s poor advice to give to a society, because there are only so many jobs at Google.

            Good, because that’s not my advice.

            My advice to society is not to do any one specific thing. It’s “avoid taking on massive debt to acquire a largely useless credential.”

            Note that this is a conditional AND here. Taking on massive debt is fine, if you’re using it to get a Harvard MBA (or even a mid-tier MBA, STEM degree, etc.). Obtaining a useless credential is fine, if you have the economic means to do so without taking on massive debt. But doing both of them is a terrible idea. Always. Everyone can follow this advice and society will be no worse off.

  62. divalent says:

    I may be late to this discussion on Grad Student tuition (and I can’t be sure someone else has already made the same points), but …

    Graduate student tuition is not a fiction or merely an accounting gimmick. In many cases, the University does get the full amount they charge, although in other cases they may not. But there are strong incentives in place to keep it high.

    1. Gov training grants that support a number of grad student slots at an institution will include tuition as one part of the total that makes up one “slot”. The university gets this tuition component.
    2. Individual research grants to a professor that has $ to fund one or more grad students will have tuition as part of the total for each student. The university gets this tuition component.
    3. Similarly for third party scholarships or endowments that go to the institution (or a department at an institution) to support grad student slots.

    IOW, the total cost of the slot = Stipend + “Full Price” Tuition. If they reduced the tuition by $10,000, the university would lose $10,000 for every student on a grant-funded slot.

    However, in some cases the institution may support slots internally, usually as some sort of guarantee that a department or program has X number of slots they can keep full regardless of fluctuations that can’t be completely controlled (with the expectation, of course, that the faculty work diligently to keep outside funding of as many as they can). In these cases (where outside direct support is not there), the university would actually be covering the costs of the slot, although the total $ burden would be less than what they would get if a Gov grant covered it. The cost to the university would be Stipend + Marginal cost of educating a student. (Students are not a trivial burden on the institution, although generally the marginal cost is far less than the “Full Price” tuition.)

    And as others have mentioned, most humanities and professional school students are often actually paying the tuition themselves. So they have an incentive to keep tuition high so that it appears roughly equivalent to what others pay out of pocket.

    So there currently is an incentive to keep the “full price” tuition as high as it can be justified, because there is profit built into it (“full price” tuition > marginal cost of education) when being paid by outside entities.

    • gbdub says:

      So the (non-profit) institutions are billing much more for “tuition” than the marginal cost of providing education, and get away with it because their funders are cost insensitive. Uncharitably, this is fraud. Charitably, shouldn’t this be called “overhead” and treated the same way overhead is in private industry?

      It kind of seems to be a case of universities using a nominally high tuition to, on the one hand, justify paying grad students super low salaries (“look at this huge benefit we give them”) while on the other justify exorbitant overhead rates (“look at this huge expense we have to cover!”) all on the backs of our generally positive feeling toward universities. A for-profit corporation doing similar things would be slammed (how much grief does Wal-Mart get for its employees qualifying for federal aid – why don’t universities get the same criticism?).

  63. Ben Thompson says:

    “…some neural net is very convinced that “either an animal face or a car body” is a fundamental concept that cleaves reality at its joints…”

    The net was trained on ImageNet, which is full of photographs taken by human beings, who are inordinately fond of photos of animal faces and car bodies, which do share some visual properties (e.g. symmetrical, smoothly shaded and contoured thing with two shiny spots). This net cleaves human visual imagination at a joint.

    • The Nybbler says:

      Some cars (notably the original Miata mentioned above, but probably many others) have their front end _designed_ to look like a face. And nearly all cars start with one feature common to faces, the two “eyes”. Add a central grill lower than the headlights and you’re well on your way.

  64. Questioner says:

    Related: Trump tax plan to hit colleges by changing some of their weird tax exemptions. Taxation has always been a little about punishing your party’s political enemies and giving tax breaks to its allies, but this is some next-level stuff here and it’s really blatant how much the new code shifts tax burden onto traditionally Democratic constituencies.

    1: Apparently you have forgotten about Obama using the IRS to go after the Tea Party. well, rest assued, we haven’t forgotten.

    2: “A little”? Massive taxes on private planes / yachts? “Sin taxes”? The tax code has always been about punishing some choices, and rewarding others, and the Left has used it for decades to go after those they don’t like

    Now the GOP, run by a New York Democrat turned “Republican”, is playing by the Democrats rules, and returning all the favors of the last century.

    You hate that? Good, you should. When you hate it enough to actually be willing to dump all the people on your side who’ve been doing it to us, we’ll see that.

    But Colleges and Universities that operate as power centers for the Left / Democrats are our enemies. We’re not going to bother taking them back, we’ll just destroy them, and every other organization that wants to be partisan for the Left.

    • Nornagest says:

      Massive taxes on private planes / yachts?

      Do these actually exist? I’ve never heard of them, and the only one Google turns up is a 10% tax on certain luxury goods (including mid- to high-end planes and boats), signed into effect by George H.W. Bush in 1991 and repealed two years later. I wouldn’t call that massive (in revenue terms it’d be trivial, which is why it was repealed), it’s not clearly partisan in its original context, and it’s certainly not relevant to 2017 politics.

    • Ilya Shpitser says:

      Some time ago, I said here that one important distinguishing characteristic of Trump supporters, other than their non-existent epistemic standards of course, is their willingness to throw everything into the fire in exchange for short term gain in the culture wars.

      Here is exhibit one, the mental giant above, who is a-okay with throwing higher education STEM, the engine of American prosperity, into the fire.

      The rural south/center picking a fight of this type with the coasts is a mistake, just like last time. It’s a mistake in the sense that it’s bad for the overall health of the US, but it is also very bad for them. Partly because they themselves will be far worse off in the long run (STEM is one of the few things with a consistently huge long term return on investment), but also partly because I think people are not appreciating what would happen if the coasts got serious about playing this type of game in retaliation.

      Remember, the coasts and urban centers is where the money, the brains, and technology lives. They are already subsidizing the rural parts of the US, in a ton of ways.

      • Questioner says:

        1: I wasn’t a Trump supporter. I supported Cruz, then stopped paying attention when he dropped out.
        But watching Trump burn down the Left is slowly changing that.

        2: I have a STEM MS

        3: STEM isn’t going into the fire, because it occasionally provides things of value to pay for it. Or, if it does go int eh fire, it will because the rest pulled it down to die with them. In which case, thats their fault for not separating from the rest

        “X” Studies? That’s going to be going into the fire, and good riddance.
        All the fields turned into pseudo-intellectual circle jerks? Let them all burn, I will dance by the fire.

        Am I willing to pay the cost of throwing everything dominated by teh Left into the fire? Why yes, i am.

        Here’s some names for you:
        James Damore
        Memories Pizza
        Sweetcakes by Melissa
        Masterpiece Cakeshop
        The Little Sisters of the Poor
        Hobby Lobby
        Not a name: “Punch a Nazi”

        You on the Left have spent decades reveling in your power to destroy those who do not sufficiently bow down to you. So yes: any time we can hurt you in any way? We will. As long as the cost to you is greater than the cost to us, we’re good to go.

        You capture some cultural heights? Doesn’t matter what they used to be worth, we’ll burn them down. Don’t want your “heights” burned down? Don’t let the Left capture them.

        Ask Hollywood, the NFL, and ESPN how being of the Left is working for them. I used to be a big pro football fan. I haven’t watched a game (or paid any other attention to them) since they let Kaep pull his act

        I used to be a big fan of my alma mater’s sports teams. I haven’t watched / listened to / read about any college sports since the NCAA started their North Carolina boycott.

        My wife really appreciates how much more time I have for her, now that I’m not wasting it on those sports, so it’s a big win all around.

        You taught us well, Now it’s time for us to show what we’ve learned

        Enjoy!

      • Questioner says:

        No, the Coasts are where the credentials are.

        The fact that you can’t tell the difference between credentials, actual education, and actual brains, is a big part of your problem.

        Another big part is your grotesque ignorance. The industrial heartland isn’t with you, which is why Trump won, and Hillary lost. And if you really want a Civil War, you might stop and ask yourself who owns the majority of the 300 million guns in private hands in the US.

        Hint: it’s not “your people”

    • David Speyer says:

      In fact, out of 146 groups flagged by the BOLO criteria, 111 were liberal and only 19 conservative. I will admit I’m getting this summary data from the left wing source where I learned about it, but you can read the Treasury department original report if you prefer. I particularly point out Figure 4 on page 17 — the largest classes of organization targeted were groups calling themselves “progressive”, groups concerned with medical marijuana and healthcare legislation, and spinoffs of ACORN.

      I could agree with saying the center was using the IRS to target groups outside the Overton window, but the Tea party wasn’t the main target.

      • Douglas Knight says:

        The new report, supposedly casting a wider, more systematic net, catches fewer cases than the old report.

        I think that the new report simply doesn’t look at organizations with “tea party” in their names. I don’t know why they exclude it, but I think that’s why the organizations it hits are so liberal.

        The old report claimed that some organizations were flagged because they had “tea party” in their names. The new report looks at organizations flagged for having one of 17 terms in their names, none of which is “tea party.” Several terms were chosen specifically because they were believed to mean “tea party,” so I don’t think the new report is claiming that the old report was completely false in its claim that organizations are targeted for using the term “tea party.” Maybe “tea party” wasn’t on the BOLO list because no one needed to be reminded of it.

        I think that’s what’s going one, but I’m not sure. But I am sure that it is important to explain why the new report has fewer cases than the old report. Perhaps all of the organizations in the old report are among the 927 organizations in this report. That would mean that more than just the 146 were investigated to the standard of the old report. And the politics of the 927 might be different than the politics of the 146. Some 428 (469?) organizations sued the IRS for being unfairly treated. Perhaps they are distorting their treatment, but they are probably telling the truth about their names. That would be a big chunk of the 927 organizations. (Also, it appears to me that the new report looks at the intersection of cases that IRS should look at with the cases it did look at, which is so stupid, I have to doubt my interpretation. But I can’t see any other way to read it.)

  65. JoeCool says:

    I suspect social desirability bias exaggerates peoples aversion to selection.

    Still shocking to me, since this seems to imply a lot of people are pro abortion but anti selection which makes no sense to me. I guess maybe they want to preserve diversity that comes with genetic disorders?

    I know a lot of media outlets created quite a hubub when they realized iceland doesn’t give birth to down syndrome people anymore. Even I get the feeling that something is lost when that happens.

    A lot of disability rights activists/polar opposite religious fundy activists really hate it.

    Then again we also have a dyslexia only sperm bank at the moment, so maybe Neuro untypical wouldn’t die out because weird people would cultivate them

  66. Luke Somers says:

    Scott, your link to the ‘openAI makes us less safe’ article in turn links back to one of your articles, the Superintelligence FAQ, but that link is dead and I cannot find it. Even if you took it down intentionally, you could put something there to redirect people to a relevant resource?

    Also, the link to the reddit comment is to one of the less relevant-looking examples on that page.