Open threads at the Open Thread tab every Sunday and Wednesday

OT25: Obon Thread

This is the semimonthly open thread. Post about anything you want, ask random questions, whatever. Also:

1. Comments of the week are Kavec talking about what work in military intelligence has taught them about probability and thinking, and John Schilling (1, 2) on waterways and “hydrologically-induced smartness”.

2. I am slowly going through reported comments to see if I need to ban anyone. Dr. Beat has been banned for one month for this comment (compare this). Anonymous commenter banned indefinitely for this and this. Also – and I guess Topher was right about us after all – Science has been banned for one week

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

707 Responses to OT25: Obon Thread

  1. Jeremy says:

    Thoughts on Truthcoin? It’s a use of Bitcoin blockchain technologies to implement distributed prediction markets where everyone’s incentive is to act fairly (including those deciding the correct outcome). According to the authors (I’m not really qualified to evaluate it), it would allow trustless anonymous prediction markets to be created by anyone (with incentives to create easily evaluated markets that will attract high volume).

    If such a system became large, it could test a lot of the ideas behind prediction markets, and perhaps offer successful rationalists a chance to earn a quick buck.

    • Murphy says:

      I had a glance through and I’m not terribly impressed by their slideshow of 300 quotes.

      I’m still unclear about how it verifies information. It seems to just be a vote system with people buying votes where the incentive is to side with whatever position the majority votes for. Common misconceptions wouldn’t appear to work well on there and anything with strong popular ideological backing/bias isn’t going to work very well at all.

    • Sam says:

      Check out, an implementation of an (essentially) identical protocol that’s a bit closer to reality.

    • AR+ says:

      I think there’s something of a problem with the selection of a proof algorithm in general. Bitcoin itself and pretty much every derivative I know of except Primecoin and Riecoin are basically digging holes and filling them up again, at huge costs of electronics and energy, far beyond the point at which this meaningfully improves security.

      It had occurred to me that the proof algorithm of the cryptocurrency that attains widespread adaptation (if such an adaptation occurs) may determine a significant fraction of all per-singularity computing power that humanity ever consumes if it’s a pure proof of work algorithm.

      • Evan Daniel says:

        Note that Bitcoin-style proof of work is no more wasteful than any other option.

      • AR+ says:

        Opps, meant pre-singularity computing power.

      • CJB says:

        I’m just amused that people thought “we need a non governmental currency! I know- lets make control of that currency entirely dependent on computing power!”

        If the NSA DOESN’T own most bitcoins being generated now-Id he genuinely shocked.

        • Pku says:

          Damn, that’s sneaky. I never thought of that.

        • AR+ says:

          They’d have to have invested in the specific sort of ASIC used to mine bitcoin to be competitive. The days when CPUs and GPUs could be used for mining are long over.

          And if they did that, they’d just be miners. Having lots of bitcoins doesn’t let you do anything to other users, and there are limits to what even a 51% attack can accomplish (for example, it doesn’t let you take other people’s coins.)

          Rather than trying to own bitcoins, the NSA is, if they’re doing anything with bitcoin at all, probably developing better means of analyzing the blockchain. Bitcoin is pseudonymous, but is arguably the least anonymous currency ever created, since every sender, recipient, and quantity of every transaction is permanently recorded in a public ledger. Having automated means of tracing paths across the blockchain between illegal sales and subpena-able exchanges would be far more useful than owning bitcoins or hashing power.

  2. Pku says:

    Request for help:
    I’m currently a math grad student (graph theory and algebra), and I’m planning to switch to software engineering/related tech field after I graduate. I know some programming and I’m pretty good at it (as in, my sophomore CS major discrete math students also frequently came to me for help with their programming homework), but I don’t have a degree/massive personal experience. I’d appreciate any advice on stuff to learn or about finding techy jobs for someone with little to no experience in dealing with tech culture. (MIT open courseware has been pretty helpful so far).
    Thanks. Anyone with useful advice gets a free invisible dragon for their garage.

    • Jesht says:

      I’m in a similar situation as a mechanical engineering grad. I can’t offer industry experience but I’m happy to share the resources that I’ve found helpful/interesting. There are a lot of great archived MOOCs on Coursera that I recommend as being engaging and high quality, if you’re interested in continuing to learn:

      Algorithms 1 and 2 by Tim Roughgarden (Stanford)
      Machine Learning by Andrew Ng (Stanford)
      Automata (Stanford)
      Discrete Optimisation (Uni Melb) – fun projects
      Nand 2 Tetris looks promising
      Compilers looks promising

      In general I’ve read that it’s good to have a GitHub that showcases some projects that you’ve done. People seem to also mention getting quizzed on algorithms and data structures in interviews.

      • Pku says:

        Thanks, I’ll look into those.
        I’ve done nand2tetris, it’s great, seriously worth the time if you get to it.
        (The one issue I’ve had with it is that the simulator for the high-level programming language is a bit buggy, so I had trouble figuring out if the problems were with my algorithms or the system, but the really interesting stuff is mostly before then anyways).

      • Shenpen says:

        Learn stuff, then learn programming specifically with the focus on how to automate the stuff you already know. More:

      • AnObfuscator says:

        In general I’ve read that it’s good to have a GitHub that showcases some projects that you’ve done. People seem to also mention getting quizzed on algorithms and data structures in interviews.

        Yes. This is extremely good advice. I regularly interview programmers, and seeing an active GitHub account is a pretty positive sign.

        FWIW, I’ve never had an interview where I was expected to regurgitate information about DS and Algos; however, I have had a lot of interviews where I was expected to solve problems using them, and be able to reason about time and space complexity.

      • Josh says:

        Couple thoughts (I’m the founder / lead engineer of a nyc-based startup):

        1. Know your market value. This seems to be EXTREMELY geographically variant. For instance, if you’re in the US in the bay area or NYC, employers cannot find enough half-way competent programmers no matter how hard they try, and you should not accept an offer that you’re not excited about (both the people / job + the compensation). If your local market is not like that, you should be aware that moving or interviewing at companies that accept remote workers might be worthwhile.

        2. Write lots of code! If you’re coming from an academic background, there are some specific jobs that qualifies you for (other people have mentioned them on this thread). If you want a general purpose programming job, the main question hiring managers are going to be asking is “okay, they are smart and know a lot of theory, but can they actually churn out production-quality code day after day”. It’s hard to learn what “production quality” means without job experience, but a good first step is to just write a lot of code, and then go back and edit that code later (and learn why some code is easier to edit than others). The more languages + platforms, the better. Write a web app in ruby. Write an iPhone app. Write some games.

      • AspiringRationalist says:

        I was a bio major, switched to programming a couple years after college, and I am now a software engineer at Google.

        I would whole-brainedly second the Coursera recommendation. I took 2 CS-ish classes in college, and those plus an algorithms class (I took Princeton’s Algorithms 1 class on Coursera) were enough to get me my first software job.

        My main advice would be don’t feel intimidated by the long list of things people say you should do to get a programming job. The important thing is to write enough code that you feel comfortable when coding, and then start applying for jobs.

        If you don’t do well at first, then you can work on building up your resume or your interview skills. Just don’t assume it’s not worth tryingl yet just because you haven’t perfected your skills.

        • Pku says:

          Any suggestions on how to find places to apply for? (Aside from Google and Microsoft, all the places I’ve run across have had ridiculously specific prerequisites. I seem to be bad at finding places).
          Also, speaking of Coursera – I’ve done a bunch of MIT opencourseware courses and such, would it be worth writing that down on my resume?

          • Linch says:

            If you don’t mind living in the Midwest(Madison is quite different from other places in the Midwest), Epic hires a lot of general-purpose programmers out of undergrad/Master’s, and cares less than most other companies about prior job experiences.

            Upsides include decent pay (actually *very* competitive after you factor in the cost-of-living expenses), many opportunities for growth within the company, a great campus, and a positive, cooperative culture. There’s also an EA presence at Madison that seems to show very positive signs of growth.

            Biggest downside is that they use fairly archaic programming languages. Second downside is that Wisconsin winters are cold(I don’t care, but you might).

            Epic’s training is pretty in-depth and if you decide you want to move on to greener pastures, anecdotally I’ve heard that it’s easy to migrate elsewhere after a year or two of Epic on your resume.

            We have an internal referral system (I don’t actually think it affects hiring decisions, but it can’t hurt) so if you’re interested I can refer you or ask a developer to.

    • Richard says:

      Does it have to be programming, or are you interesting in testing? If the latter, you can simply sign up to uTest dot com and start finding bugs and earn small amounts of money from bug bounties.
      If you get good at finding the bugs, you earn ‘badges’ and because there is no college level education in testing, a good badge from uTest is quite often enough to get you hired as a tester.

      If you want to get good at testing, I would recommend starting by reading James Bach’s blog over at satisfice dot com

      Thanks for the offer of dragon, but I already got one and am not sure if I can fit another without extending the garage.
      Come to think of it, that is an excellent excuse for a DIY project, so yes please.

      • Anonononononymous says:

        My understanding is that the salary and earning potential of testers is a fair bit lower than that of equivalently skilled programmers; be warned.

        • OTOH Microsoft, at least, expects testers to also be programmers, so a very good tester with some programming skill can possibly get hired there, and then transition after some time to just being a regular (non-tester) programmer. Testers and engineers have exactly the same pay scale.

          • Eric Rall says:

            Microsoft recently abolished testers as a separate discipline. We’re all “Software Engineers” now.

            Before the merge, there was a reputation that it was harder for testers to get promoted than for devs because the middle-to-upper tiers of management valued development accomplishments more than testing accomplishments. This may have been true in some parts of the company, but I didn’t see much evidence for it in my personal experience.

        • AnObfuscator says:

          That depends. Manual testers, yes.

          However, I work currently as an SDET (after a few years of regular software engineering). I am paid like a regular software engineer, because I *am* a software engineer. My work is mostly building a sophisticated test harness for automated testing of a rather complex distributed system.

    • olivander says:

      If you want to know what kind of questions you’ll be asked in interviews, Cracking the Coding Interview is a good resource. You’ll definitely have to answer these types of questions if you interview with a big tech company like amazon, apple, google, or microsoft. Startups are a lot more variable, most do similar algorithmic questions but some want work samples instead. I have a github profile, but noone asked me questions about anything in it so I’m not sure how much it helps.

    • Emile says:

      I’d recommend working on open-source projects on github (or something like it), or making your own projects open-source on github.

      1) it will give you more contact with tech culture, geek norms etc.

      2) a well-filled github is a good sign when evaluating a candidate

      (source: software developer for about ten years, though my own github is pretty empty – I consider that a liability if I was to look for a new job …)

    • Forlorn Hopes says:

      With a math degree I’d consider looking into jobs in big data. Most big technology firms are now hiring data scientists under one name or another so if you have the maths skills for statistics it’s a good job market.

    • Shenpen says:

      I am not a big fan of the idea of programmers a separate class of people whose main expertise is that they can code. What worked in my career is being a domain (business admin) expert who can also code. Coding is about automating stuff. You better know the stuff you want to automate or else you rely on specifications, in which case you are easy to replace / outsource.

      So my opinion is, there should not be programmers i.e. people who only know how to automate stuff. There should be people who know stuff, and also know how to automate it.

      (People who call themselves only-programmers generally tend to know stuff like “computer stuff” i.e. hardware and communication standards, such as HTML, and then they automate that i.e. they automate how a HTML file should be presented on a screen in an app like Firefox. But, for example, our Scott here could learn about programming without caring anything about “computer stuff” and still be able to amend and customize the business logic layer of medical diagnostic software, because he is a domain expert of that. He need not care about the “computer stuff” much the same way when I program how to calculate manufacturing material requirements I can almost forget it is running on a computer, it is abstracted away from my business logic layer. )

      At any rate, know stuff, then learn to automate it. What is the stuff you know? Math? Let’s say, numerical analysis? You can automate that with R and with some Python stuff like Python Pandas, most likely in the Anaconda distro.

      • I used to work with some business types who’d switched to coding, and their code was bad. And no one could do anything, because they were the bosses.

        • Shenpen says:

          If they switched to coding when they were already bosses, that is too late. They should have majored in business and minored in programming.

          Also if they use a general purpose framework like .NET they will fuck it up. They should use special purpose frameworks that are harder to fuck up like ABAP.

      • Orphan Wilde says:

        The ability to turn specifications into automation seems trivial, until you deal with the actual specifications people actually produce. I’m currently writing business logic for medical diagnostic gear, by which I mean I’m writing this comment instead of doing that, largely because I’m waiting on a clarification. (My domain-level knowledge in this subject: Only above zero because I’ve done this so many times before, but entirely irrelevant to what I’m doing. Same with insurance, annuities, healthcare, human resources, finances, image processing, and whatever other domains I’ve worked with that I can’t recall right now) There is an art to turning poor or nonexistent specifications (I’ve never gotten good specifications, and I typically run through as many projects in a month as most busy people do in a year) into a working product, and it is hard to find people who can do it.

        On the other hand, I’ve worked with countless companies that keep hundreds of programmers employed doing largely-irrelevant busywork, such as adjusting the way the company webpage looks in Mozilla Firefox. These people are bodies, and are quite replaceable.

        Tangential aside: This is because of disturbingly-common incentives that reward managers for hiring more employees, resulting in ever-growing departments, which is offset in the private sector at least by periodic corporate restructuring/layoffs, and in the public sector by hiring freezes. I’ve seen companies outsource just so a manager could say they manage a hundred employees, thus securing a higher pay scale for themselves. What isn’t typically said about outsourcing incentives – if you’re hiring people to get nothing done, you might as well hire even more of them for the same amount of money to get nothing done, since you’ll get paid more in the process.

        • Shenpen says:

          But if you were a doc, you would need no specifications at all… wait, in this case it is wrong, medical stuff is too important to let doctor-programmers just wing it without writing exact documentation, at least after, which works a lot like a spec in hindsight. But there are many, many other fields where it is OK.

          Of course not in the kind of giant business that employs hundreds of programmers. That kind of giant business is a really weird thing to me, not only never worked at one, but rather never even seen one just heard their names like Apple or Microsoft. The average number of programmers at the firms I tend to work at is 1-2, their total number of employees maybe 100 and that is considered big. Often, their total number of employees is 20 and then I work for them through a consulting business which may have a whopping 2 employees, including me.

          Granted, German-speaking countries have a way of overdoing smallness largely because it is incredibly hard to find customers. So I see businesses with one owner and one employee specializing in three different services like business software, network security, and something else, because they cannot find enough customers.

          I think this way of all businesses being really big is some secret of the US that somehow the rest of us are unable to penetrate. Competitive pressure tends to make Euro businesses smaller, not bigger. Cost-cutting. I knew a business that was covering a the distribution of a major shoe brand in a country with only 10 employees. One guy left, took 70% of their customers, and has 1 employee beside himself. So just gets smaller and smaller.

          Needless to say, I hardly know the term manager and the incentives they face. I always, literally always get managed by the owner. Who has different incentives.

          I have no idea why competitive pressures did not kill managers yet in the US. Maybe there is some other, opposite force, such as having a huge unified market. Perhaps if it was really hard for a shoe distributor in Maine to distribute shoes outside Maine because of language etc. the same pressures would kill the managers just as well.

          Or maybe it is customers not being loyal. I have no idea really. Maybe in the US business grow large because there is none of this “I always bought from Joe so I will always buy from Joe no matter what price I could get from others, I don’t trust them” logic that defines yurop. Maybe a better functioning legal system makes switching easy and thus some corps grow big.

          • Corwin says:

            Two words : labor costs.

            A few more words : in Europe, an employee earns, net, about 1/4 to 1/3 of their total cost of employment.

      • youzicha says:

        Some subsequent studies failed to replicate the results in “The Camel has Two Humps”‘. Some attempted reproductions could not see any affect on learning outcomes. The paper itself was never formally published, and one of the authors have published a retraction: ” I need to make an explicit retraction of much of what it claimed. Dehnadi didn’t discover a programming aptitude test. He didn’t find a way of dividing programming sheep from non-programming goats.”

      • Jonathan Paulson says:

        This is a pretty common idea, but it works less well than one might expect: it turns out to be hard to translate ideas into code (in many cases, harder than learning the domain-specific knowledge).

      • Shenpen says:

        Here is my background profile, maybe you need to look for people like me:

        – BASIC at 8 years old, TurboPascal at 14 years old, the classic teenage hobby programming, making Nibbles clones

        – Go to business school

        – Start as a consultant of business software which means I can edit configurations like these group of customers sales will be posted to that G/L account, but NOT code

        – When clients downright refuse to accept how the software works, what can I do, desperately try to change it by coding, because I had literally no other choice: either quit and be unemployed with only 6 months job experience on my resume ( = suicide) or do what the clients demand i.e. change the code

        – Suck at it horribly, like, I was not even aware I need to read records from the database and write them back, I thought it is just IF-THEN…

        – Realize the best thing is to imitate closely the standard coding: this is a HUGE advantage for a beginner programmer, to not start from scratch but to customize an existing coding so he or she can see clear examples

        – Suffer like a pig on sleet ice working 14 hours a day until I saved that project from crashing and burning (so they merely hated to use it but did not refuse to use it)

        – Repeat it with 2-3 other clients

        – Slowly realize now I am good at it

    • Jeremy Jaffe says:

      Ok, so I too was a math grad student who wanted to get into software development.
      I attended one of the coding boot camps that teaches web development in 9 weeks.
      I attended Appacademy (
      but flatiron and dev boot camp are also good choices. They are hard to get into but not for someone with a math background – since they basically give you some programing challenges as part of your application which if your smart you should be able to do.
      After completing, getting a job was hard since I lacked any industry experience. I eventually got accepted to work at Epic ( ) (I start work in 2 months)
      Epic is a company that, for a variety of reasons, seems to prefer hiring people with Academic backgrounds. This shows itself in their application process – for example they ask for your transcripts as a standard part of the application which is a very rare thing to do.
      I would recommend applying to work their …
      Although I am biassed as I would like to have a bigger SSC community in Maddison Wissconsin

      • Anonymous says:

        Oh, boy. Epic. They have a… reputation. I sincerely hope you’re coming into this with full knowledge of that reputation. If not, then I strongly suggest you seriously research what’s been written about them.

        (Brr, MUMPS. I get chills even thinking about it.)

      • Alex Welk says:

        I’d second working for Epic but only if you’ve got a great work ethic and willing to work quite hard (for some great pay and benefits relative to the Madison area). My fiancé works there and I got an offer but declined in favor of something using my electrical engineering degree. I also wouldn’t mind seeing more SSC people in the Madison area.

      • Aegeus says:

        Hey, I work at Epic too! I started in June. So there’s at least 1 other SSC reader in Madison.

      • Scott Alexander says:

        Contact drethelin (uses that name on Twitter, Tumblr, and here) – he either is or was heavily involved in the Madison LW community and can introduce you to it.

    • Shmi Nux says:

      See what’s in demand and learn it, filling up the holes in general CS education along the way:

      If any of the technologies mentioned sparks your fancy, use it to start an open-source project on github. If you are good, odds are the jobs will find you.

      Re salaries, for those interested:

    • AnObfuscator says:

      Several pieces of advice:

      (1) Write code, and get feedback. I suggest Reddit’s DailyProgrammer. There are some brilliant solutions on there, and by submitting and reviewing the solutions of others, it will help you expand your ability to create code.

      (2) Learn frameworks. For better or worse, modern software engineering is less “come up with clever algorithms” and is more “string together frameworks to solve your problem”.

      (3) Learn how to engineer software. One of the major problems I see with people coming from other fields is that while they know how to “write code”, they don’t know how to build larger, complex systems well.

      To this end, I would suggest learning design patterns and unit testing. For design patterns, these are great books:
      Design Patterns: Elements of Reusable Object-Oriented Software — This is the bible. Also called “Gang of Four” or “GoF” design patterns. Also called “The blue and white design patterns book”. It’s a little dense, and a little dated, but it’s a classic.
      Head First Design Patterns — A little newer and more accessible than the bible.

      As I learned Agile software development and unit testing OTJ, I can’t really offer any good books on those.

      • brad says:

        I’d agree with this post. Not so much for getting a job in the first place, but for doing well after you do. Algorithms are important, math is important, being able to pick up syntax is nice, but end of the day you are building a product with other people and that means boring process issues are critical. Design too, but the industry is set up to mostly insulate you from that.

      • James Picone says:

        Don’t go overboard on the design patterns though. If you ever find yourself writing a factory-factory, consider re-evaluating your life choices.

      • Shenpen says:

        One big problem when people talk about programming is that they think their subset of programming is the whole of it. That is what made stuff like Joel Spolsky’s writings unreadable to me. GOF Design Patterns and everything you mentioned is largely about the category I would call “technical programming”. “software engineering” or “application programming” are probably good terms to describe it too.

        Technical programming is, by far, the most popular subset of programming so I do not begrudge you present it as if it was the only subset. But the important part is that this subset of programming is largely suitable for people whose main interest and expertise is computers or technology.

        If OP’s main interest is math, e.g. numerical analysis, then programming in R or Python Pandas in interactive notebooks like will probably very different from OOP design.

        Or for example, imagine a guy like me who is interested in business, not technology. I sit in the office and some purchaser asks me to update a thousand items with a program because he does not have the time to do it manually. Depending on which software we use I may whip up something like this: do you think it has much to do with OOP design?

        Of course, technology programming is by far the most popular kind of programming, in fact many people use programming jobs and tech jobs interchangingly. So it is OK you present it as first and foremost, but really tech jobs are people who are good in tech. Math people or business people or medical people need to learn entirely different kinds of programming. For example if your main expertise is insurance, you most likely will end up doing something like this, and good luck using the GOF there:

        And of course doctors get MUMPS but I won’t even link to that.

        • Adam says:

          Don’t forget scientific and embedded computing. One of my friends is a meteorologist who studies large storm cells and almost all of his daily work is programming simulations, but I don’t think he knows shit about object oriented design patterns, does everything in Fortran, and is selected mostly for his knowledge of physics and complex systems dynamics, not generalized programming skill. My wife does radar design, sometimes software, usually C or Fortran, but sometimes has to implement the tracking algorithms directly in hardware programming FPGAs. I don’t think she even knows a single object-oriented language and was selected for knowledge of digital logic, signal processing, and adaptive filtering, not generalized programming skill.

          I picked both of these because they’re examples of when having a very strong applied math background can give you a huge advantage over someone whose comparative strength is software design.

          As for R, oh man it’s basically the most convenient and easy to use interactive wrapper of BLAS and LAPACK libraries ever built. I start hating it when it comes time to extend it and find the copy semantics and ‘every bracket and operator is a function call setting up its own child environment’ makes it nearly impossible to do anything efficiently actually in R, the built-ins are so great because they’re written in C, not R, and now I have to learn how to use those stupid SEXP pointers that have virtually no documentation.

          • CatCube says:

            You’ve pointed out one of the things that bug me about professional programmers: sneering about FORTRAN. I get that the language doesn’t have a great computer-science theoretical model, but for those of us who just want our computer to do the arithmetic bitchwork for a scientific algorithm, FORTRAN works just fine.

          • AnObfuscator says:

            Indeed. Fortran is simple to learn and extremely fast for numerical algorithms. With the vast volume of well tested & performant existing scientific computing libraries out there, it makes little sense to abandon Fortran at this point.

            The language has also evolved considerably since it’s punch card days… I hear it even has OOP features now… 😉

            “I don’t know what the language of the year 2000 will look like, but I know it will be called Fortran.” — Tony Hoare

    • MaskedInMonotonicity says:

      Pku (and also Jesht, along with anyone else who may find this helpful):

      I was in a similar (or at least comparable) position last year: I had an undergraduate degree in (pure) Mathematics, followed by some interesting but unfulfilling work as a data scientist/statistician and looking to become a programmer/software engineer despite minimal training (one CS course in college, minimal (truly minimal, not false modesty) tinkering on my own).

      The software company I currently work at has an explicit and well-followed policy of hiring smart people who like to learn things (mostly) irrespective of what they already know. It’s worked out well for us so far. I’ve been here for about a year, and I’m having a fantastic time.

      The company is called Appian (yes, after the Roman road), and the culture is conducive to both happiness and learning (your mileage may vary, but low turnover rates, high ratings, and my own informal surveying suggest it is unlikely to vary by much). If you are open to moving to the DC Metro Area, I would highly recommend applying. Also, if you would like a foot-in-the-door referral, feel free to send either contact information or a resume to the burner address , which I will check daily until 9/15/2015, and I will respond from my work email to set you up with recruiting.

      Full disclosure: if you are referred through me, you are guaranteed to to receive a phone interview and not be screened out by our software. If you are hired after being referred through me and stay with the company, I will receive both a pat on the head and a small amount of money. Nevertheless, I fully stand by every claim made by me in this post.

      Additionally, I have heard not-great things about Epic. I am originally from the midwest, and the second-hand opinion I’ve garnered from the reports of friends and acquaintances in both Development and Testing (n = 11) is that the environment is both insular and isolating, the work-life balance is unfortunately lopsided, and the culture is akin to a pep rally. I interviewed there and was offered a position, which I declined because what I saw there reinforced these impressions (though, you know, something something confirmation bias). I would urge caution.

      As a more general aside, Cracking the Coding Interview is fantastic preparation for most software-engineering technical interviews. I would also brush up on combinatorics, which will be, like all well-applied mathematics, unreasonably effective.

      • Pku says:

        Thank for the offer!
        Do they do internships? I’m still about two years away from finishing my PhD, so I’m still about a year away from looking for a full-time job.

        • Chalid says:

          If you’ve still got a couple years, spend some time now looking at job listings that interest you. Look for skills that are particularly in demand and see if you can find an excuse to do them as part of your PhD work. This makes you more marketable, gives you something to talk about at interviews, and helps you figure out what you like.

          I’d also echo the advice upthread about looking at data science roles as being a perhaps better intellectual fit for a math PhD. (This varies by position, of course.)

        • MaskedInMonotonicity says:

          They do, but mostly for undergraduates. I know they tend not to for Masters candidates, and I really don’t know about doctoral students. Either way, I can make introductions with Recruiting (who are generally friendly people) who can start your process and put a pin in it, or at the very least walk you through the environment and make sure it’s somewhere you’d feel at home. I’ve made a few introductions for people (all but one of whom have ended up joining), and I know they’re perfectly happy giving someone the run-down and then marking their calendars for when to reach back out.

    • Moshe Zadka says:

      E-mail me at last-name <dot> first-name @ and I’ll be happy to talk you through some ideas.

      Source: I’ve been in the industry for >15 years, worked at companies like Facebook and VMware 🙂

    • Bugmaster says:

      I’ve interviewed quite a few people, and I can give you some advice on what not to do during interviews:

      * Know how to write a simple loop. Seriously. I have a feeling you’ve got this one, but still, you wouldn’t believe how many people fail.

      * Don’t ever say, “I refuse to answer such a low-level question”, especially when the question was, “write me a simple loop”. If you don’t know the answer to something, don’t try to bullshit your way out of it, just say so.

      * If you put an item on your resume that says something like, “STRONG mathematical skills, especially in the area of probability”, or “I am a BRILLIANT coder” (yes, in all caps, these are real quotes), then you’d better make sure that you are, in fact, STRONG and BRILLIANT. Your interviewer will not just take your word for it.

      * In terms of actual knowledge, make sure you know how all the common data structures and algorithms (e.g. lists, hashmaps, quicksort, Markov chains, whatever) work. Not just how to use them, but how they are implemented. This seems like a petty thing, maybe, but people will ask you questions about them. Also, if you don’t know how a tool works, you are very likely to shoot yourself in the foot with it one day.

      * Depending on your job, knowledge of parallel processing / multithreading may come in handy; if you are going to be involved in any kind of high-throughput computing or client-side UI, it’s practically a requirement.

    • Jonathan Paulson says:

      Different advice from others: Just apply for jobs (I guess summer internships?) at the places you want to work for. This is more important than taking classes, or doing open-source projects, or doing GitHub; it will be useful experience (both the experience of applying and actually doing the job is useful), and good for your resume.

      Either you will get interviews, or you won’t (I bet you will, because math is pretty close to CS; I know several friends who did math and college and are now software engineers). If you do, you will get to do real-world interview practice, and you will know that you can pass resume screens, and you might do a summer internship. You will definitely be in a better position to move into programming than you seem to be now.

      If you don’t get interviews, you will know your resume needs improvement, which probably means classes, side projects, etc. Ask your programmer friends (or the internet) to read your resume, and improve it.

      Programming hobbies are a great way to improve your skills and your resume. But the best way to get a tech job is to go out and ask for one, not procrastinate by learning more or building your resume.

      (Source: I did CS and school and have been in industry for a few years now)

    • James Picone says:

      Caveat: I’ve never worked in the US, and I tend to work on low-level C/C++y stuff, not fancy web things. Your mileage may vary.

      In my experience technical questions in interviews tend to be very simple. Here’s a selection of ones I’ve seen:
      – “What is polymorphism? When would you use it?”
      – “Can you talk a bit about the difference between static variables, local variables, and member variables, paying some attention to where they are in memory? What about local static variables?”
      – “Here’s some code. Can you point out some errors?” (The errors were referring to a variable with different names in two different places, a line of code being in the wrong place, and a private non-virtual member function that nothing called and so likely wasn’t called by anything ever).
      – A longish design thing where the interviewer had me explain how I’d design a logging system for dumping stuff to a file without blocking code logging stuff, and then said ‘okay now we’re logging into a database, how do you change the design?’.
      – ‘Explain the difference between a class and an instance’.

      One or two of these places had a actually-write-some-code step later in the interview process.

    • Brian Nachbar says:

      For what it’s worth, I came out of undergrad with a math major and a CS minor, and nothing outside of coursework to put on my resume, and A-list companies gave me interviews. One you’re in the interview your resume is all but forgotten (at least where I work now), and if you do well you can get hired.

      So once you get good at algorithm problems (many resources for this, I suggest old Google Code Jam problems), you may even be able to get interviews right away. Probably just make sure you have a programming language on your resume so they know you actually code.

    • Pku says:

      I’m somewhat overwhelmed at the level of help you’ve all shown here. It’s honestly beyond anything I expected, and I don’t know how to thank you all. It’s the first time I’ve ever seen a request for help on the internet with dozens of helpful responses and not a single trolling/douchey comment.

    • sweeneyrod says:

      I don’t know how valuable it is considered by employers, but I’ve found online freelance work incredibly easy to get. I’m still in school, so I’ve got no CS qualifications (other than a GCSE) but I’ve been inundated with more job offers than I have time to accept (although that might be because I set a fairly low salary). The work is mostly front-end web development.

  3. Newbie says:

    Scott, I’m curious on your thoughts on this piece, critical of the superintelligence aspect of AI risk:

    • Muga Sofer says:

      Definitely not even a little bit Scot, buut:

      >The extraordinary claim that machines can become so intelligent as to gain demonic powers requires extraordinary evidence …

      >By the 1980s, however, they discovered fundamental limitations that show that there will always be diminishing returns to additional processing power and data. Although these technical hurdles pose no barrier to the creation of human-level AI, they will likely forestall the sudden emergence of an unstoppable “superintelligence.”

      I … am incredibly curious to see the citation for this claim. (Surely no-one would say this based on nothing?)

      >AI technology we already have is poised to make threats like those posed by nuclear weapons even more pressing than they currently are.

      Wait, what? D:

      >>Bostrom’s descriptions of how machines might rapidly improve their intelligence make it clear that he does not appreciate that the knowledge possessed by reasoning programs is much more important than how those programs work

      So … this is wrong, right? I mean, it’s just … wrong. You can’t beat a powerful chess program with a terrible chess program that has a huge database of chess games.

      Ultimately, looks like a ton of interesting claims. I’d be really intrigued to see the arguments leading up to them.

      But instead, it comes across (to me) as more ike … “as everyone knows, superintelligence is impossible, so it’s kind of odd that all these experts and smart people inexplicably disagree with me on this obvious fact that everyone knows is true.”

      Also, um, I don’t think this person realises that “The likelihood of goal mutation is a showstopper for Bostrom’s preferred schemes to keep AI “friendly,”” is … kind of received wisdom in the field?

      I mean, most of MIRI’s work seems to be focused on goal stability in self-modifying tiling agents, but the author presents this as a huge original insight that upsets everything.

      • Murphy says:

        >So … this is wrong, right? I mean, it’s just … wrong. You can’t beat a powerful chess program with a terrible chess program that has a huge database of chess games.

        There’s actually a section that touches on this in Artificial Intelligence: A Modern Approach. It takes us into the sticky area of distinguishing behavior from intelligence.

        Imagine someone who isn’t very bright, IQ in the 80’s but does have a simple app on his phone.

        He fills in the state of the board and perhaps some information about his observations of the person he’s playing against (whether you look bored, happy, sad, tired, etc) and a moment later the app simply searches a big table with historical entries for the the experimentally identified optimal move to make against an opponent when the board is in that state and the opponent is tired.

        The app isn’t smart, it’s just a collection of knowledge/information and the player barely understands the moves in chess but they could still beat someone who is very good at chess who lacks access to the table of information.

        Is the person with the app actually more intelligent?

        The related section in Artificial Intelligence, A Modern Approach.

        “The Brain Prosthesis Experiment
        The Brain Prosthesis Experiment was touched on by Searle (1980), but is most commonly
        associated with the work of Hans Moravec (1988). It goes like this. Suppose we have developed
        neurophysiology to the point where the input/output behavior and connectivity of all the neurons
        in the brain are perfectly understood. Furthermore, suppose that we can build microscopic
        electronic devices that mimic this behavior and can be smoothly interfaced to neural tissue.
        Lastly, suppose that some miraculous surgical technique can replace individual neurons with
        the corresponding electronic devices without interrupting the operation of the brain as a whole.
        The experiment consists of gradually replacing all the neurons with electronic devices, and then
        reversing the process to return the subject to his or her normal biological state.”

        Patricia Churchland (1986) points out that the functionalist arguments that operate at the
        level of the neuron can also operate at the level of any larger functional unit—a clump of neurons,
        a mental module, a lobe, a hemisphere, or the whole brain. That means that if you accept that the
        brain prosthesis experiment shows that the replacement brain is conscious, then you should also
        believe that consciousness is maintained when the entire brain is replaced by a circuit that maps
        from inputs to outputs via a huge lookup table. This is disconcerting to many people (including
        Turing himself) who have the intuition that lookup tables are not conscious.

        • Anonymous says:

          they could still beat someone who is very good at chess who lacks access to the table of information.

          If I’m sitting across the board, I don’t need access to the table of information. I just need to know that they’re relying on the table of information. If so, my first ten moves are going to be utterly absurd. I won’t throw away anything too major, but I will definitely get the game outside of whatever table you’re using. There is a finite number of possible positions, but we’re well beyond “atoms in the universe” land. My intuition is that ten moves will be enough, but worst case scenario, I just continue with absurd moves until my opponent looks at his phone, gets a confused look on his face, and then takes a good ten minute think.

          The only possible issue is that I have to absolutely crush him in the middlegame. I can’t let enough material be traded so that we pop back into an endgame database with him having even the slightest advantage. Endgame databases will also be skewed toward close games, so if I can turn the tide and get a 3 or 5pt lead before too much material is gone, he’ll have no chance of recovery.

        • 4bpp says:

          This reminds me of an argument (possibly made by the other Scott A somewhere, but I can’t find a citation) that the reason most people intuitively say that they do know how to add (500-digit number) and (500-digit number), but don’t know how to factor (500-digit number), even though in both cases they can describe a terminating algorithm that will eventually produce a result, is that only in the former case is the algorithm “efficient”, in the sense of polynomial-time or something close to it.

          By analogy, one could argue that a brain with neurons partially or completely replaced with electronic ones is conscious not because it exhibits behaviour indistinguishable from a meat brain which is axiomatically conscious, but because it does so within a polynomial multiple of the space taken up by the original. A look-up table brain is no longer to be considered conscious because even with simplifying assumptions, it will at least take up exponentially more space.

      • Harald K says:

        “You can’t beat a powerful chess program with a terrible chess program that has a huge database of chess games.”

        Yet. But in other traditional AI domains, big data (huge database) and simple algorithms (“terrible program”) win out. I won’t rate it as impossible that machine learning approaches eventually become essential in these sort of games too.

        • Newbie says:

          That’s my impression as well. A lookup table or even a set of heuristic rules represents the result of an enormous amount of precomputation, experimentation, etc. Given a choice between betting on the smartest person in the world who’s just been told the rules of chess and the hardest chess ai that can run on your phone that uses a set of heuristics and lookups, I’d assume the ai would win because even if the smartest person in the world does theoretically perfect processing, she’s unlikely to quickly develop the level of domain knowledge the ai is pre-built with.

          The general point is that having intelligence in the human sense of having the ability to learn can frequently lose to intelligence in the sense of having far more data about the problem.

      • Adam says:

        I don’t know if it’s wrong specific to chess programs, but most of the high profile advances in intelligent devices and program behavior from the last few years has been driven by access to larger data sets, not by better algorithms. That’s really Google’s comparative advantage. Most of the cleverness has come in the form of innovations in dealing with large data and managing a distributed workflow.

        As for actual chess, Deep Blue itself wasn’t doing anything more intelligent than prior attempts at chess-playing programs. It just ran on specialized hardware that allowed it to brute force larger search spaces. We could probably make huge generalized search gains implementing databases in TCAM, but the cost and energy usage compared to disk storage is prohibitive.

        That’s probably what he means about diminishing returns to scale, but I don’t know. His point seems to be largely that advances in computational intelligence require access to and processing of increasingly larger data, which is a difficult engineering problem involving lots of energy and physical infrastructure, rather than just re-writing code, making the idea of a near-infinite slope recursive self-improvement loop unlikely if not impossible.

      • Nornagest says:

        I … am incredibly curious to see the citation for this claim. (Surely no-one would say this based on nothing?)

        It sounds like they’re talking about Amdahl’s law and other facts in that general space. Which has profound design implications and rules out some naive approaches to… well, most things, really. But it’s hardly a knock-down argument.

      • Jonathan Paulson says:

        So … this is wrong, right? I mean, it’s just … wrong. You can’t beat a powerful chess program with a terrible chess program that has a huge database of chess games.

        A lot of the recent success of machine learning (e.g. Google Translate) has been applying very simple algorithms to massive amounts of data (fancy algorithms take too long when you try to apply them to massive amounts of data).

        • Raph L says:

          It’s interesting, fashions change. A few years ago, a lot of the advances in translation quality was simply the application of relatively simple statistical techniques to ungodly amounts of text. But these days deep neural learning is giving statistical techniques a serious run for their money, largely because so-called “embedding” can do a good job filling in the gaps where there is no exact match for the phrase in the input data. Source: Jeff Dean’s CIKM keynote.

      • Loquat says:

        >>Bostrom’s descriptions of how machines might rapidly improve their intelligence make it clear that he does not appreciate that the knowledge possessed by reasoning programs is much more important than how those programs work

        >So … this is wrong, right? I mean, it’s just … wrong. You can’t beat a powerful chess program with a terrible chess program that has a huge database of chess games.

        You can if the powerful program has basically no experience playing chess. No matter how smart you are, if you don’t know very much about the world you’re going to have a hard time outsmarting less intelligent people with vastly greater experience.

        For example: in a previous AI risk thread, a poster claimed that obviously a superintelligent AI would be able to design super-robots and produce them in sufficient numbers to conquer humanity. But leaving aside the question of how the AI would get control of adequate manufacturing facilities, how is the AI going to suddenly design super-robots if it didn’t know anything about robot design to begin with? Humans have been working on basic autonomous robots for some time, and we’re still having a few issues. Logically, even the smartest AI would have to go through several rounds of trial and error to work out all the flaws – and don’t just say it’d do that in simulations, nothing short of a literal deity is ever going to be smart enough to think of every possible monkey wrench the real world can throw at you.

        • Pku says:

          OTOH, the smart AI might be aware of that and plan its attempts ahead in a way that would be non-catastrophic for it if the first few attempts failed.

  4. Richard says:

    I was reading the discussion on vegetarianism in the comments on the last post when the RSS feed told me it was a new open thread, so I’ll keep it going here.

    “don’t eat meat because you’re an enlightened moral being who cares for the welfare of all living things to me”.

    I simply don’t get the logic here, sorry.

    First, let’s change ‘welfare of all living things’ to ‘welfare of animals’. Plants are living things, and with the exception of edible fruits, I don’t think they benefit much from being eaten.

    Presumably, the animal being eaten is already dead, so that eating it does not affect it’s welfare. What does affect an animals welfare is how happy it was while alive and how painlessly and quickly it was killed.

    Thus, if you care about animal welfare, the object must be to maximise the number and happiness of live animals. The way to do this is NOT to just leave them alone. You only need to spend one summer removing rotting seal carcasses off the beaches in order to start questioning the wisdom of banning seal hunting. Nature is notoriously bad at managing populations and incredibly good at imposing suffering in the form of starvation or nasty predation methods. “nature red in tooth and claw” and all that.

    The key to maximising animal welfare is careful population control and increasing the carrying capacity of the land through responsible farming and wildlife management. Said population control will involve killing animals because getting deer to go on the pill is unfeasible.

    I see nothing morally wrong with eating the flesh of contented animals after they are dead and I suspect that those who do need to spend more time around animals both wild and responsibly farmed ones, then make up their mind which ones are better off.

    Some time ago (~six months?) I wrote a post in a thread on here about responsible farming, but I couldn’t find it just now, it explains my position in more detail.

    edit: it was here

    • Froolow says:

      The point you try to prove is, “If a thing is dead, it does not care about being eaten”, which I agree with. But your argument is *actually* “If a thing is dead there is no reason not to eat it”. Which I don’t agree with; one trivial case where I don’t agree with it is eating the bodies of dead humans (although it doesn’t upset the dead human, it might upset their relatives, or society more generally). A more weighty case where I disagree with it is in the intentional killing of a sentient thing to eat it (arguendo animals are sentient, or might be) – even if society was ok with eating dead humans I would object to killing a human for its meat; humans have projects which are irrevocably interrupted by their death, and I think animals do to (arguendo again – not the key point here)

      Even though I think your argument might have the above flaw, I think it is probably actually true in principle – there is some level of guaranteed comfort in my life at which I would accept being killed at 60 for food, and similarly I can imagine making a win-win deal with proto-cows (or rather, their genes) where their genes would prefer to become completely ubiquitous worldwide at the expense of us eating them. You argue that animals are better off in humane farms than the wild, and I (mostly) agree with you.

      However I think the real damage of this argument is that it ‘proves too much’. If you think animals are the sorts of things to which rights should apply – even if that right is only, “Maximise total animal happiness” – then you’re pretty much committed to agreeing that most industrial animal farming is in the same sort of moral basket as slavery. That has implications for behaviour which non-vegetarians rarely follow through on.

      • Richard says:

        My argument is actually: “If a thing is dead, there is no _moral_ reason not to eat it.” There may be others, like kuru or societal norms.
        With regards to animals having interruptible projects, my claim is that a managed population results in more animals living longer and with more resources to spare for activities other than searching for something to eat, which would benefit such projects.

        I would not claim that factory farming is equal to slavery, but it’s certainly on the same continuum somewhere.
        Consequently, I do get >90% of my meat from animals I have reasonable evidence were content while alive. (When eating out or when abroad, I have little chance to check beyond insisting on high quality meat. I go for wild-caught fish or venison when possible. Also, lean meat is just about always happier than fatty meat provided we’re talking the same species)

        • Zykrom says:

          If killing animals is bad, paying to have them killed is also bad, and if that’s bad then encouraging others to pay to have them killed is bad.

          Most of the time when you eat an animal, you’re doing one of the above.

          If you’re a human and want to commit to not doing those things, it makes sense to cultivate a disgust reaction to eating animals.

    • Froolow says:

      Forking because it’s a different topic:

      It has occurred to me that if effective altruists (and the general rationalist sphere) take their claims seriously, the answer to Richard’s question is WAY more important than anything else they could be thinking about.

      If a chicken life is worth even a fraction of a human life, switching from eating factory chicken to humane chicken (if Richard is right), or factory chicken to no chicken (if the general vegetarian argument is right) might be one of the most cost-effective interventions you can perform to generate QALYs for sentient creatures – especially since going vegetarian is probably cost-saving in the long run.

      To put it another way, rationalists are happy to change their behaviour based on a high-payoff low-probability outcome like cryonics. But many seem unwilling to change their behaviour based on a less counter-cultural high-payoff low-probability outcome – namely that, “Animals have sentience equivalent enough to human sentience that their welfare matters more than your desire to eat their meat”.

      Why is this?

      • Jiro says:

        Because the entire argument relies on the fact that humans are not good in formalizing their beliefs as numbers and writing them down.

        If a chicken life is worth a small enough fraction of a human life, then saving chickens wouldn’t be worth it. It’s just that people when asked about what fraction sounds plausible will write down ones that look small but are much larger than they actually believe in. Nobody writes down 0.0000001, and even that may be too high–who would sacrifice a human to save 10,000,001 chickens?

        • Froolow says:

          People are very bad at formalising their beliefs (especially regarding very small numbers), but it seems reasonable to use revealed preferences to elicit the same sort of information.

          The CDC reckon around US 150 people are killed in a collision with a motor vehicle of some sort each year. Say 50% of these collisions are people crossing the street. There are about 350m people in the US who – we will say – cross an average of five roads a day every day. This makes the risk of crossing the road in the US 1 in about 8 billion.

          I would absolutely cross the road to save a chicken that was suffering, suggesting I would roughly prefer a human to die than 8 billion chickens to suffer.

          I’d probably actually take a much greater risk TBH, but I like animals an unusual amount whereas I think most humans would think you almost have an obligation to cross the road to help a chicken in distress.

          • Jiro says:

            Most humans who are not EAs will treat differently saving a life of someone who is in front of him (and to some extent any concrete example) and saving a life in the abstract or on a statistical basis. I can’t speak for you personally, but insofar as your behavior in this example generalizes to others, your example is measuring these factors, not the relative valuation of humans and chickens. A proper example would have to be neutral on these factors. One building contains a human and another contains 8 billion and one chickens. There are gas leaks which will make the buildings explode and you only have time to find and turn off one gas valve. Ignore any factors that make the chickens useful to humans (this many chickens could stave off a lot of starvation, for instance.) Do you save the human or the 8,000,000,001 chickens? Everyone’s going to save the human.

          • Froolow says:

            I agree people act differently when faced with concrete examples of suffering rather than abstract statistical examples, but I think meat is closer to a concrete example than an abstract one; you actually physically handle the dead body of an animal to prepare and eat it.

            Also I’m not sure people would save a human over eight billion chickens. Chickens are maybe just on the cutoff of ‘no morally important feelings’, but certainly with higher animals like dogs, cows or pigs I think people would pick the animals over the human at way less than a few billion animals. Substitute great apes or dolphins for the dogs / cows / pigs and I think you might not even get to triple figures of animals before people start switching – certainly people choose to donate to animal charities vs other charities at rates that suggest animals rank more highly than humans in the abstract for at least some people.

            But since we can’t perform those sort of neutral experiments, I reckon asking whether people would do X risky thing to save a chicken is a reasonable way of estimating the correct order of magnitude. You maybe don’t even need a real chicken suffering – just ask, “Would you do X if I (credibly) promised to go and torture a chicken if you didn’t?”

          • Jiro says:

            I don’t think meat is that close to a concrete example, but even assuming it is, then you’re just using one concrete versus abstract example to make statements about another concrete versus abstract example. However, you’re using it to make a statement about the wrong part of the example–you’re using it to make a statement about the relative valuation of humans and chickens–a non-(concrete-versus-abstract) factor. You can’t do that.

            As for animal charities, most people donate to charities for reasons only remotely related to how much they actually value the benefit to the recipients of the charities.

            But since we can’t perform those sort of neutral experiments, I reckon asking whether people would do X risky thing to save a chicken is a reasonable way of estimating the correct order of magnitude.

            No, it isn’t. That experiment measures people’s inability to compare concrete risks to non-concrete or statistical risks, not their relative valuation of humans and chickens. Just because you have nothing better than a bad experiment doesn’t mean it’s okay to use the result of the bad experiment.

          • Froolow says:

            Actually, sorry, I was wrong in my last post. You write:

            > Most humans who are not EAs will treat differently saving a life of someone who is in front of him … and saving a life in the abstract or on a statistical basis.

            But I wasn’t talking about most humans, I was talking about EAs (and rationalists) alone – arguing that EAs (and rationalists) spend a lot of time thinking about stuff which is potentially less important than working out their human-animal tradeoff ratio for common food animals.

            Sorry I didn’t spot that earlier, although I did enjoy the digression the conversation took, so not wasted time at all in my opinion.

          • Baby Beluga says:

            @Jiro: I’d save the 8,000,000,000 chickens. And I don’t think that’s even that weird! 8,000,000,000 is a huge number.

            To give you a sense: don’t think about a building filled with 8,000,000,000 chickens, because when you think about that, you think of “a building filled with chickens,” and buildings aren’t all that big. But this would have to be a huge building! Think of it this way instead. Every minute for the rest of your life, you watch 250 chickens die, which is enough chickens to entirely cover the floor of the room you’re currently sitting in and then some.

            Wouldn’t you rather watch a single human die than watch 250 chickens die, every minute for the rest of your life? I would.

            Also, for what it’s worth, I think my break-even ratio for killing animals:killing people is very different from my ratio of animal suffering:human suffering. The first ratio is big (but certainly smaller than 8,000,000,000), because animal lives don’t seem all that fun or interesting, and are much shorter. The second ratio is barely more than 1, because I’d guess the experience of pain and suffering isn’t something that has changed much from animals to us.

          • Jiro says:

            8 billion is the cube of 2000. I think it is possible to have a building sized roughly as a cube 2000 chickens on a side.

            Wouldn’t you rather watch a single human die than watch 250 chickens die, every minute for the rest of your life? I would.

            I wouldn’t want to watch any arbitrary thing every minute for the rest of my life. My disinclination to do it would be because of that, not because of the moral value of the chickens.

          • Hedonic Treader says:


            You maybe don’t even need a real chicken suffering – just ask, “Would you do X if I (credibly) promised to go and torture a chicken if you didn’t?”

            This confounds the direct harm with the indirect harm of rewarding torture blackmailers.

          • AlphaGamma says:

            Is this post intended to be an extended “why did the chicken cross the road?” joke?

          • Whatever Happened to Anonymous says:

            I’d save the chickens, think of all the humans you could feed with that amount.

          • Jiro says:

            I’d save the chickens, think of all the humans you could feed with that amount.

            What? I said:

            Ignore any factors that make the chickens useful to humans (this many chickens could stave off a lot of starvation, for instance.)

          • Whatever Happened to Anonymous says:

            I missed that specification, apologies.

          • Deiseach says:

            Wouldn’t you rather watch a single human die than watch 250 chickens die, every minute for the rest of your life? I would.

            Thing is, if I knew I was going to be seeing 250 chickens per minute die for the next thirty years, I imagine the first ten minutes would be horrific, but after half an hour or so I’d be desensitised, so that after about a week it’d be “Goddamn chickens, die faster!”

            Whereas watching one human die, and the knowledge I let him die in preference to saving chickens, would cause me much more mental anguish.

          • Anonymous says:


            Such a building does not exist and would be incredibly difficult to create. Consider: the largest building (by volume) in the world is the Boeing Everett Factory (, with a volume of 472mil ft^3. If we pack chickens at 1/ft^3, we would still have over 7.5bil chickens to place somewhere. In fact, the top 11 largest buildings in the world would not even hold 2 bil chickens, and this is assuming the buildings are empty inside, save for chickens. Even with 2/ft^3 we still are drastically short of places to shove chickens.
            I suppose we could fit 8 or 9 chicks per ft^3, in which case we could potentially fill the Boeing Factory.

      • Tom says:

        I believe I read a post of Scott’s on exactly this topic. His premise was essentially that depending on the value you assign to animal life, the value of all animals is either insignificant or completely eclipses human life.

        • Froolow says:

          You’re quite right

          I don’t remember reading this, but now I’m really worried that I did, forgot that I’d read it, then re-described it to this thread thinking, “I bet this is exactly the sort of thing SSC audience would be interested it”.

        • Adam says:

          There’s something viscerally unappealing about that argument. It’s like you’re effectively saying if I could figure out a way to breed a sufficiently large number of termites that each have non-zero value, I could eventually breed enough that my termite colony matters more than all other life in the universe.

          • Jiro says:

            What’s viscerally unappealing about that argument is that

            1) people’s ideas about animal suffering aren’t very consistent and that argument rubs your face in the inconsistency

            2) people may say they are concerned for animal suffering, but they are actually concerned about self-righteousness, purity, and/or status grabbing and the conclusion that animals are either supremely important or unimportant is not good for any of those.

            If termites have a non-zero value and you could breed an unlimited number of them, you really could breed enough that they matter more than anything else. I choose to bite the bullet and say that termites have a value that approaches zero and that animals are unimportant.

      • Doctor Mist says:

        To put it another way, rationalists are happy to change their behaviour based on a high-payoff low-probability outcome like cryonics.

        You’re confusing “is” and “ought”.

        Whether freezing my head will let me live 800 years is an empirical question which will, eventually, admit a factual answer. Whether a chicken is worth 1/4 or 1/100 or 10^-23 of a human is not.

      • Anon says:

        > To put it another way, rationalists are happy to change their behaviour based on a high-payoff low-probability outcome like cryonics.

        Er. There are more veg*n LessWrongers than there are LessWrongers signed up for cryonics.

      • Albatross says:

        Aside from factory vs humane farming vs vegetarianism I think the most aspect is hunting.

        Any farm is kinda like a prison, while hunting is an animal growing large in its natural habitat and then being very quickly killed. Vegetarianism leads to vegetable farming and the destruction of natural animal habitats, while hunting is responsible for dedicating more land to wilderness.

        Hunting is basically maximum number of animals with maximum natural happiness while alive. The increase in natural habitat would also be a positive good in its own right.

        • houseboatonstyx says:

          This. On almost all points.

          In (eg) the US, the hunters’ political influence plus the environmentalists’ influence is stronger than the developers’ influence, preserving habitats. In the Commons (ie the ocean), un-regulatable fishing may cause extinction of the target and damages the system for all creatures.

    • anonymous vegan says:

      “First, let’s change ‘welfare of all living things’ to ‘welfare of animals’. Plants are living things, and with the exception of edible fruits, I don’t think they benefit much from being eaten.”
      You can’t abstain from eating at all, and fruit based diets are extremely difficult to follow. You harm living things less by eating plants, than by eating animals. How do I know this? One: I think that all morally reasonable people will agree that the moral weight of animals is immensely greater than that of plants, which are inferior creatures. And two: the animals that you eat, themselves consume a huge amount of plants as they grow up, therefore, a diet that includes meat and dairy will inconvenience plants more, not less.
      “Presumably, the animal being eaten is already dead, so that eating it does not affect it’s welfare. What does affect an animals welfare is how happy it was while alive and how painlessly and quickly it was killed.”
      Therefore if I kill a human being who has no friends, as long as it’s painless and nobody learns about it, I didn’t harm anyone’s welfare? I think not having their life cut off should be considered part of a creature’s welfare and I think most people would intuitively agree.
      As for the idea that more should be done to improve the life of animals – that doesn’t refute veganism.
      Isn’t the best way to treat a pig – best for the pig – to take care of it WITHOUT killing it for food at all, at least until it dies of old age, having enjoyed all of its natural lifespan? This is certainly what the pig would prefer if it were intelligent. Such a policy would be incompatible with large scale meat consumption.
      Anyhow, kudos to your farm for treating its animals well, and if you, Richard, go around telling people to only eat meat and dairy that come from ethical farming, then as a person concerned with animal welfare I feel grateful to you and see you as a valuable ally. But if you argue frequently against veganism (and I don’t know if this is the case of course), then the good you do to animals might be undone, because then you risk encouraging people to eat, not ethically raised animal, but horribly factory farmed animals. And the lives of factory farmed animals are much worse than those of wild animals.

      • Jiro says:

        the animals that you eat, themselves consume a huge amount of plants as they grow up, therefore, a diet that includes meat and dairy will inconvenience plants more, not less.

        That would seem to imply that not only should you not eat animals, you should kill as many as possible to avoid harm to plants.

        • anonymous vegan says:

          I also said that the moral weight of animals is immensely greater than that of plants.
          I also said that killing is bad.

      • Richard says:

        If we are talking about a single pig, then yes, not killing it is to be preferred. If we are talking about all pigs, the number of pig-QUALYs is increased by maximising not only the happiness of each pig but also the number of pigs. Due to population mechanics, this requires killing quite a few of them.

        This is a lot more evident when we’re talking wild animals, which is really the most realistic alternative as you wouldn’t get a lot of farm animals without a market for the product. My dogs and I spend enough time searching for diseased and dying wildlife that convincing me they would be better off without hunting is a bit of an uphill battle. At least around here, the problem is not enough hunting, and experiencing a single scabies epidemic up close should be enough to convince anyone. (I realise that the situation for the rhino is rather different)

        I argue a lot more frequently against factory farming than I do against veganism, so I suspect we’re mostly on the same side there. My claim is: raising animals well > not raising them at all > raising them badly. The same goes for pets of course, which is why I prefer rescue dogs.

        • houseboatonstyx says:

          Content note: singular used as generic.

          > My dogs and I spend enough time searching for diseased and dying wildlife that convincing me they would be better off without hunting is a bit of an uphill battle. At least around here, the problem is not enough hunting [….]

          Also arguing within your same side here, what about not enough predators? My take has always been, that serious injury or sickness for a wild animal only lasts till a predator finds it (perhaps an hour or two, positively correlated with the size of the animal).

          So am I wrong, or is your area perhaps lacking enough wild canines (etc), such that your rescue dogs can find the easy prey before they do?

          This isn’t a practical suggestion of re-introducing wolves or whatever to an area like yours; I just want to check my own assumption about what happens in the more-nearly-wild. I would be surprised (perhaps wrongly) if even an adult deer immobilized by injury, lasted very long if there were even a pack of coyotes around.

          • Richard says:

            The thing with predation is that it works rather badly. The classic example is lemmings:

            1: There are a few lemmings and low predation.
            2: The lemmings multiply and provide food for more predators
            3: The lemmings multiply faster than the predators and starve or die of epidemics
            4: The predators are now far too many and will also starve/fall ill
            5: Repeat from 1

            The lemmings example is just the most dramatic because of numbers. Around here, we got the same cycle with hares/foxes and deer/wolves unless people step in and manage both populations.

          • houseboatonstyx says:

            @ Richard

            So at Stage 3 you are finding starving hares, and at Stage 4 you are finding starving foxes?

            Content warning: theoretical nonsense below.

            A while back in this forum, some posters were theorizing that all wild animals world-wide should be exterminated, and all wilderness itself paved, because their lives were almost all suffering — illustrated by images of all sick or injured prey animals suffering for what I consider an unrealistic length of time.

            Thank you very much for all your good information about current reality. If you haven’t closed the window on me already, I suggest you do so now. 😉 I don’t want to draw you off into nonsensical world-wide forever speculation, but I will sketch out what I’d reply to your point.

            In a wide, healthy, bio-diverse wilderness, different animal populations rise and fall for many reasons. In a large elastic system, when one group is too small, another can expand to fill its niche (or perhaps move in overnight). Or one or both can move to a better area (if there is one).

            I’m not disputing that the cycle you describe does happen in parts of even an ideal and large wilderness. But there, over-population of prey = Stage 3 = abundance of food for neighboring predator species, seems unlikely to last very long. Stage 4 = starving predators would be different, and how long that would last might depend on how high up the food chain that predator is.)

        • Baby Beluga says:

          I agree with you that raising animals well > not raising them at all > raising them badly. However, that’s a reason why veganism isn’t always correct, but not a reason why veganism isn’t correct in the 21st century US.

          Like, if you walk into a Chipotle and buy a chicken burrito or whatever, you are creating demand for raising animals badly, not for raising them well. If we lived in a society where all animals were treated wonderfully, factory farms didn’t exist, and meat was several times more expensive as a result, it’d be a different story.

          • Richard says:

            What I advocate is creating a demand for ethical meat by being willing to pay the premium for high quality meat.
            The easiest way to do this is simply demanding low fat meat because it is quite tricky to raise lean and unhappy animals profitably.
            The fat content of free-range contented pigs is around 10% of the factory farmed ones.

            One interesting tidbit is that when NZ removed farm subsidies, it forced the farmers to move from factory farming to free-range in large numbers because grain based feed was no longer profitable.
            Thus if you find meat from NZ that is labeled ‘grass fed’ in your local supermarket, your odds of getting the flesh of contented animals are quite good.

          • Hedonic Treader says:

            Richard, humane standards require more trust than avoiding chicken. Perhaps if there were really trustworthy independent audits or something.

            I will say, however, that funding chickens would still be suboptimal in this situation. You could save the money and spend it on more efficient pleasure generation. The most efficient way would probably be a charity that promotes and funds hedonium, i.e. future technology specifically optimized for pleasure-per-input.

          • Richard says:

            @Hedonic Treader
            I honestly didn’t get what you are trying to say there. Would you mind rephrasing?

          • Hedonic Treader says:

            If I understand you correctly, you are advocating that it is more ethical to eat high-cost humane meat than no meat because this will allow more animals (e.g. chickens) to exist and have good lives.

            However, this is basically an existence donation. If you are willing to give existence donations to chickens, you should be even more willing to give existence donations to sentient entities that experience more positive wellbeing per dollar than chickens. The optimal path for such altruists should therefore be to fund the invention and existence donations of artificial entities specifically optimized for pleasure-per-dollar/joule. Otherwise it’s an opportunity cost.

            Sorry if that wasn’t clear.

          • Richard says:

            @Hedonic Treader
            Ah, that made more sense, thanks and I’m willing to accept that argument.

            Two things though:
            Many people will eat meat. I am not convinced that the negative utility generated by eating factory farmed meat < the utility gained by the donation from the premium you would have been willing to pay for better quality and ethical farming.
            Happy animals == tasty animals. The premium I pay for the flesh of contented beasts gives me an immediate and non-zero benefit. (and probable long-term health benefits) A pure donation does not give me such.

          • Hedonic Treader says:

            Richard, yes, those people who will not forgo meat (and who refuse faux meat of all forms) should probably gravitate toward high humane standards. I’m not convinced happy animals == tasty animals, but I’m not a gourmet myself. 🙂

            One last aspect to mention, meat affects wild animal populations and their suffering/happiness, e.g. compare the discussions in the following links if you don’t know them yet (I mostly mentioned chickens because they seem to displace fewer wild animals):


  5. olivander says:

    Are there any postmortems of Metamed? It seems like an important example of a startup that tried to apply less-wrong style rationality, but its homepage is now dead and I haven’t even found the normal ‘startup shuts down’ techcrunch post.

      • AlphaCeph says:

        I totally expected something like this to happen. Metamed is a great idea, assuming that a lot of its target audience are A-grade rationalists who would prefer a piece of great, actionable, research-backed advice scrawled on the back of a dirty napkin to a bunch of useless platitudes in a nice brochure.

        So basically, if Vulcans had health problems, then MetaMed would be in the money.

        Unfortunately, they tried to sell to Earthlings…

        • Vaniver says:

          I totally expected something like this to happen.

          I recommend predicting this ahead of time next time; you can then win many more Bayes points.

          • suntzuanime says:

            It’s often considered rude to tell someone they’re totally going to fail until they’ve already done so.

          • rsaarelm says:

            It’s often considered rude to tell someone they’re totally going to fail until they’ve already done so.

            This is why you publish predictions you don’t want publicly legible just yet as SHA1sums.

          • Vaniver says:

            @suntzuanime: If you can visualize the failure mode, this is very useful information to people before that mode. “I don’t think you will find a market for the substance, you need to sell the symbol also” helps them see that failure mode, come up tests, and pivot sooner, whereas simply saying “I don’t think it will work” does not help in the same way. (It only provides a bit of outside view information.)

        • Jiro says:

          The fact that someone created glossy brochures is legitimately useful as a signal. It shows that you spent significant money and effort on the project. Of course, this doesn’t actually prove that you spent anything at all on the project itself, but it’s still useful to weed out projects on which people are not willing to spend money or effort at all; unwillingness to do so is correlated with the project being bad, even though it does not *prove* the project is bad, and a proper Bayseian customer would recognize this.

          You can’t just say “signalling is boring and expensive, so I’m not going to do it”.

        • vV_Vv says:

          The world is too stupid to understand me – said every sore loser ever.

          MetaMed could never provide evidence that they overperformed their competition: concierge doctors.

          Doctors have licences that certify that other doctors think that they know their sh*t, they have a government-backed oligopoly to give “Medical Advice” and “Prescribe Treatment” (which MetaMed couldn’t do), and they have established reputations.

          MetaMed was just a bunch of smart guys with more or less high status in the “rationalist” commuity and little or no track record of domain expertise. The closest thing to an expert they had was probably Scott, who bailed out early and went to work as a traditional doctor.

          MetaMed business model was based on EY’s f*ck-the-experts-I-have-superpowers school of “rationality” that we were discussing in the previous post. They tried to market it on LessWrong and the other “rationalists” venues to what they thought was a receptive audience.
          People did not buy it. They crashed and burned. Period.

        • houseboatonstyx says:

          Metamed is a great idea — I agree. Needs a rational customer — I disagree. If you’re selling directly to patients, then all it needs is some rich, desperate patients.

          When told there is no cure, some people will try anything they can afford. Find some expensive, last-resort treatments or products, target their vendors’ customers. Study how those vendors are doing their marketing — how they reach rich people who would be turned off by dodgy ads.

          And/or, depending on your actual costs for doing the research, find some expensive professionals in medical/healing work, offer them a free sample: “What are some problems your clients are desperate to solve?” Research one or more of those problems, give the professional a free report, perhaps asking zim to do a beta test with some of zis clients. If your report proves helpful, then that will spread through zis grapevine. Desperate people want friends to try the same thing that helps them.

      • Adam says:

        I hadn’t heard of MetaMed until right now, but my own frustration with healthcare right now has nothing to do with access to information. I basically have a very good idea of what is required to change my health for the better, but it requires surgery and finding a surgeon who would take my insurance was the bottleneck, to the point that I needed to change insurance providers.

        Of course, one data point and I have no idea what the relative prevalence is of problems solvable by information versus problems solvable by access to treatment.

      • Deiseach says:

        Reading the linked post makes it sound like they fell into the EA scrupulosity trap: every cent, every minute, not devoted to the most efficient method of saving the largest number means YOU are personally responsible for the deaths of millions! So we must spend as little money on non-essentials as possible, and marketing is a non-essential since the service we are providing is so obviously a good thing, it doesn’t need any help to sell itself!

        They forgot they were a business and needed the services of marketing and sales (and I never thought I’d be sticking up for marketing, but I know I’m hopeless at selling and even I recognise you need to be able to SELL the product, no matter how wonderful it is).

        Yes, spending money AT THE START on nice, coloured, “ooh this looks all fancy and science-y” brochures and reports would have persuaded more people than “Huh, this looks like it was knocked up on Word, anyone can do that“.

        The scrupulosity (no unnecessary spending! certainly not on fripperies like fancy typefaces!) meant that they ended up doing less good than they could have. “A spoonful of sugar helps the medicine go down”. I’m sorry MetaMed didn’t make it, but they can’t blame the customers, they have to realise where they went wrong, even if it’s for as dumb a reason as “Duh, we didn’t print our reports on glossy paper!”

      • I’m curious about the part about the difficulty of finding competent suppliers. Do most businesses have that much trouble? I’m tentatively concluding that Metamed tried to skimp on the administrative effort needed to find good suppliers and get rid of bad ones, but this is a guess.

    • moridinamael says:

      I’m sure this sounds condescending and whatever, but that blog posts reads like The List of Things You Learn When You First Enter the Business World.

      Like, holy shit, people expect information-products to be delivered with professional-looking packaging/formatting. First of all, you learn this within your first year of working at any company. Second, rationalists all know about the Halo Effect (etc) and how much appearances matter to perceptions of credibility. None of this should have been surprising.

      A corollary learning to the first is that you want to enlist the services of consultants and contractors based on their reputation, not by what they tell you they can do, because every incentive forces them to exaggerate what they can do (“the symbol”) whereas reputation correlates with what they can actually do (“the substance”). Acting shocked that there might be some mismatch between contractor promises and capabilities (or acting like this observation is itself mind-blowing) is baffling.

      Anyway, this blog post annoys me because it seems to be trying to say “we failed because people are too stupid to understand our great product” and I read it as saying “we failed because we’re bad at business.”

      • Deiseach says:

        Ironically, too much information can result in you communicating less than you want to do. My go-to example for this from work was when I got lumbered with designing an invitation-cum-notice about the opening of a new literacy centre which was being mailed to a list of the local Great and the Good for the formal ribbon-cutting and photo op.

        All that was basically needed was: event, host organisation, date, time, location done with some bit of fancy typeface and borders and making it look more expensive than it was (you can do a lot with coloured ink and different weights and textures of paper). The service co-ordinator got over-enthused about how good the first draft looked, and wanted more and more added in, but all still fitting on an A4 page, until the end result (despite all my struggles to encourage her to leave the damn excess out) was a mess.

        It ended up with a full page full of bumpf that realistically nobody was going to read – to cram all she wanted to include on it, I had to keep reducing the point size, so it was a full page of small print – but instead would dump in the bin, which meant the important information she wanted to communicate was going to be trashed along with it.

        Sounds like MetaMed had a bit of the same problem: so invested in cramming in all the “But you really need to know this” that they over-rode the advice about “Yes, but sometimes the aesthetic really needs to come first” and they ended up with printed material people didn’t trust because it looked shoddy.

  6. Michael Keenan says:

    Hey Scott, have you considered writing for The Atlantic? They were recently calling for freelance writers (and they mentioned that they can pay “more than we’ve been able to in the past”):

    we’ve made a blueprint of what we’re looking for in pitches from freelance reporters and writers.

    We want your riveting, original, weird, and wonderful stories about science, technology, and health.

    When it comes to topics, we’re expansive and inclusive. So “health” means not just nutrition, exercise, and illness, but also relationships, psychology, sex, family, etc.—all the things that make up a human life. Technology, similarly, is not just about gadgets or Silicon Valley startups. It is fundamentally about people: inventors and engineers and researchers who, prompted by their experiences in the world and the thinkers who came before them, stitch together new systems and bring into existence ideas that really do change the way humans interact with one another.

    We like stories that are both serious and silly. Basically, if there’s something you’ve always wondered about, or a question you can’t find a good answer to, that’s a good place to start.

    • Zebram says:

      Near the bottom of that Atlantic ad it said the following:

      “We want a diversity of voices. Young white men are always welcome to pitch, but we’re hungry for other perspectives.”

      I feel like Scott would end up writing like 10 pages about that statement.

      • Anonymous says:

        …this is the strongest reason for him to put together a submission.

      • Terracotta says:

        “We want a diversity of voices. Young white men are always welcome to pitch, but we’re hungry for other perspectives.”

        Did a quick tally of the authors that they showed in their list. There are 23 unique authors in that list. Of those 23 authors, 16 of them are women and 7 of them are men.

        Of the men, there is 1 asian man and 6 white men.

        Of the women, at least 10 (probably 12, but pictures of all of the authors were not available) are young and white women. There were also 2 black women and 2 asian woman.

        Almost all of the authors looked young (2 of the white men and 1 of the white women looked older), but I don’t have enough confidence in my age assessment skills to further subdivide based on age.

        Based on this, it seems to me that “diversity of opinion” is an applause light for young white female perspectives.

        • Deiseach says:

          Based on this, it seems to me that “diversity of opinion” is an applause light for young white female perspectives.

          What sorts are their readers? If the people who read “The Atlantic” are predominantly young white urban professionals (or whatever the most-desirable demographic is nowadays), then their contributors will probably be predominantly young white urban professionals because non-white/non-urban etc. people don’t read it and so have no idea they’re asking for contributors.

      • Anonymous says:

        I think it’s code for “we want more black and hispanic people.” But you can’t just say “white people are invited to apply, but we are hungry for a diversity of opinion,” as that might offend some white uterus owners, and uterus owners are an officially recognized oppressed group. Young white MEN, however, are a very safe category to subtly disparage.

        • Terracotta says:

          That’s a good point that I hadn’t thought of.

          Thanks for the Straussian reading.

          • Tarrou says:

            The point is valid, but not Straussian, I think.

            Does not “Straussian” imply deeply hidden or inverse to the plain text? This is overt and obvious, unless one has been conditioned by long social and media practice to ignore slights both micro- and macro- against white males, due to their supposed “privilege”.

      • Cauê says:

        This has been happening a lot in the last (3? 5?) years.

        The Guardian’s was the most dramatic recent fall, from something I respected to basically Salon. I mourn the Atlantic as well, although they still have some good content. Vox was maybe the biggest disappointment, as I mentioned below. The rise of HuffPo didn’t help, and neither did Buzzfeed. I may have shed a tear for Cracked at some point. And then there’s the “enthusiast press” in some communities, which… well, let’s not go there.

        I’m not imagining this trend, am I? I mean, I guess I might have gotten more sensitive to both clickbait and certain political fashions, but I don’t think I was the one who changed this much in the past few years.

        • Adam says:

          I don’t much care about politics, but the rise of clickbait seems very real to me and it’s what stopped me from reading basically any news site at this point. If I want to know something now, I’ll usually just go to the WDI, BLS, some record of the federal budget, Pew Foundation, whatever, and get the numbers directly. Whatever value there may have once been in long-format reporting has been drowned out completely by blogging and the same two thumbnails telling me over and over that some jellyfish-looking thing in a dude’s hand can turn me into Mr. Universe and how stupid Dallas drivers feel that they didn’t know this law.

          • Brock says:

            Adblock Plus can rid you of the stupid thumbnails.

          • Adam says:

            A better version or what? I’m looking at a stop sign with ‘ABP’ in the middle of it sitting next to my url input box right now. I don’t think it can adequately distinguish between thumbnails that are links to legitimate related stories and thumbnails that are sponsored; plus, at least half of the shitty clickbait thumbnails actually are legitimate links to related stories.

          • Brock says:

            Assuming that the thumbnails are from Taboola, add this to your filters:


            Similar filters will work for other purveyors of “sponsored content”.

            Also, “Fanboy” maintains lists that are more comprehensive than the default one.

          • Adam says:

            Thanks. I think that might have helped. I tried Buzzfeed and HuffPo right now. Buzzfeed only had links to its own content and HuffPo had no links at all, just beautiful bare white space to the right of the story.

        • Nornagest says:

          No, it’s real. The only news sites I read anymore are the WSJ and the BBC: the latter harbors some clickbait within its opinion and analysis sections, and tends to fall victim to stuffy British incomprehension whenever topics adjacent to colonial customs are in the news, but its headlining articles are still usually trustworthy. The former’s still pretty solid overall, albeit heavily paywalled and thick with boring business news. Some people seem to like Al-Jazeera; I don’t have much exposure to it myself.

          As to news sites I used to read… the NYT was always pretty upfront about its political leanings, but now it’s trying to be Jon Stewart with better fonts. The San Francisco Chronicle went from tolerable if you don’t read the comments to flat-out unreadable; I can’t even trust it for local news anymore. The Atlantic went from hit-and-miss to mostly miss.

      • vV_Vv says:

        “We want a diversity of voices. Young white men are always welcome to pitch, but we’re hungry for other perspectives.”

        Isn’t this thing actually illegal?

        • Adam says:

          Nah. There are all kinds of exemptions from Title VII. You can discriminate against communists, discriminate for national security reasons, you can give preference to Indians if you’re near a reservation, you can discriminate to redress a pre-existing imbalance (which seems to be what they’re trying to do here), plus independent contributors don’t count as employees anyway, so they aren’t even subject to legal protections.

    • Deiseach says:

      They’re paying their freelances? Quick, everybody grab that opportunity!

      No, but if they’re really willing to pay (instead of “Getting an article in our prestigious publication will get your name noticed and help build up your portfolio for prospective employers, why do you want filthy lucre on top of that?”) then it’s interesting to see, as I’ve been led to believe the current model (particularly when it comes to generating online content for the webpages of publications) is to invite people starting out or wanting to get into journalism etc. to provide unpaid content in exchange for exposure.

      It certainly sounds like it would be a good opportunity for Scott, if he can manage the free time, to dip his toe into the waters of getting published. I don’t know if pop science articles for mainstream media count towards professional repute, but they probably can’t hurt too much (so long as it’s a step above “Dr Bob Advises”). And if Scott ever harboured notions of becoming a stand-up comedian, there is certainly precedent for moving from medicine to comedy via writing for magazines 🙂

    • Buckyballas says:

      Somewhat related: Conor Friedersdorf, a frequent Atlantic contributor, is a fan of this blog.

    • Scott Alexander says:

      Writing for real news sources is hard in the same way writing for LW is hard, but a thousand times more so. And I have other projects that I prefer to use my do-actually-useful-things energy for.

      Also, if I ever started writing for a news source, it would be Vox – they have a lot of opportunities available, they’ve been nice to me before, and they’ve got a few lines of connection with the rationalist crowd and are good at publicizing things like AI risk and effective altruism.

      • onyomi says:

        I am really pleasantly surprised by the quality of Vox, especially given that I didn’t have a super high opinion of Ezra Klein prior to that.

        • Cauê says:

          You guys are weird.

          (I’m surprised, is what I mean)

          Back when Vox was launching, saying they’d do “data driven journalism”, I was actually pretty excited (despite previous opinions about Ezra Klein). That, 538 and Upshot could have been the salvation of internet journalism. I really tried to like it.

          But, of course, turns out “data journalism” can easily just be the same kind of partisan BS we had before, now with a yellow coat of paint, and I should have known better.

          • Cauê says:

            GamerGate and Donald Trump might not seem obviously connected, but they are: both are expressions of a disturbingly prevalent belief in the United States that not only is it right and good to hate women, but that hating women is so right and good that anyone who tells you not to hate women is a threat to core American values. Some believe it is such a threat that it is appropriate to punish them by, say, blanketing them in online harassment or calling in a SWAT team to their house.

            Yesterday from Vox. Come on, guys, the bulshit is neither minor nor subtle.

        • Held in Escrow says:

          I’m going to agree with Caue here, albeit perhaps coming from the other direction. I was really exciting for Vox because I was a massive fan of Ezra Klein’s Wonkblog… but its actually ended up being slightly more longform HuffPo. I mean, I’m fairly dedicated blue tribe and my much more blue friends are even more disappointed with it. Not because it isn’t blue enough, but because it’s shitty propaganda a lot of the time. If you’re going to argue for a point I believe in do it well dammit!

  7. Emile says:

    I recently ran into this blog, which seems well-written and has some overlap with the interests here:


    * (complaint about how people complain too much, on how policy implications need to be treated more seriously, etc.)
    * (avoiding the toxoplasma of rage)

    • Cole says:

      Thanks for the links, she has some good blog posts. I do wander if she will be successful in staying away from social issues. She clearly has some talent for writing about those topics.

    • 27chaos says:

      “In fact I don’t think I ever encountered words that grated at me more when I was doing my undergraduate degree than “Policy Implications”.”

      Already I like this woman.

    • birdboy2000 says:

      I’m very sympathetic to the idea of the social issues diet – indeed, I wonder if it doesn’t go far enough. This stuff has a way of leaking into every corporate-owned website, because corporate websites prioritize controversy (which brings clicks) over civility and give a wide license for abusive behavior, especially when the “right” people are doing the abusing. (Admittedly, this listed most of the presently big ones, but I think the problem is an institutional enough one it requires a heuristic to address.) I have half a mind to extend it well beyond the sites listed and avoid using communities that are either owned by publicly traded companies or funded by venture capital altogether, although often community websites owned by non-media or non-social networking companies (i.e. official forums of sports team/video game X) can avoid these pitfalls, so I’m wary of extending the idea too far.

      The dilemma, of course, is that when you cut out facebook, reddit, twitter, and tumblr (not explicitly mentioned, but I think that omission is because they don’t use it in the first place) you also cut out the largest sites on the web and a lot of your ways to keep up with people you care about. Replacing them is an admirable goal, but not an easy one – and the alternatives the post offers are mostly of interest to people who care about economics.

    • The two articles aren’t exactly consistent with each other– the first is rejection of arguments that demand change without considering the difficulties and costs of change.

      The Social Issues Diet recommends ignoring problems that could be solved if people just stood up for themselves, but fails to look at why people are so bad at standing up for themselves or what it would take for people to become better at standing up for themselves or what might happen if they were.

  8. NE says:

    Just wanted to, ehr, “raise awareness” and comment that I’m running adblock after the Apptimize ad came out. The picture just brings terrible memories of working at similar small companies and being coerced to act as if my coworkers were my friends just because the relevant authorities think that this will lead to the company making more money, and that employee feelings at that level are something that’s ok to tamper with (the coworkers were not bad people at all, just generally didn’t connect that much with them, that’s all).

    Definitely still have a great opinion of the blog, have gotten people to read it after recommending it and all that, sorry for talking more about the negative than the positive : )

    • Scott Alexander says:

      If I host any more job ads, I will be sure to put in a recommendation for pictures of people sitting on their own and looking miserable.

        • AR+ says:

          That line was way better when I incorrectly read it as, “You’re Never Along With a Stand.”

      • Deiseach says:

        It would certainly be more realistic; I wonder how many people get turned off by ads featuring smiley, chirpy, good-looking people all laughing and having a high old time? When everyone knows that work is not like that, and if you spend all your time laughing and joking with your workmates, the boss will be having a little chat with you soon.


        • Adam Casey says:

          Are you tired of politicians who will give 110% to help hard working families? Vote for us, a bunch miserable bastards who promise to put in the bare minimum to help our key demographic: lazy single people.

    • Robert Liguori says:

      A few years ago, Gallup did a meta-survey, and identified 12 questions or so whose answers correlate most strongly with the success of a company. One of those was “Do you have a best friend at work?”

      Many companies, upon giving the Gallup test to rack up more meaningless metrics so that management can pretend they’re doing something, immediately put into place a bunch of measures to increase the best friend quotient. Very few asked “Wait, is this question even coherent? Should we try to reword it for the IT crowd, who will read ‘A best friend? Incoherent question, ignore.’?

      (Of course I can imagine the question does correlate; companies that use stack ranking and otherwise pit employees against each other will not succeed. But that doesn’t mean you have a general factor of correlation, any more than scurvy means you can improve health generally with vitamin C megadosing.)

      So of course, at a team meeting, when asked how we could improve our metrics, I suggested cash payments of $1,000 for every member of every team who answered next year’s survey with perfect scores. When this was pooh-poohed, I pointed out that the correlation didn’t exist between having a best friend at work and doing it well, it correlated between saying to Gallup that you had a best friend and doing well, so if we were going to pretend that correlation equaled causation, we might as well be rigorous about it. Then, just for fun, I estimated the $1,000 number as how much each of us were being cost based on lost productivity from meetings, taking the survey itself, and so forth, and was quickly shut up.

      You have a lot of freedom once you’ve realized that you can comfortably retire in your current role and pay grade.

      • Thomas says:

        You have a lot of freedom once you’ve realized that you can comfortably retire in your current role and pay grade.

        I want to be you.

      • chaosmage says:

        > Do you have a best friend at work?

        …sounds like it’d correlate with duration of employment, which of course must correlate with the longevity of the company.

        > You have a lot of freedom once you’ve realized that you can comfortably retire in your current role and pay grade.

        Sure. When you cease to value x, endangering x isn’t costly anymore, making lots of courses of actions cheaper. x = “promotion” is far from unique in that…

      • Robert Liguori says:

        Well, the big trick is to 1: Work in technology, and 2: Ruthlessly trim your expenses. I’m childless, live in a townhouse 5 minutes from my job, drive an old used Camry, cook the majority of my own meals, and so forth.

        As for the promotion thing…well, I have some co-workers who are playing the political games really, really hard to try to get promoted, for what to me looks like a really wonky return on life effort investment. Most people won’t be promoted to upper management, and most people who do get significant pay raises in my industry do it by changing companies entirely, where me not playing the games here and now won’t matter at all.

        The thing is, in my company at least, there are two different standards. There’s what you need to do to avoid getting reprimanded or fired, and what you need to do in order to become a candidate to move up through the ranks, and there’s an incredible amount of daylight between those two standards, such that if you structure your lifestyle to require the money you’re getting now rather than the money you hope to be getting in 5 or 10 years after playing and winning The Game of Thrones Office Politics, you can opt out of a good percentage of what makes working in the modern buzzword-obsessed corporate environment so obnoxious.

  9. BD Sixsmith says:

    I’ve tried this before and it was a miserable failure but I think it would be fun to have some fiction recommendations. I’ve been re-visiting The Name of the Rose, which pulls off the rare trick of being a “good read” and an elaborate feat of intellectualism. Hardly can claim not to have skipped a page or two but given that there are more than six hundred of them, and that eleven tenths of those contain something to amuse or intrigue, that is no major criticism.

    • Pku says:

      Jonathan Strange and Mr. Norell is pretty good (just found out it got a TV show). On more classical notes: Jack London (White Fang/Call of the Wild for something longish, To Build a Fire for something short – all available on Guttenberg), 100 years of solitude (my brother claims Love in the Days of Cholera is better, but I haven’t read it), or most Murakami. A more regional recommendation is “A tale of Love and Darkness” by Amos Oz. It’s a great portrait of growing up in Jerusalem (and I hear the english translation’s pretty good).

      • BD Sixsmith says:

        I love Murakami. The tropeophilia that surrounds his work can be annoying – cats! ears! – but I think one of the reasons he attracts people who like that kind of thing is that he’s a great writer on loneliness.

        • Pku says:

          Huh… I’ve never thought about it that way, but that’s exactly right (another comment my dad made about it today: he spends so much time describing mundane things like laundry, because that’s all the characters have).
          I like it, because it turns loneliness from something painful and unendurable to quiet, relaxing solitude. It’s the literary equivalent of chamomile tea.

    • yli says:

      I’ve been entertained by The Instructions by Adam Levin recently. It’s a lot like HPMOR, insofar as it’s about a 10-year old genius wreaking havoc at a school. On the other hand it’s also solidly tasteful and “literary”, unlike HPMOR. It’s kind of like an Ender’s Game that you can read without feeling guilty about how you’re enjoying pornography. It’s delightful to see that it’s possible to pull that off. I found it by way of this good review.

      • johnfn says:

        I loved The Instructions. It was one of my favorite books.

        But I have to say that that article about Ender’s Game bothered me: calling Ender’s Game “pornography” seems overly reductionistic. (fwiw, I enjoyed Ender’s Game, though it’s not one of my favorite books.) The blog author’s point, as far as I can see, is that Card sets up a tense situation, and then resolves it, and so the blog author says “aha, this is pornography.” The truth is that this tension/release cycle is at the heart of a large part of popular literature, *especially* YA (just consider The Hunger Games, Harry Potter, Name of the Wind…). It’s everywhere.

        So to single out Ender’s Game specifically as “pornography” seems remarkably unfair to me when Card is just following a formula that has already been laid out (and doing better at it than most). The whole article bothers me – it almost reads as an attack on Card himself rather than the book.

        I’d agree that Ender is a Mary Sue, but that’s why we have the word “Mary Sue” – to describe that category of problem. We don’t need to start calling every book with a Mary Sue porn.

        Rant aside, do read The Instructions. But please don’t read it because it’s not “pornography”. It has the same sort of tension release cycle as just about any other read. Just read it because it’s an awesome book. And if you wanna feel smart while reading it, I suppose that is okay too. 😛

        (I found that the book got better and better as I kept reading. It kinda dragged until I got about a quarter of the way through. Then it picked up and up – till the point where I read something like the last 200 pages all in a single sitting.)

      • Deiseach says:

        Oh, good grief, that “Ender’s Game is pornography” blog post.

        Right, let’s get the disclaimers out of the way first: (a) I haven’t read “Ender’s Game” nor do I have any intentions of doing so, since the description of the story doesn’t interest me (b) I don’t know off the top of my head if I’ve read any of Orson Scott Card’s work, I think I probably have read some short stories but since nothing leaps to mind out of memory, that tells you how much they meant to me (c) I’m a Christian, yes, but that does not mean I’m disposed to support Orson Scott Card on those grounds: being a Catholic, I think Mormonism is misguided in its theology at best, heretical at worst (we argue over Christology amongst other things).

        So when I rip into this blog post, it’s not on the grounds of one Christian mindlessly defending another, are we clear?

        Now, the post makes good points about writing to a formula and setting up situations so they can be resolved to give the reader the guaranteed thrill, and about the problem with wish-fulfilment fantasies and Mary Sue/Marty Stu characters.

        The classic example of formulaic writing is probably the mass of the pulps, but we see it in everything: romance novels, indeed many genre novels, are written to a formula, manufactured boyband pop hits are written to a formula, blockbuster movies and the latest instalment in a horror franchise (Friday the 19th Part 26 Hellraising Freddy’s Revenge of the Killer Jasons’ Saws in the Hostel of the Rings) are written to a formula, soap operas are written to a formula. And yes, it’s like the same formula as porn: set up, tension, climax, resolution – rinse and repeat.

        Is Ender Wiggins a Marty Stu and is “Ender’s Game” a geek wish-fulfilment fantasy? Not having read the book, I don’t know. I do know professional writers get proper books published by legitimate publishing companies that involve massive Mary Sue/Marty Stu characters (I’ll never forget one fantasy novel where the author had very plainly fallen in love with her main character and poured on the Marty-Stuification to the nth degree, to such an extent it was embarrassing that something like this had got past an editor).

        I did think the criticism on those grounds sounded a bit hard, since I’ve never seen those points raised in reviews (that Ender’s penance at the end of the book is really gratification of vanity in the guise of self-pity), but not having read the book, I can’t opine on that view.

        And then Jesus.

        Suddenly in the middle, we get a whole paragraph about Jesus and geek wish fulfilment fantasy for Christians, and I went “What?”

        And then we segue back into “Ender’s Game” and criticising Scott Card on grounds of his religious beliefs, and I have to say – I think this is the main driver of the post. That the blogger was determined to see Ender as a Christ-figure, read this backwards into the book, and of course gave it a kicking on those grounds because it was a case of “give a dog a bad name and then hang him”.

        I don’t know if the impetus was that this person has a personal problem with Christianity in general or Mormonism in particular and, knowing that Scott Card is a Mormon, was determined to ferret out the “hidden secret Mormon/Christian proselytization” in the book (the mentions of the Narnia novels incline me to that interpretation, but I can’t read their mind); if they dislike Scott Card for whatever reasons either for his religious or political views, or simply on his prose style; or what bee is in their bonnet.

        But if in the middle of a review of a SF, romance, crime or horror novel, the reviewer sticks in a paragraph about Deng Xiaoping, Richard Nixon, Stalin, or Ludwig Wittgenstein (and if none of these are characters or even referred to in the novel), then you begin to suspect the reviewer is riding their own hobbyhorse.

        That’s what I think is going on here in this case, I’m sorry to say.

        And damn it, if it’s pornography to like a story with a beginning, middle and end, then I’m a consumer of porn!

        • stillnotking says:

          I’d be less sympathetic to the blogger’s identification of Ender as a Christ figure if it weren’t for the later books, in which the allusions were so blatant that the appearance of literal stigmata wouldn’t have seemed out of place.

          But yeah, that reviewer comes off as a bit of a jerk, and also the sort of person who is far too invested in the idea of making young people read the “right” YA novels — always a bad sign.

          • Tarrou says:

            There’s one profound difference between Ender and a “Christ figure”. The essential nature of a “christ” is to be both innocent and self-sacrificing, but Ender inverts this. Ender is the one with the physical and moral capability for violence and vast destruction. Christ takes on the guilt of the world in a moral sense only, Ender actually commits the crime(s) necessary to bear the guilt of a race.

            The whole point of the Ender series is to make the reader identify with Hitler, basically. It may be messianic in a sense, but to draw too close a parallel with the “lamb of god” would be to misunderstand or misconstrue both biblical text and Card.

          • LHN says:

            Ender takes on the guilt (both by his own conscience, and by the public who identify him as the Xenocide). But it’s really questionable whether he can be said to have committed the crime. He didn’t have mens rea: he thought he was commanding a simulation, and was dramatically quitting by breaking the rules he’d been given. Leaving aside that he was a minor who’d been subjected to years of military indoctrination in a setting isolated from other influences.

            (For the same reason, the guilt of his superior officers is at least complicated. They put a child in charge of potentially genocidal weapons without telling him. But they laid down a strict doctrine that they were only to be used for ship-to-ship combat, and were clearly genuinely shocked when he violated that. They were also honestly, though wrongly, under the impression that they were fighting a war for species survival. That said, they’re guilty of something, and their discussions make it pretty clear they’re aware of it.)

          • Deiseach says:

            Well, if you’ve read the later books, and you’re pretty sure Ender is meant to be a Christ-figure, and that’s your problem with the books, then you should say that from the start.

            Dropping it into the middle diverts the point being made, and the point of how fiction of type L can be different from fiction of type Q because of the set-up of a formula for tension-climax-resolution was interesting and worthwhile.

            But the impression I got was that this was an early post about “Ender’s Game” and that the rest of the series hadn’t been written yet, so either it was extremely prescient about Ender-as-Christ or it was jumping the gun. Either way, I’d much have preferred “Okay, I don’t like religion, I don’t like religion in my SF, and I particularly don’t like Orson Scott Card’s version of religion or his version of SF combined, so here goes” from the start. It would have been clearer and more honest.

      • switchnode says:

        Oh come on, if you’re linking controversially critical reviews of Ender’s Game you have to link this one.

      • Pku says:

        Speaking of things that are somewhat like HPMOR but more “literary”, I’d really recommend the Dresden files – It’s got the “awesome wizard using magic in creative ways, in a world full of awesome, complicated plots”, but (for better or worse) without the arrogance/status porn. It’s great, and it’s one of the few book series that get better as it progresses – It starts out pretty good, but gets better as the world/characters are fleshed out (and the author got more experienced).
        (The other main example of a series that improved in later books was a series of unfortunate events).

    • DavidS says:

      Maybe Anathem by Neill Stephenson. For me, great combination of the ‘good read’ and ‘elaborate feat of intellectualism’. Warning, however: it uses a lot of its own language (a bit like Clockwork Orange or something) – there’s a glossary but I found it worked very well if you read it and picked up the meanings naturally, as they’re more nuanced than just renaming a rabbit a smeerp. And lots of them you can get a sense of from context, or because they’re contractions/portmanteuas that we don’t have. But I know some people have less tolerance for starting a book where occasionally there’s a word they don’t get.

      Basically, it’s a bit like the literary equivalent of starting to watch the Wire…

      • I’d second that, although I’d really recommend all of Neal Stephenson. Diamond Age, Snow Crash, and his latest SevenEves are all quite, quite good. (In SevenEves, social media almost causes the extinction of humanity, and it flows organically from the plot. I just had to throw that in there.) I didn’t like the Baroque Cycle as much.

      • johnfn says:

        “Basically, it’s a bit like the literary equivalent of starting to watch the Wire…”

        You know, I’m glad to hear you say that, because when I watched it I couldn’t understand it at all and I figured I must have been an idiot.

      • Brad (the other one) says:

        This is kind of of topic, but I had a weird problem watching the wire: the dialogue seemed unrealistic.

        I’m not talking about cops talking about cop stuff or anything like that – I mean, how literally every other word out the mouths of the characters is a vulgarity. I don’t know *anyone* who talks like that (ecept perhaps people on certain imageboards); they curse like they’re on a quota system or something.

        This is the same problem, by the way, Game of Thrones seems to have with nudity; even when it makes no sense and is clearly gratuitous, bam, naked people rambling around in the background of an expository scene. Yes, HBO, I know we’re paying extra to see you, no, I don’t need this to be convinced to watch the next season of True Detective.

        • Nicholas Carter says:

          There’s certainly a subculture in which the inter-community nudity taboo has been almost completely shredded, which has many members for whom swearing has become an adverb of a completely non-vulgar type. If you presume that members of this community of that type are the target demographic for something, it makes sense.
          (I seriously have some acquaintances that are naked pretty much the whole time they are in their home, and sometimes get naked in a non-sexual way in my home, for several hours.)

        • Devilbunny says:

          The language is not particularly extreme for ghetto culture. Perhaps you just don’t know any truly ghetto people?

    • nydwracu says:

      Bonfire of the Vanities. Back to Blood. The Illuminatus Trilogy if you haven’t already. Sartor Resartus. Mishima is popular these days but I haven’t been able to find a good translation of anything he’s written.

      • yli says:

        Second the recommendation of Bonfire of the Vanities and Back to Blood. Tom Wolfe is so much fun to read. He has a very distinctive style though. If you don’t like it, you won’t like any of his work. If you like thinking about status and signaling, like a lot of people here, you’ll probably have a good time with Wolfe.

        Any reason you didn’t recommend his other two novels, A Man in Full and I am Charlotte Simmons? A Man in Full is probably my favorite of his. It’s hard to get interested by reading the synopsis (“So it’s about a real estate mogul in Atlanta? Who cares?”) But for me, it delivered the same goods are BotV except with the benefit of one novel’s worth of experience.

        • nydwracu says:

          I liked the two I recommended more than the two I didn’t. (I liked A Man in Full more than I Am Charlotte Simmons, which struck me as a satire of a media trope that may or may not actually exist. But that’s his “I was Roissy before Roissy was Roissy” novel, and it’s worth reading on those grounds alone, especially given that there are probably a lot of Adam Gellins around here.)

      • BD Sixsmith says:

        I read and liked Ivan Morris’ translation of The Temple of the Golden Pavilion.

      • dinofs says:

        I’ve been reading a good amount of Mishima lately and the translations never really bother me — I can tell that the prose is a bit clunky at times, but for the most part the ideas and images still shine through. But then again I don’t read Japanese, so I have no idea if what I’m getting is the “real” Mishima.

        Actually I’ve never really minded translations in general, even from languages I can read. I think it’s like subtitles — they drive some people crazy but I don’t see the problem. I’ve even talked to people who say they refuse to read anything in translation, full stop. What is it about translated literature that bothers people so much that they’re willing to miss out on the vast majority of all writing that isn’t in their handful of spoken languages? I’m not asking because it bothers me, I’m just wondering where this comes from.

        • anon a gnon says:

          I read Japanese semi-fluently, talk regularly with translators/editors of fansubbed anime+visual novels (think *very* long books written in power point form).

          I’ve noticed essentially two groups of people who read translations: ones who are fine with “literal” translations and ones who believe that translations should be ”what the writer would write, if they wrote stuff in English”. I fall closer to the latter group and talk regularly with someone who believes that it’s basically impossible to talk with the English fanbase about a work you’ve read in Japanese.

          There are some writers who are an absolute joy to read in Japanese, who demonstrates a masterful command of the language, with constant wordplay and lots of cultural references. As may be expected, the translators don’t necessarily do the greatest job of this (most of these end up bland). The blandness tends to come from translators who prefer literal translations.

          I bet that even as you scale up in the skill of translation, a similar scaling of expected translation quality happens, so that even with non-amateur translators you see complaints of bad versions. I remember the edition of Brothers Karamazokv had some awkward phrasings, that Murakami’s South of the Border, West of the Sun got castigated for having a borderline unacceptable translation, and that Gregory Hayes’ version of Marcus Aurelius’s Meditations was widely praised for retaining the aphoristic tone of the original diaries.

          Losing the voice, especially for a writer who you greatly admire, could seem like an absolutely catastrophic event to your enjoyment. In fact, I’ve seen about three or four people “reread” old works in the original language and talk about how they took away many different things from the original, compared to the translation.

          It wouldn’t be surprising that someone who 1) believes in general that translation is low quality and 2) believes that a good portion of the joy in writing is in how the prose is put together would try to confine themselves to only their native language, where at least you could get good prose that doesn’t go through the noisy process of translation.

          ‘course, my view on that isn’t to stop reading translations; it’s to learn the language.

          • Held in Escrow says:

            I suspect much of the “must be literal” crowd among the (non-japanese speaking) fanbase tends to come from those who want to “write how the author would have done in English” tend to be horrible at writing in the first place.

    • Lemminkainen says:

      Cixin Liu’s “The Three Body Problem” is excellent in that regard. Mark Danielewski’s “House of Leaves” is very compelling and also mind-bending, if you have a tolerance for postmodernism and text tricks (although Danielewski is smarter about using these than most authors are)

      • tcd says:

        I will second “The Three Body Problem”. The recent English translation is very well done. If I have one problem with the story it is that while 75% of the story is truly excellent science fiction, the remaining quarter was a bit disappointing (I found out after finishing that it was part one of a trilogy).

        I think Liu may have missed an opportunity to cut about a quarter of the fluff and add another 200-300 pages to the story and he would have ended up with one of the best single volume science fiction stories in a long time. I am still very much looking forward to the upcoming books.

    • Anatoly says:

      I think some SSC readers might like Helen DeWitt’s The Last Samurai (no relation to the Tom Cruise movie). It’s a novel that hard to describe in a few words. The protagonist, throughout much of it, is an absurdly precocious boy who is not at all like Ender. If you’re fascinated by languages, music, and the questionable meaningfulness of, you know, everything, that might just be a novel for you.

      • James says:

        Your last sentence makes this sound like my kind of book. I’ll keep my eyes peeled for it and pick it up if I ever see it.

      • Yes, one of my all time favorite novels. Highly recommended. Also got me into Kurosawa films, which is a plus in its own right.

      • Tau says:

        Yes! This book is amazing. I don’t know how to say what it’s about, but it has stayed with me for a long time.

        Also, I recommended it to my mother and she got about ten pages in and just couldn’t anymore. I think maybe SSC readers should not recommend it to their mothers.

        …actually, I think maybe I will find that book on the shelf and read it again, it’s been a while.

    • Alejandro says:

      If you haven’t read them yet, an obvious recommendation is Borges’ short stories and essays, given that The Name of the Rose is (among other things) an elaborate pastiche of Borgesian themes and motifs, as is lampshaded with (ROT13 for spoilers) gur anzr naq oyvaqarff bs gur ivyynva.

    • Brock says:

      Name of the Rose is on my need-to-reread list.

      Have you ever read any Nabokov?

      Lately, I’ve been checking out books from Postmodern Mystery. Name of the Rose is on their list, so you might like its other suggestions.

      • brad says:

        I’ve tried to pick up a famous Russian novel three times — Crime and Punishment, Anna Karenina, and The Master and Margarita — but didn’t manage to make through any of them. And I’m usually not the type to give up on books. Thinking back I don’t remember not enjoying them for the same or similar reasons, but I think it is unlikely to be coincidence.

        • houseboatonstyx says:

          Start with Ayn Rand, then work backwards to some Russian novel she admired?

          May need hearing a heavy Russian accent, or disregarding most commas; Rand uses them to indicate a change of pitch.

      • Brock says:

        You probably shouldn’t think of Nabokov as a Russian writer in the context of applying a “Will I enjoy this?” heuristic. His prose can be a bit on the flowery side, but it’s nowhere near as ponderous as Dostoevsky or Tolstoy.

        Lolita is generally regarded as his best novel, and contains some of the most exquisitely crafted prose ever written in English. I’m partial to Pale Fire. It’s an odd work, consisting of a 999-line poem by John Shade, a Robert-Frost-like character, and line-by-line commentary written by his delusional literary stalker, Charles Kinbote.

        • Nornagest says:

          Pale Fire is amazing. I’ll give a nod to Ada, too; it’s Lolita-esque in some ways but also has some odd SFnal elements that remind me of Gene Wolfe’s The Fifth Head of Cerberus.

      • BD Sixsmith says:


        I did like Lolita, though I couldn’t help feeling uncomfortable about it. Can’t be sure if that’s my fault or the fault of the novel.

        • Brock says:

          If you enjoyed Lolita, I suggest giving Pale Fire a try. Approach it like a puzzle.

          Here’s a suggestion that may seem somewhat off-the-wall, but like Name of the Rose it’s a distinctive spin on the detective novel: Dirk Gently’s Holistic Detective Agency. Douglas Adams is better known for the Hitchhiker’s Guide series, but I rank Dirk Gently ahead of them.

          • Who wouldn't want to be anonymous says:

            >. Douglas Adams is better known for the Hitchhiker’s Guide series, but I rank Dirk Gently ahead of them.

            Yes. Ten times yes.

          • BD Sixsmith says:

            Yes, I loved Dirk Gently: the first a lot more than the second, though.

          • Brock says:

            Well, it sounds like our taste in fiction is pretty much in line. I’m now interested in getting recommendations from you!

    • Saul Degraw says:

      Check out the Cornish Trilogy by Robertson Davies, it is a great series of novels on academia and the arts with a lot of dry wit. There is also A Dance to the Music of Time by Anthony Powell which follows the lives of a bunch of Brits from the 1920s to the 1960s. E.L. Doctorow died last week and Ragtime is brilliant. Don Carpenter is also a brilliant novelist but his novels might be harder to find.

    • Skaevola says:

      Did you try any of Eco’s other books? I enjoyed Foucault’s Pendulum and The Prague Cemetery, but I couldn’t get into his other fiction. Both of those two books carry on the high level of intellectual sophistication, while having interesting plots. You’ll especially like Foucault’s Pendulum if you have any interest in secret societies.

      He has a new book title “Numero Zero” coming out this November.

      • BD Sixsmith says:

        Foucault’s Pendulum has been on my to-read book for an embarrassingly long time because I’m just the sort of person who would love it. Get the sense that his other novels will fade into history, though.

      • Jaycol says:

        Baudolino has actually been one of my favorites since high school, but to say it’s not for everyone is an understatement. Talk about “fading into history.”

    • PsychoRecycled says:

      K.J. Parker (who has now admitted that they are Tom Holt, which is massively disappointing to me) writes fantasy-without-magic (mostly) fiction which includes a reasonable amount of technical accuracy: in a book they wrote where bows and the making of bows featured heavily, they actually made all of the bows described in the book except one.

      For a sample of their writing style, they have a reasonable amount of short fiction online: here are the links. (Depending, it might not be obvious that I meant to direct you towards ‘External Links’.) I would particularly recommend A Small Price to Pay for Birdsong.

      • Anatoly says:

        Just out of curiosity, why has the admission so disappointed you? (I haven’t heard of either author before).

        • Richard says:

          without speaking for PsychoRecycled, I see two immediate possibilities:

          KJ Parker has been widely assumed to be female for some reason, so discovering ‘she’ is actually Tom might be a surprise.

          KJ writes ‘serious’, usually ‘dark’ fantasy, Tom writes totally bizarre humour fantasy. To me realising the two are one and the same causes a small amount of cognitive dissonance. Mastering both ends of the spectrum is really impressive though.

          BTW, knowing they are the same makes Interview with K. J. Parker by Tom Holt a much funnier read.

          • PsychoRecycled says:

            Mostly column a, with a little bit of column b. I had grown attached to the idea of a female retired lawyer who picked up metalworking and siege warfare in her spare time also being an excellent writer. I maintain that my version of Parker is a lot more fun, and I liked having a mystery.

            The cognitive dissonance is largely due to my having been pretty darn convinced that K.J. Parker specifically wasn’t Tom Holt: a French magazine used the female pronoun a few times, and someone on Wikipedia claimed that they represented Subterranean and that, while they couldn’t comment on Parker’s identity, they could unequivocally state that Parker wasn’t Holt.

            I’m now picking through the statements made by Parker, and trying to figure out if any of them are factually inaccurate, i.e. where they lived, what they did for a living.

    • johnfn says:

      I’ve been reading Red Rising. It’s great! It’s pretty much Ender’s Game… IN SPACE!

      … oh wait, Ender’s Game was already in space? eh, whatever…

      Yeah, though. It’s not a particularly intellectual book – the prose is pretty nice, but that’s really all – but honestly, I’ve read too many intellectual books and kinda forgot about the raw joy of reading a really strong page turner. Definitely would recommend if you have that problem.

      • Held in Escrow says:

        I really enjoyed Red Rising. It’s fairly predictable and derivative but I was hooked all the way through. The second book wasn’t quite as good as said predictability was replaced with ass pulls, but I still have high hopes for the third in the series.

    • Dead Milkmen fan says:

      I am an infrequent SSC commenter and feel like an atypical reader (“SSC-curious” or something), just so you can calibrate (look mom, I’m learning the lingo!). And I say if you like Name of the Rose, you might like Mason & Dixon by Thomas Pynchon.

      You may have heard that Pynchon is weird or difficult or postmodern or other off-putting things; you may have read other Pynchon novels yourself and come to the same conclusion. But M&D is very concentratedly Ecovian to a degree TP’s other books aren’t.

      (That said, I don’t want to downplay the Pynchonian wackiness that pops up here and there: there are extended jokes about George Washington’s hemp crop and the origins of ketchup. Plus a surreal subplot about Jesuits and Feng Shui masters operating, secret-society-style, in colonial America.)

    • sweeneyrod says:

      Blindsight is a brilliant sci-fi book about consciousness and first contact, among other things. The Culture series by Ian M. Banks is more good sci-fi. The Bartimaeus trilogy is a YA fantasy series. Worm is a web serial about superpowers. Reading Dune made me realize that Less Wrong is just fan-fiction of it.

  10. Shitposter says:

    In a recent post you wrote

    At one point, [Eliezer] bought and sent me a [nutrition] book he was interested in so that I could review it and tell him if it made sense. I told him it was wrong, and he listened.

    What was the book + explanation for why it was wrong?

  11. Shenpen says:

    I’d like to test an aspect of Moldbuggian theories of history: was there a definite point in history where bicoastal liberal urban Puritan-Protestants in the US pulled a Rousseau i.e. basically revolted against Calvin’s “total depravity” theory and based their philosophy on the complete rejection of the consequences of original sin and instead started believing in human perfectibility?

    Or was it actually so right from the beginning, all along? Folks like Cotton Mather came up with the idea of moving from England to America meant a return to the Adamic pre-sin condition (original sin canceled, so to speak), shining city on the hill and all that, was it more or less an explicit rejection of both Calvin’s total depravity theory and the Catholics more moderate version of original sin and its philosophical, political consequences (people being vulnerable to temptations towards sins etc.) ?

    I mean the only way to do that is to openly break with Calvin’s most important ideas. Did they do so? Could Puritans have the courage to break with Calvin, is this likely? John Knox was a Calvinist and American has Knoxvilles all over…

    • Deiseach says:

      Wasn’t that Transcendentalism? The whole “plain living and higher thinking” idea? Nathaniel Hawthorne’s “Young Goodman Brown” certainly gives the Puritan ideal a good kicking in that it claims that constant vigilance for looking for witches and the Devil under every bed created a society where people were so convinced of their own depravity they became the very terrors they feared while maintaining an outward façade of probity.

      Modern American Calvinism, at least the little bit of it I’ve encountered, seems to be very prickly when accused of the whole “total depravity/the elite are the only chosen and saved/preaching to the reprobate will not save them/double predestination” theology and softpedal the “chosen to be damned” element of the foundational theology; even the Five Point Calvinists (except for a very few hold outs) are more inclined to emphasise the necessity for preaching so the unchurched can attain salvation rather than the uselessness of trying to inculcate a saving faith in those not predestined to salvation.

      Then again, I think the American versions of Christianity are fascinating. What turned out to be a temporary fad in a minor offshoot of a denomination, preached by an Anglo-Irish clergyman, turned into this huge thing in American theology – I mean Dispensationalism and its best-known element, the Rapture.

      I was so completely ignorant of “the Rapture” and its allied theology that, until I started interacting online with American evangelicals, I didn’t even know what the Catholic position on it was, or if we even had a position. I had to look it up! (Turns out we’re officially amillenialists). That’s the fascinating thing: what are obscure minor points or not even showing up on the radar in global and historic Christianity are considered (by virtue of the fact that Americans are talking to Americans about American experiences) to be the big, important and most importantly “All Christians everywhere believe that…” parts of the religion.

      • Leif says:

        FWIW, here’s what Matt Slick of CARM (the Christian Apologetics and Research Ministry), a Calvinist, has to say about total depravity: and about the five points:

        I don’t know enough about theology to say whether this agrees or disagrees with what you said. I’m just linking it because it seems like a relevant source. CARM comes up first on a lot of Google searches for Christian topics, and Matt Slick has been (heavily mocked as) a guest on The Daily Show, so it’s arguably a big deal.

      • CatCube says:

        As a Lutheran, (and therefore fellow amillenialist) I can barely keep from rolling my eyes when people start banging on about the Rapture or about the thousand year-reign on Earth.

        I didn’t know until recently that Luther had proposed that Revelation be made an appendix (or outright declared apocryphal) because of what the tendency to take an obviously symbolic book hyper-literally had on people’s theology.

        • AnObfuscator says:

          Yeah, it’s inclusion in the Canon was highly controversial and contentious, as I recall. The Eastern Orthodox church was particularly reluctant to accept it, if memory serves.

      • Brad (the other one) says:

        If I may ask, do you know what the (official) Catholic position is on 1 Thessalonians 4:17?


      • Nick says:

        It might amuse you, Deiseach, to know just how Americans characterize our own Catholicism. I’ve run into the same story now in a few classes on Catholicism in history: “American Catholicism was viewed with great suspicion of heresy”—followed by a long section on modernism or something—”until, that is, the popes in the twentieth century came around and admitted the American Church was the awesomest, most flourishing Church ever!” Do the Irish have their own narrative of their Catholicism?

    • nydwracu says:

      La Wik:

      Many Congregational churches claim their descent from Protestant denominations formed on a theory of union published by the theologian Robert Browne in 1592. These arose from the nonconformist religious movement during the Puritan Reformation of the Church of England. In Great Britain, the early Congregationalists were called separatists or independents to distinguish them from the similarly Calvinistic Presbyterians. Some Congregationalists in Britain still call themselves independent.

      Congregational churches were widely established in the Plymouth Colony and the Massachusetts Bay Colony, later New England. …

      Without higher courts to ensure doctrinal uniformity among the congregations, Congregationalists have been more diverse than other Reformed churches. Despite the efforts of Calvinists to maintain the dominance of their system, some Congregational churches, especially in the older settlements of New England, gradually developed leanings toward Arminianism, Unitarianism, Deism, and Transcendentalism.

      If you mean human perfectibility as in the Oneida Perfectionists, what you’re looking for is the Oneida Perfectionists.

      • Shenpen says:

        What I have in mind is mostly how prog political ideas derive from puritanism / protestantism. While Voegelin provided a decent hypothesis for that, I thought this largely happened in Europe (Rousseau being the most important link, who came from Calvin’s Geneva and ended up influencing the French Revolution) and got imported to the US somewhere around the secularization wave of the late 19th century, Moldbug presents a different hypothesis, namely that prog ideas have always been more at home in the urban Northeast US, it is not a late import, but a long co-evolution between the US and Europe or perhaps to some extent actually exported from the US to Europe.

        My point is, if Voegelin’s immanentizing-the-escathon, secularizing-gnosticism process actually happened in the US and not in France, we must see some Protestant groups actively pulling a Rousseau move at some point in the US. I.e. starting to preach something along the lines of human nature is naturally good and virtuous (i.e. original sin cancelled) and only the poor organization of society causes all the problems.

        Oneidas are a bit late given that they were founded in 1848. 1848 was precisely the year when apparently everybody went crazy prog in Europe. If Moldbug is right and this was imported from the US, American Protestant churches pulling this Rousseau move and immanentizing the escathon must have happened no later than 1800 and perhaps far, far earlier, even pre-Revolution.

        • Deiseach says:

          1848 was certainly the Year of Revolutions; we even had one (albeit a very small one) in Ireland, but I don’t think that was particularly influenced by American religious thought.

          This was post-Famine Ireland, when the physical force movement of the Young Irelanders had achieved ascendancy over Daniel O’Connell’s constitutional progress policy. The influence was very much from Continental Europe. Now, if Europe was influenced by America, then perhaps – but in what way? French Republicanism seems to have been the major impetus. America was important as representing a democratic, constitutional republic and as an example for the various nations of Europe with their own individual emphases (doing away with monarchy without running into a corresponding Terror for the French, unifying disparate states into one nation for the Italians, breaking free of colonial influence, etc.)

          If American religious thought influenced American politics and policy which in turn influenced European politics and policy, then sure, at second- or third-hand it had an effect, but I don’t really see a wholesale importation of specifically religious or secularised-turned-into-civic-religion ideas from America to Europe (for religion, as I’ve said with Dispensationalism, it generally turns out to be the other way round: small crackpot groups here become huge denominations or non-denominational movements when exported to America).

          • Shenpen says:

            That’s roughly similar to my analysis of it: Moldbuggian theory is not entirely correct, the international “Cathedral” – the unelected rule of intellectuals, journalists and civil servants who comprise the US Gov or EU elites and their private-sphere satellites – is not entirely an American product. I mean, that is obvious that 1848 was already more or less an attempt of a power grab by those elites, just by looking at how Kossuth was a Mason (Freemasons are a rather obvious part of the Cathedral, the main reason they were constantly called out by conspiracy theorists is that they were more into actually coordinated and visible organization as opposed to the far more spontaneous org of other arms of the Cathedral) or e.g. Mazzini was a rather obvious Cathedral type fellow as well etc. but the whole thing was probably far more rooted in French Jacobinism than Anglo-American Puritanism.

            Of course post-WW2 one could say the power center of the Cathedral shifted to the US – it is still not clear if the EU as such is a product of competing Cathedral elites closing their ranks into a supra-national governance in order to compete more efficiently, or actually just separate geographical wings of the same structure.

        • Jaskologist says:

          I’ve long found Moldbug’s accusations of crypto-Calvinism to be one of his odder claims. If we’re looking for the common thread between American progressives, the French Reign of Terror, and the Communists in Russia and China* that provide so many NRX object lessons, it’s not Calvinism, it’s atheism. The religious impulse is inherent in human beings, and most of your major religions harness this impulse towards productive ends. Try to repress it altogether, and it explodes in unpredictable and dangerous ways. Very often it gets shunted into politics instead, becoming fundamentally about making sure your neighbor is holy and free of eye-splinters.

          As Solzhenitsyn put it, “we have forgotten God, that is why all of this has happened.”

          *Of course, much of my analysis may be void if you accept Moldbug’s expansion of “progressive” to mean “everybody who likes democracy.”

    • houseboatonstyx says:

      Or was it actually so right from the beginning, all along? Folks like Cotton Mather came up with the idea of moving from England to America meant a return to the Adamic pre-sin condition [….]

      Screwtape’s nephew Wormwood in this AU fanfic* unpacks the change in definition of ‘Puritan/Puritanism’ suggested by the canonical Screwtape in “Letter X” of _The Screwtape Letters_.

    • Douglas Knight says:

      I think Moldbug’s answer would be “the Quaker doctrine of the Inner Light…All the American Protestant sects, or at least all the Northern ones, became heavily Quakerized during the 19th century.” (He mentioned the Quakerization here, too.)

      I do not see how total depravity or perfectibility gets you any policy. Everyone believes that people can become better. Everyone believes that environment has some effect. Preaching is an attempt to make other people better. Disagreement is only about specific policies, not general principles. (Inner light does seem to be related to preferring the carrot over the stick, but I don’t see as much related to total depravity or perfectibility.)

      I also do not see that total depravity has gone away. eg, microaggressions: getting rid of your macroaggressions is not good enough. Nothing is good enough; you must always work toward perfection, even though it is not attainable.

      America doesn’t name things after people as old as John Knox. Forts Knox are named after General Henry Knox, as are most Knoxvilles.

  12. Emile says:

    Any recommendations on some good short online readings to get a better understanding of statistics?

    I’ve realized that my understanding of some issues is still imperfect (for example I’d also like to have a better intuitive understanding of how correlation and “part of variance explained” relate, as per Scott’s recent post), so it would probably be a better use of my time to read up on that rather than the comments of the recent drama thread 🙂

    • Zebram says:

      I’d probably read a Schaum’s outlines if I were you, depending on how much time I had.

  13. J says:

    Thanks for sharing your sense of humor with us, Scott. For several weeks now I’ll randomly smile when I ponder Big Parasitic Worm.

    • J says:

      (Weird, I can’t seem to edit the parent comment (hm, worked just now), and the second comment I made with a pair of links to hilarious images looks like it’s silently being held for moderation, perhaps, but with no obvious indication of that?)

  14. Anonymous says:

    Why do Mealsquares take square root of optimized?

  15. Zebram says:

    Hi Scott,

    Have you done any articles on moral nihilism? I did a quick search of your site and couldn’t find any articles, but that would be a topic I would be interested in. I would consider myself a moral nihilist and an existential nihilist. I was wondering if you have ever considered that morality does not exist in either an objective or subjective form and if you would like to go through, from your perspective, the evidence for and against moral nihilism.

    • Kiya says:

      Does not exist in a subjective form?

      I must be missing something here. Surely I can declare “I define morality as [insert partial ordering over the space of possible actions]” and call that a subjective morality endorsed by me.

      • Zebram says:

        I’m not sure I’m understanding you correctly, but it doesn’t seem like the subjective morality you are defining is anything but a collection of actions. By morality, whether subjective or objective, I mean actions that have some sort of quality like ‘oughtness’ or something similar to that.

        • Adam says:

          It’s not the set of actions he’s defining. It’s an ordering of the relative desirability of actions. Since that seems to be a thing people actually experience, the feeling that some actions are better than others, by that definition morality has subjective existence.

          That’s different than actions themselves having a quality we can point to and identify as “desirability.” As an analogy, think of the way people define attractiveness in a mate. You might like blondes better than brunettes, taller people better than shorter, and weight these in such a way that you’ve produced a partial ordering of human females. However, the only identifiable traits they have are things like hair color and height. ‘Attractiveness’ here is a real thing, but it exists solely in your mind, not in the objects of your attraction.

          • Zebram says:

            Ok, I see what you are saying. I don’t think any such moral desirability exists. We certain desire certain things and outcomes, but I argue that what we think are moral desirability ratings are in fact non-moral feelings which have been mislabeled with moral terminology through years of conditioning since we were 2 or 3 years old.

          • Adam says:

            That sounds plausible enough. All systems of morality are effectively designed to produce world states deemed better than others, which necessarily requires all of them, regardless of the specific approach they take, whether consequentialist, deontologist, virtuist, or something more esoteric, to be based upon the subjective valuing of one world state over another.

            If you haven’t already, I strongly suggest reading the research of Antonio Damasio, in particular Descartes’ Error, which suggests that the subjective valuing of one outcome over another is the means by which human brains evolved to produce action in the world and is thus a fundamental component of instrumental rationality.

            Given we’re social animals, this gets dressed up and reified as ethics to convince other people that they should value the same outcomes you do.

      • So if you prefer murdering someone to injuring them….?

        Your theory has flaws. Morality is about *social* goods, like justice and equality, and it’s about *interactions*, like murder and theft. For those reasons, you can’t satisfactorily explain it as solipsistic preference satisfaction.

        NB not saying preferences completely irrelevant.

        • Deiseach says:

          I think Zebram might say that we condemn murder as “immoral” but that has no real basis. Our animal co-inhabitants of this planet don’t have the concept of “murder”, and in our past as evolving primates neither did we. Killing rivals who compete with us for resources (whether those resources be food, mates, or status in the pack) is a normal part of nature; it is neither objectively nor subjectively moral or immoral, e.g. it is not “better” to injure a rival rather than kill them, as (a) in the wild an injured animal will probably die anyway and will certainly suffer (b) a still-living competitor is still taking resources away from you, so killing them serves your interests better as it’s all about surviving the pressures of natural selection so handicapping yourself by not disposing of rivals is inefficient and foolish.

          The only reason for refraining from killing rivals or only injuring rather than killing them is when the group gets big enough, and develops enough of a social structure, that it is more beneficial to the interests of the group as a whole to preserve individual members rather than letting pack-members or group-members kill one another over competition.

          Stigma against murder is part of social organisation, we assign a moral/ethical value to it in order to inculcate the value of going against our own personal interests in the interest of the group, and as a method of control/coercion over those who might be inclined to ignore the group consensus: do this thing and you will be shunned/punished for violating the norms of the group.

          • stillnotking says:

            Refraining from killing rivals is also a normal part of nature; many species engage in nonlethal dominance contests, including ones that aren’t very social at all (big cats). Presumably this is because a life-or-death contest would carry significant risk for both animals, and could be a Pyrrhic victory even for the stronger. But that’s the Outside View — the Inside View might seem very much like our ideas of “mercy” or “restraint”.

          • houseboatonstyx says:

            @ StillNotKing

            I’d disagree about animals and mercy. I bet it feels more like behaving honorably, being a good sport, not hitting a man when he’s down, and other manly virtues.

          • TheAncientGeek says:

            In summary, you have argued that moral realism is false, that morality has no basis in the non human world; and that morality *does* have a basis in human society.

            But nihilism, the claim that morality has no basis, that all moral statements are false or meaningless, cannot follow from an argument against realism alone….you need to show that alternative bases, such as social constructions, are also inadequate….and that’s what’s currently missing.

    • 1. Is morality something we would want or need?

      2. Could we, would we, construct it, if it doesn’t exist?

      3. Doesn’t morality already exist?

      • Zebram says:

        1. Probably many of us want it to exist. I don’t think there is any such thing as ‘needs.’ Everything is a want. Or it could be a ‘need for a specific purpose’, but not an intrinsic ‘need.’

        2. Not really. We can certainly come up with a bunch of rules that are useful to meet certain ends, like that make society materially prosperous, or ‘prosperous’ in other ways. But that’s not how about 99% of people define morality. To most people, morality does not just mean a social construct.

        3. Many people claim it exists. I see no evidence for it. And when I say that, I am defining morality as actions which ‘ought to be done’ or ‘ought not to be done’ and other similar definitions, not just an arbitrary code that I live by.

        • 1. The idea of an instrumental need, that is required for some specific goal seems quite cogent to me, and so does a need for a whole bunch of desiderata….and it seems to me that morality can deliver a variety of desiderata.

          2. You need to distinguish definitions from theories.

          3. There is plenty of evidence for a wide variety of oughts.
          Ethical constructivism regards morality as a social arrangement that places obligations on people,to refrain from murder, theft, etc, in exchange from not being murder, etc. So moral oughts are explained as a basically ordinary sort of ought, a kind of contractual obligation.

          If you think that meal oughts must be something more special, you need to say why.

          • Zebram says:

            3. That’s not really evidence for an ‘ought’. I’ve never found any evidence for ethical constructivism. I don’t understand your explanation exactly, but it seems you are saying there is such a thing as a contractual obligation, which I’ve never found any evidence for. There is certainly a social construct that we should do what we agreed to, but that is all.

          • TheAncientGeek says:

            There’s evidence for contractual obligation, in that someone who breaks a contract faces sanctions. I don’t think you can talk about obligations in the absence of sanctions, and moral sanctions clearly exist.

            So there are things people should do, and they are punished if they don’t do them. For me, that passes the duck test for obligation.

            Again, it would be helpful if you could say what you think an obligation is….more helpful than repeatedly stating that you can’t see obligation in places where other people can see it.

            I don’t see what you mean by seeing no evidence for constructivism. You mean you see no evidence for defacto moral thinking and behaviour? But there is plenty.

            No evidence that it was explicitly constructed? But most constructivists think it evolved gradually.

    • brad says:

      Nihilism of all sorts is extremely dull. It’s almost always impossible to prove the nihilists wrong, for arbitrarily strong values of prove — which nihilists will inevitably demand, but on the other hand you can’t build on it to anything that is interesting to talk about.

      Nihilist: You don’t even exist, you are a figment of my imagination. Prove to me otherwise.
      Me: I can’t.
      Nihilist: You see, I’m right!
      Me: Bye now.

      • Zebram says:

        If you find it boring, fine. This is why I’m asking Scott to discuss the topic and not you.

      • AnObfuscator says:

        Philosophical skepticism is a real problem, though. “How do you know anything is knowable?” is a hard question to answer, and I haven’t encountered a really solid one.

        The only way I managed to move on was accepting that certain assumptions are just really damn useful, and getting over the whole “unprovability of knowledge” thing.

        However, I do still find it to be a useful “nuclear option” for people who are utterly unable to conceive that their philosophy could be wrong.

        • Nicholas Carter says:

          The Fully Solipsistic Counter-Argument is that:
          1. The Mind Being Formed By It’s Experiences
          2. Awareness Being The Mechanism of Experience
          3. Reality Being A Relative Property
          C. If the world isn’t real, my mind is a product of fiction, and thus equally real. Since I’m concerned primarily with my own tier of fictitious-ness (my utility function is less real that any reality above me in the same way yours is), it’s really quite irrelevant to *me* what hypothetical more real beings might feel about the truth value of my Experiences to their own environment.

    • haishan says:

      “Moral nihilism” is not really a cohesive thing — it’s just the opposite of moral realism. What kind of moral nihilist are you? Like, are you a non-cognitivist, or an error theorist, or what?

      • Zebram says:

        I’d say I fall into the error theorist category. I don’t think the non cognitivist view accurately represents what most people mean when they make moral statements.

  16. Deiseach says:

    What has cheered me up recently in the news is the (so far) good prognosis for the Ebola vaccine. Nice to see the Triumph of Science 🙂

  17. Machine Interface says:

    Book recommendation based on past writings of Scott about the power of rules in the making of organised religion: “Religion for Atheists”, by Alain de Botton, an essay on what a secular-based religion would look like and what benefits it could bring. A bit superficial and lacking in rigor overall, but makes some interesting individual points that are worth considering and might resonate with the readers of this blog (notably a criticism of the current model of education in the west, and an advocacy of meditation and other religion-associated methods of self-care).

    • Shenpen says:

      This is an excellent, but slightly creepy personal take on that:

    • Tarrou says:

      There’s no need to wonder what secular religions look like. Communism in its various forms serves quite nicely, as does our current raft of secularized psuedo-liberal cults of cultural masochism. Politics is most often the vessel for the religious sentiment and need of non-religious people.

      • merzbot says:

        On the “cult” thing: see this post. Also this one. If leftism is a cult, it’s a highly noncentral example of one.

        Moreover, consciously trying to become a less prejudiced person is no more masochistic than any other form of difficult self-improvement, like exercise or intense study.

        • Whatever Happened to Anonymous says:

          >Moreover, consciously trying to become a less prejudiced person is no more masochistic than any other form of difficult self-improvement, like exercise or intense study.

          It also lacks the tangible benefits, though.

          • HeelBearCub says:

            @Whatever Happened to Anonymous:

            >Moreover, consciously trying to become a less prejudiced person is no more masochistic than any other form of difficult self-improvement, like exercise or intense study.

            It also lacks the tangible benefits, though.

            You seem to be taking the position that being aware of and reducing biases has no tangible benefits.

            Steel-manning, eliminating that which might be considered prejudice, but is merely true would not have benefit. I think there is ample evidence that one’s prejudices are quite likely to include items which are not true, however.

        • Cauê says:

          I don’t think he meant “leftism”, but a certain subset of it. And as long as we’re quoting Scott, he apparently agrees, at least in some part:

          • Machine Interface says:

            While liberalism may have characteristics in common with that of a folk religion, and SJW with that of a death cult, the latter seems to lack the benefit of psychological stability and of social support brought by more organised religious movements (indeed, SJW members seem to live in a constant state of fear, anger, frustration and paranoia, provide little concrete support to each other, and spend more time attacking and destroying each other than their enemies), and the former seems to lack the rigor and formal hierarchy associated with more succesfull religious organisations, like the Catholic Church or the LDS Church.

            Thus, while they are “secular religions” in a sense, this is not in the interesting sense, in the way that Methodism or Orthodox Judaism are religions.

          • HeelBearCub says:

            (indeed, SJW members seem to live in a constant state of fear, anger, frustration and paranoia, provide little concrete support to each other, and spend more time attacking and destroying each other than their enemies)

            So, Baptists then?

            😉 if it is necessary.

    • merzbot says:

      The Satanic Temple describes itself as a serious nontheistic, nonsuperstitious religion, if you’re interested. Though I’m not 110% convinced that’s not a lie to make their anti-Christian activism (which mostly takes the form of “if you’re going to do a religious thing on government property, you need to let us do our unpalatable religious things on government property too, so you shouldn’t do them at all!”) more effective.

      • Adam says:

        You could just join the actual Church of Satan, which I’m pretty sure takes that seriously and has existed for fifty years without much in the way of political activity. The Satanic Temple is just the specific branch of it for people that care about political causes and want to troll Christians, but it’s only one branch of the larger church. They’re not even very liberal generally. Core tenets of the Church philosophy include social Darwinism, the ending of welfare and affirmative action, and support for capital punishment.

  18. Are there any EA causes that attempt specifically to increase happiness or reduce pain rather than prevent death? Do current top listed causes/methods (deworming, malaria nets, direct cash transfers) do a good job of making life more enjoyable rather than simply longer-lasting, relative to other potential causes?

    • haishan says:

      Thing there is you pretty quickly run up against the hedonic treadmill, right? At least in WEIRD societies — maybe it’s easier to improve happiness set points in Africa.

      • Lemminkainen says:

        I recall that several years ago, The Economist published an article which included data about GDP/capita and happiness. The two were fairly strongly correlated, although there were a few outliers (East Asian countries tended to be unhappy for their wealth level, while Brazil and Venezuaela were much happier than comparative-rich countries), and the effect seemed to cut off around “as rich as one of the poorer Western European countries.” So, at some point, the hedonic treadmill does kick in, but most poorer countries are far from that point.

    • Roz says:

      Calculations (always?) use a disability weight factor, as well as just looking at number of life years saved. This is kinda about making people’s live happier, and definitely about making them less painful.

      Would be interesting to look at things in the opposite frame, negative disability weights for things that make people happy above and beyond health, though I suspect they wouldn’t appeal to EAs due to low cost effectiveness.

    • haishan says:

      I think the other thing about this is that, if you’re focused on increasing happiness rather than saving lives, EA loses its strongest argument — Singer’s drowning child. Everyone thinks you have an obligation to save the drowning child; (almost) nobody thinks you have an obligation to make a stranger on the street happier, even much happier. (Do you buy antidepressants for strangers?) So you have to really buy in to hedonic utilitarianism — I know you personally do, but I also think you’re wrong.

    • Hedonic Treader says:

      Are there any EA causes that attempt specifically to increase happiness or reduce pain rather than prevent death?

      Yes, the Foundational Research Institute is focused on reducing suffering. Not sure how effective they are.

      An alternative is going vegan (or lacto-vegetarian with reduced lacto) and spending more money on yourself in return. This effectively reduces the amount of pain without trusting other people to do it for you.

    • Leo says:

      Deworming is clearly all about quality of life. Having no worms is pleasant and possibly makes you go to school more, do better there, and have a more successful life later. But worms almost never kill you, so deworming won’t make you die less.

    • FacelessCraven says:

      I’m very sure this doesn’t count as Effective Altruism, but just for a bit of perspective, Child’s Play seems to be about increasing happiness and reducing pain.

  19. hidenameortheyllgetme says:

    You forgot (?) to post your usual topics not to be discussed so let me tell you of my theory about how third-wave feminists created Dogecoin in order to hide IQ differences between dog races…

  20. Tracy W says:

    Disgusting topic, is there any real science about any connection between how your poo looks (as opposed to constipation or diarrhea) and your overall health?

    • Acedia says:

      Color can be indicative of a number of things, I believe. Here’s paper from this year about its usefulness in screening for biliary problems:

    • from my knowledge, unless it’s pale, like tar, or red..color is probably not a concern

    • roystgnr says:

      Black stool can be an indicator of colon cancer (i.e. “You might die soon.”) and/or internal bleeding (i.e. “You might die very soon. Go to the emergency room. Now. Why are you even still reading this?”). This caused me some brief panic a few weeks ago until I discovered that a toddler’s black stool can also be an indicator of pigging out on blueberries.

    • Tracy W says:

      Thanks everyone for their answers. So colour is an indicator, but not consistency? (Like Roystgnr, I spend quite a bit of my life right now looking at toddler poo).

      • Pku says:

        For some weird reason, this made me really want to have kids (I’ll get to look at poo ALL DAY!)

        • Anthony says:

          If you have kids, or dogs, you will quickly discover you can tell which crayons were eaten by observing the poo.

      • Anthony says:

        Consistency is an indicator of things which don’t matter much. Unless there’s lots of diarrhea, in which case there may be dehydration, and the diarrhea might be caused by something nasty. But one semi-liquid movement just might mean something upset the digestive system a little.

  21. Oleg S says:

    Hi all,
    Could anyone point me to comprehensive and reasonable ethical (non technological) objections against cloning and usage of CRISPR technologies in humans?

    • Ever An Anon says:

      No, but I can give you the five-second version.

      Remember all those magazine ads from the turn of the 20th century featuring radium-based “medicine” that make you cringe uncomfortably thinking about three-headed babies? Well in a sense the guys pushing that snake oil were on to something, after all radiology is a very important part of modern medicine. But the technology to use radioisotopes medicinally just didn’t exist yet: the best science available at the time still led the Curies to early graves.

      A lot of what we’re doing in the lab today looks like what radiation looked like in the 1910s or 20s, albiet likely safer for ourselves. We have had promising breakthroughs but a lot of the fundamentals are just not here yet. Right now anything more than simply diagnosing or predicting genetic problems is a risky proposition, and that doesn’t seem likely to change soon (although obviously I could be mistaken, it’s hard to predict the future).

      • Oleg S says:

        Ok, technology may be too immature for large scale clinical / market applications. What about research then? What are the most pronounced ethical objections agains basic research of human cloning in western world?

        • Ever An Anon says:

          Depends on what you mean by research and cloning. Not trying to be glib but I’m not sure what exactly you’re thinking of here.

          If you mean researching how to clone animals increasingly similar to humans, then we’re doing that already AFAIK. I’m not really interested in cloning personally, so not up on the latest research, but I recall that there’s a lot of money in agricultural cloning (cows and such) so presumably some of that is trickling down to basic research on mammalian cloning.

          If you mean “trying to clone a human just to see if we can” then you’re deep in mad scientist territory. Even tried techniques often fail in the lab, and in this case failure means dead or disfigured kids. Waiting at least until we can reliably clone nonhuman primates before trying would be prudent.

          • Oleg S says:

            I mean not cloning whole organizms (indeed, we are not ready for that yet), but but making experiments on non-implanted human embrios to study possible mechanisms of failures of CRISPR/cloning.

            From utilitarian perspective experiments on human cells may provide great benefit because of possible future applications of these techniques to prevent genetic diseases. One can argue that experiments on fertilized human eggs are more ethical than attempts to clone primates. So, what sort of reasoning officially prevents scientists from making experiments on human embrios?

          • Adam says:

            The Dickey-Wicker Amendment already forbids research that requires destroying human embryos, at least research funded by DHHS.

          • Oleg S says:

            Thanks! The Report of the Human Embryo Research Panel, cited as a foundation for Dickey-Wicker Amendment is exactly what I searched for.

    • Irenist says:

      @Oleg S:

      Are you a utilitarian, a consequentialist, or what? I think “reasonable” is going to depend a lot on what your ethical stance already is.

      • Oleg S says:

        I mean the critique and anti-cloning arguments that sound reasonable enough to be heard by policymakers. Something a member of comitee on research ethics can cite and use.

        • Adam says:

          The history of U.S. policymaking in this arena suggests that most research injunctions had been religiously motivated at least through the late 90s, though it’s increasingly shifted to left-wing quackery about Monsanto taking over the world. In neither case was ‘reasonable’ a good measure of whether an appeal would succeed.

  22. David Moss says:

    Scott, you’ve talked before about the way in which post-Ferguson white people’s views about the police/race actually improved (diverging from black attitudes which of course declined) and I’ve found that example/graph useful too.

    So does this change your views on this issue at all?
    [gallup poll showing that white’s have also increasingly judged that black are treated unfairly in a variety of spheres in the past couple of years]

    • whateverfor says:

      Well, it’s not like the cops stopped killing black people. Of all the “bad cop” stories to hit national news in the last five years, Ferguson was probably the one where the cop was most justified in his actions. When Ferguson was the chosen pushed story, people rationally thought if that was the worst it got there wasn’t actually a problem. Then the Gray/Bland/DuBose/etc. stuff happened and people updated.

      • David Moss says:

        That’s certainly plausible. I had thought though that part of the reaction against Ferguson was due to the issue becoming tribally politicised, so all the loud campaigning had a backlash effect and polarised opinion. These stats seems to count against that, since these cases have still been part of a heated, politicised debate, and yet white opinion seems to have largely shifted in the same direction as black opinion.

        • Tarrou says:

          I think there’s a sub-category here. There’s a very large coalition across the political spectrum pushing against overaggressive policing. Almost everyone is in on it, but as with the tale of the elephant and the blind men, they all see different parts of the problem.

          Social conservatives have been the targets of seemingly punitive “John Doe” investigations and SWATTing in recent years, and so have become very wary about the overreaches of prosecutors and police.

          Libertarians have always hated the over-militarization of police and the drug war, and have much stronger stances on civil liberties than even liberals.

          Liberals have always wanted less criminal penalties for just about anything you can think of, unless it’s perfectly legal and constitutional political speech by people they don’t like.

          Blacks have always born a disproportionate brunt of policing (you can argue over whether this is because of evil racisms or a higher offending rate, but it is incontrovertible), and mistrust of and antagonism toward the police is a basic part of black culture now.

          So pretty all the major groups in politics would like the status of police to be lowered, and the ability they have to project force to be curtailed. The only thing stopping this from happening tomorrow is the fact that none of these groups is willing to hand the other a “win”. Blacks and liberals will not, under any circumstances, countenance a restriction of the police unless the reason is racism. And conservatives and libertarians won’t* unless the reason isn’t racism.

          And for all the publicity over so many questionable uses of force, there has yet to be a single credible allegation of actual malice or racism on the part of the cops. It’s all just assumed because the cops are white (ignoring all the instances when the cops aren’t white or the victim isn’t black or brown).

          *Rand Paul is the exception here, which may signal which side is going to cave.

  23. Mark says:

    I don’t want to work and feel as if maybe I am missing something. It’s one of those things, like, “am I mad, or is it everyone else?”
    It just seems so pointless – spend eight hours a day working on a check-out when the work can already be automated – drive a bus all day – sell people things they don’t need – design rubbish that stops people from thinking about and doing things that are really important.
    I want to get a job for the social aspect, but the work itself is just mind killing.

    I mean, what are the actual important, good jobs, that need to be done and can be done by a normal person? That don’t require a vast fortune and years spent on training? That aren’t closed to outsiders?
    What should I be doing to actually improve the world?
    (The price mechanism for labor isn’t working.)

    I am middle aged, and have a reasonably good mind (no genius), but I don’t like computers, have a low tolerance for bullshit (corporate and small talk), am not especially personable, and like taking afternoon naps. What the hell am I supposed to do?

    • Saint_Fiasco says:

      >What should I be doing to actually improve the world?

      You should get the most high-paying bullshit job you can find and then donate to an efficient charity.

      • Mark says:

        What is the best way to get a high paying bullshit job?

        • Saint_Fiasco says:

          Depends on the market. You don’t like computers and are not a genius, so perhaps one of the trades?

          Then again, I’ve heard horror tales about firemen requiring college degrees and other barriers to entry in America, so perhaps you could move somewhere else?

          • Godzillarissa says:

            If you’re already suggesting a bullshit job, why is dislike for computers still an argument?

        • bluto says:

          Get into the finance industry, by hook or by crook.

      • Murphy says:

        I’m guessing that lack of fuzzy points to accompany the utilons would be a problem with that if personal fulfillment is wanted.

    • Murphy says:

      It depends on your skills but it’s pretty tough to find high paying, fulfilling, high impact jobs with low barriers to entry.

      What are your skills/interests/etc?

      • Mark says:

        This sounds really awful, but I don’t have any professional skills (nothing special anyway). I’m kind of trying to work out what job I should go into and then work on getting the skills for it. I’ve basically always worked in customer facing bullshit jobs, but I don’t actually think I’m very well suited to that kind of role and have sort of developed a severe dislike of (unnecessary) speaking because of it.
        Interests – uh… I like sitting on grass, looking at the sky etc. I think I have normal interests plus I like reading and ideas…

    • You can do socially useful things by volunteering, or taking a minimum wage care worker job. It sounds like you have some requirements about remuneration.

      • Mark says:

        Yeah… I think I need to get around to trying this, at least for one day a week or so. Thanks for the suggestion.

    • NZ says:

      You should do carpentry at home, and then sell the pieces for a lot of money. Then, as other commenters suggested, you can donate the money to whatever cause you think is worthwhile.

      Carpentry is meaningful in itself–you’ll get a spiritual buzz out of designing and building things with your hands.

      • Nornagest says:

        You should not expect to sell home carpentry for a lot of money, especially relative to the amount of time it takes, unless you’re willing to spend a lot of time and effort marketing yourself or are lucky enough to have a ready market already for some other reason. It helps to have been doing it for many years, too, but that’s not strictly necessary.

        On the other hand, it is pretty easy to use it to save a lot of money that you would otherwise spend on furniture, if you are the kind of person that wants furniture originating somewhere other than the thrift store, the side of the road, or Ikea. It helps that the shabby-chic Pinterest look is popular right now — the materials for it are dirt cheap.

        • NZ says:

          When I said “a lot of money” I really just meant “enough to cover your costs and meet your needs.”

          I think unless you’re really just a noodlehead, dedicating 8-10 hours per day to learning carpentry will almost certainly result in craftsman-level work within a year or two. The learning tools are all available for dirt cheap (e.g. carpentry books abound at garage sales and cost about a quarter). The actual tools can be pretty cheap too especially if you get them second-hand.

      • Mark says:

        The idea of carpentry is attractive, I’ve certainly thought about it before, but as a *complete* beginner, I’m not really too sure about how to start getting into it. Should I just go and buy some bits of wood? It is one of those things that I’ve always wanted to try but have never really had the motivation to get started with.

        • bluto says:

          One of the best first projects in carpentry is making tools (really basic might be a table saw pusher or a popular one is making a surface plane and working up to benches and vises). These allow a new carpenter to get practice, make something useful even when it’s a bit roughly made, and give the maker much feedback about what works and doesn’t because they’ll be using the product so have good ideas about how to improve future projects. Finally, the materials are frequently less expensive than commercial tools.

          If you live anywhere with trees, there’s almost always scrap wood available frequently for the cost of moving it (and scrap wood may be available elsewhere).

        • houseboatonstyx says:

          In my remote area of retirees, there is instant money well over minimum wage for very basic handyman work, even with limited tools. There is skilled carpentry work available as you get more skilled.

          This is a lovely peaceful area with population/income too small to support a big deal local service — or to pay someone from town maybe $80 just to drive out here.

        • NZ says:

          “Should I just go and buy some bits of wood?”

          Yes. Your local hardware store or hardware megastore will have scrap lumber for cheap (something like 80 cents a plank). Buy some of these, choosing the straightest planks you can find.

          For about a hundred bucks you can get a hand saw, a power drill, some drill bits, some sandpaper, and some wood screws. You can bring that price way down by searching garage sales or Craigslist, or by informing older male relatives that you’re interested in woodworking and see if they just give you some of their tools.

          Also, like I said, at garage sales you can pick up some carpentry how-to books for about a quarter each. More carpentry how-to information is available for free online, if you don’t mind putting up with a computer for an extra bit of time while you’re reading it.

          First, build the simplest thing you can think of that’s still useful. A box, a bookend, a couple dowels on a base to hold dirty boots, that sort of thing. This will give you a feel for how wood responds, it will give you practice using the tools and how they fit in your hand. Don’t worry about making the things fancy, but do try to make each thing as well-built as you can. If you screw up that’s good–it means you’ll learn better the next time. Screwing up on some small unimportant projects is the best way to do it.

          If you keep doing that, even for just 10-20 hours a week, you’ll be a much better-than-average carpenter than most people within a year or two. People will be asking you to build them bookshelves, desks, or even hand tools. You’ll get to enjoy working by yourself, on your own schedule, with your own hands. You’ll be creating real value by building beautiful, unique, and useful objects. You’ll get to smell freshly cut wood all day. You won’t need any computers. It’s ideal, really. And your investment in tools and materials can always be incremental so that aside from that first hundred bucks you’re always in control of your overhead.

    • moridinamael says:

      Unless you subscribe to some specific philosophy that imbues acts with satisfactory meaning, everything is pointless, including both work and leisure. If you can’t find any form of work which feels meaningful to you, then perhaps you need to upgrade your skills in finding or applying meaning to actions and experiences.

      If you for some reason believe or feel that this approach wouldn’t work for you, then just forget about meaning and try to find enjoyment. Work is actually fun for most people, or can be. Think about The Office versus Office Space – two comedies specifically about practically the same setting, one emphasizing the humanity and community and simple enjoyment of professional life, the other emphasizing the emptiness and futility of it. It’s entirely up to you which perspective to take.

      • Mark says:

        Wow… I thought they were *both* about how dreadful work is! Maybe that says it all about me…(I’ve only seen the UK version of the office though )

        There are lots of things I do find to be meaningful, but just not anything that anyone has ever paid me to do – I feel tremendously happy to spend time with family and friends, to read, watch tv, drink beer, etc. etc.
        I just get the feeling that with work, at the moment, the sort of “economically liberal” idea that pay will express information about what people actually want requires that we aren’t all just following the same social signals off a cliff. That’s what it feels like to me, when I am at work – that a horrible social monster is eating my life. And I don’t want to do it.

      • Marc Whipple says:

        Work you enjoy is one of the best kinds of play.

    • Urstoff says:

      Step 1: Chillax, bro
      Step 2: Re-evaluate previously held contempt for life

    • Adam says:

      I largely agree with you. Aside from a tiny number of companies building better medical devices and more intelligent home-cooling systems and what not, very nearly all commercial activity is a process of brainwashing people into feeling inadequate if they don’t purchase something they didn’t need until you brainwashed them into feeling inadequate without it, and that doesn’t seem like a worthy activity to engage in.

      On the other hand, the available evidence suggests that the single most important thing to objective measures of quality of life, aside from basic enjoyment, is an economy that grows faster than the population, and it doesn’t seem to matter much what exactly drives the growth. Even asset bubbles work until they pop. If only we could figure out basic enjoyment.

      Personally, I try not to worry too much about meaning. The lowest-paying job I’ve ever had in my life was detailing restrooms in the middle of the night at a theme park, but it was arguably also the most enjoyable. No supervision, I only had to work until I was finished but got paid for eight hours, free admission to the park, immediately tangible results to my work. If only age didn’t come with larger bills.

    • Tracy W says:

      Work as a nurse’s aid at a hospital or a residential home or the like. What you do has a direct impact on people’s happiness. It’s often directly unpleasant, (eg cleaning up poo) and you see a lot of sad stuff, but on the other hand you can work the morning shift and get your afternoon nap.

      Source: I worked doing this one summer at uni. I was at a secure unit for people with behavioural problems so I got attacked a few times. They liked to employ men as the greater physical strength helped with moving patients. I’m not particularly personal but after a few weeks I built up some relationships with some patients that I then was mostly assigned to, which was quite normal, everyone had particular people they got on well with.

    • Nicholas Carter says:

      Don’t increase the pressure, lower the resistance. What we’re looking for here is a way to get your living expenses down below $12,000 a year. Then you can do just about anything a little bit, but I recommend substitute teaching if you aren’t opposed to noise and a little bit of substance-less power exercise.
      But no, for the most part, I do not like doing the things I have convinced people to pay me profitably to do, although I have had some success with payment in kind, favors, and barter.

    • Anthony says:

      Become a plumber. You’ll be improving the world, one bathroom at a time, and making decent money doing so.

    • Loquat says:

      This may or may not be a good solution for you personally, but any job that involves educating others is IMO useful to the world. I’m currently doing health insurance sales and service*, and while the service work could mostly be automated, much of the sales work involves explaining to people how the system works and what kinds of options they have – and it’s not like nobody’s tried to automate that, there’s plenty of good information on government healthcare sites alone, but a substantial percentage of the population would really prefer to have it explained by a live person so they can ask questions.

      *In the US, specializing in Medicare, if anyone’s wondering. Medicare could really stand to be simplified, particularly the prescription drug benefit which was designed in a completely ridiculous way to begin with, but it keeps me in a job so I won’t complain too much.

  24. merzbot says:

    Do you guys have any strategies for disciplined self-study? I’ve been trying to teach myself some calculus but I’m really not getting anywhere without the structured nature and mandatory work of an actual class.

    • Mark says:

      Cancel your internet and sell your TV.

      • merzbot says:

        Thanks, but I can’t realistically cut all distractions from my life. I’ve tried the whole internet blocker thing; I just make excuses to myself to disable it or find other ways to distract myself. What keeps me focused on coursework is “oh shit if I don’t do these problems I’m going to fail the exam I better work my ass off on this instead of reading blogs,” not a complete lack of distractions. I think I need some way of replicating that.

        • Mark says:

          I’m not sure about calculus alone, but you can often pay to take examinations (without having to pay for tuition) with professional associations/ correspondence universities. I know that the international program of the University of London has some calculus exams in its Maths and Economics course, but it is quite expensive. (I only know that because I signed up for the exams in a failed attempt to motivate myself to study for them.) They also do a Mathematics diploma for graduates.

          Yeah – I haven’t had much success with internet blockers either – the only time I’ve managed to successfully self-study for a prolonged period was when I moved house and didn’t sign up for the internet for six months. I made a note of anything I wanted to look up, and then printed it out the next day at work!

        • Murphy says:

          Depending on your personality type finding someone to compete with on a friendly basis doing things complementary to the course work can help. It worked for me in uni anyway.

          Picking some projects of your own, again related to the coursework but of your own choice helps. Combine with the above for best effect.

          I found informally tutoring classmates helped me as well as I’d keep ahead myself for the sake of that.

        • Adam says:

          If the problem is really you can’t do it without an actual class, take an actual class. Nearly every community college in the U.S. at this point offers online Calc I and Calc II for a few hundred bucks.

        • H says:

          Displace yourself. Go live somewhere else for a while, bring only a backpack. It doesn’t have to be miles away. Force yourself to change your habits, or obliterate them all together. It’s the only thing that works for me, anyway. Change is so refreshing.

        • Wrong Species says:

          I’ve been thinking about how to get more motivated when I get back to school. One of my problems with doing schoolwork is I know that it will only affect me so I just don’t do it and then fail. However, if it’s a group project then I will put it in my best effort because I don’t want to let anyone down.

          So I had this idea that I could take a certain amount of money and give it to charity every month. Every time I don’t do an assignment on time or blow a test, then I have to give a little less money to the organization. So if I decide to give $50 to deworming each month, and then fail a test then I have to deduct 20 dollars from the amount I give. From an EA perspective, it’s terrible to let people die just to get yourself motivated but it’s better than not giving anything. Anyways, you could easily adopt that to help you study on your own.

    • as Mark said, kill the distractions

    • Whatever Happened to Anonymous says:

      Is convincing a friend to do it with you out of the question?

    • Brock says:

      The Coursera Calculus One and Calculus Two MOOCs are fantastic!

    • Alexander Stanislaw says:

      Why do you want to self study?

  25. Error says:

    I got imprisoned on TVTropes yesterday and ran into Don’t Shoot the Message. It feels like a perfect summary of how Scott seems to view the social justice movement.

    I’m not really going anywhere with that; I just wanted to share the connection.

    • Urstoff says:

      So basically every Aaron Sorkin show? Or, rather, how a smart progressive must feel about every Aaron Sorkin show?

      • ddreytes says:

        Oh, good lord, yes. Those aspects of his shows frustrate me to death. And the really annoying thing is that I still get drawn in by the dialogue and the characters, which I really like.

        So basically, yes, I have a complicated relationship with Sorkin’s shows. A well-chosen example.

    • stillnotking says:

      That’s certainly true for me. And Jerkass Has a Point pretty much sums up my attitude to neoreaction.

  26. Chalid says:

    Does anyone have suggestions on how to navigate these enormous comment threads?

    For example, if I look at the site when there are 300 comments, and then come back when there are 320, is there a good way to find which are new? I realize that the new ones are in green but I don’t see a good way (short of endless scrolling) to find them. And if I use the comments RSS I lose the thread structure.

    • Zykrom says:

      look in the top right of the screen, also ctrl+f searching “~new~” works

      • Chevalier Mal Fet says:


        I’ve always been either scrolling looking for green outlined boxes, or else manually searching for the names of posters who the top right of the screen has identified as posting recently.

        This makes everything so much easier! God bless you and your progeny.

        • Emile says:

          or else manually searching for the names of posters who the top right of the screen has identified as posting recently.

          Note that you can also click on those names (it’s a list of recent posts, not of recent posters).

      • Chalid says:

        Thank you!

  27. Emily says:

    My usually relatively non-outraged/not-that-political facebook feed now has multiple posts about abortion from people whose understanding of the universe would, I feel, be improved by reading Scott’s “FETAL ATTRACTION: ABORTION AND THE PRINCIPLE OF CHARITY” post. However, my life improved somewhat when I stopped getting into fights on facebook, and I’m not sure I can suggest this without starting a fight. Oh, well.

  28. Brock says:

    To what extent did the proliferation of pornography on the internet – specifically girl/girl porn aimed at a straight male audience – contribute to the mass change in attitude toward homosexuality we’ve seen over the past 20 years?

    Should same-sex marriage supporters – and I count myself among them – be sending a thank you note to the Hollywood porn industry?

    • Jiro says:

      Why be specific to girl/girl porn?

      You can find tons of material (much of it will be written, but it still counts for these purposes) involving male/male pairings for a female audience.

      • Brock says:

        I’m thinking that girl/girl porn aimed at men is much more popular than written “slash” erotica aimed at women, but I might be wrong.

        I’d hypothesize written “slash” erotica would have the same sort of effect, in proportion to its popularity.

        • Jiro says:

          I have seriously considered the idea that slash leads to, not acceptance of homosexuality, but acceptance of porn by feminists. It’s really hard to read about Harry Potter taking it up the ass from Draco Malfoy and then complain that porn objectifies women.

          • Eggo says:

            Oh trust me, they manage it. The best tumblr comeuppance is posting their own shota-rape-piss-porn in response to their anti-porn-made-by-men rants.

    • Saul Degraw says:

      My guess is little to none. There is a lot of historical evidence that men (including men in charge) have always found lesbianism to be less threatening than male homosexuality. Jews considered lesbianism a lesser sin because female-female sex did not waste semen or eggs (hence it did not violate “Thou Shall Be Fruitful and Multiply). Lesbianism was legal through a loophole in pre-Weimar Germany because the male legislatures simply couldn’t conceive of it when they wrote the penal code section against homosexuality.

      I think that people born after 1975 simply grew up in a world where homosexuality was more open and became more open as time went on. I think MTV helped here especially the second and third seasons of the Real World. The second season featured an openly lesbian (and quite butch) woman and her friendship with a southern conservative guy. The third season featured an openly gay men and AIDS educator and showed his friendship with a conservative woman from Phoenix. The producers also showed him cooking with his lover in a domestic scene and a commitment ceremony. In 1992 and 1993, this was presented in a matter of fact kind of way and that was quite and quietly radical. I think that people between 10-15 must have been influenced by this especially.

      • John Schilling says:

        This. Note also that we discussed Jeremy Bentham’s late eighteenth century defense of homosexuality here a while back, which offered no defense of lesbian relationships because it did not so much as acknowledge the existence of such relationships. This at the same time and in the same country that the Ladies of Llangollen somewhat infamously cohabited in what certainly looks like a lesbian domestic partnership. Not long after, the term “Boston Marriage” was invented for such relationships, and while it carried a hint of scandal and disapproval, there was little overt repression.

        Not clear whether people were blind to the fact that women in such relationships might engage in carnal as well as boringly domestic pleasures, or simply didn’t care. But it’s pretty clear that nobody was intensely threatened or disgusted by female homosexuality, as has generally been the case for adult male homosexuality. Porn exhibits the sexual side of female homosexuality in a manner specifically calculated to not induce disgust or fear, and thus changes nothing.

      • anonymous says:

        Indeed the human mind seems to be wired in such a way, that even today, lesbianism is less likely to awaken that whole set of feelings such as jealousy and disgust or fear.

        I think this is the reason that many, probably most men aren’t the least bit jealous if their woman has an affair with another woman (as opposed to another man), whereas most women become jealous if their man has an affair with anyone, man or woman.

        Similarly: I heard that among polyamorous lesbians while it might be given for granted that one’s female partner has a right to have other female partners, many have problems with allowing male partners. Also many monogamous lesbians only want to date women who define as lesbian and not bisexual.
        Whereas I never heard of gay men forbidding their partner specifically to date women, or having problems because their partner defines as bi.

        Another thing regarding lesbianism, many people in the past were unable to conceive of it because of the definition of sex as penetration. By that definition, what does lesbian sex even mean?

        To ask which comes first, the (still present) view of sex as penetration, or lesbianism being less likely to awaken fear and jealousy, is a chicken and egg problem.

        • gattsuru says:

          Whereas I never heard of gay men forbidding their partner specifically to date women, or having problems because their partner defines as bi.


          Maybe it’s just the specific circles I walk in, but at least within the parts of the gay community I’ve seen there’s a much stronger presumption of monosexuality than of completely closed relationships. It’s not the most common practical issue for some complicated reasons, but fear of a man leaving them for a woman is definitely a thing.

    • Wrong Species says:

      I’ve been watching Friends recently and the one thing that amazes me is that this show is 20 years old and there are more gay characters than black characters. And this is right before internet porn became big or at least around the same time.

  29. Odoacer says:

    I’ve been reading a bit about the cardinal and heavenly virtues recently. It seems to me that quite a few would not be considered virtues by modern secular society*, e.g. patience, chastity, humility, and temperance. What traits do you think would be considered modern virtues?

    *Different modern subgroups have different virtues. E.g. diligence is highly regarded in many subgroups.

    • Irenist says:

      It seems to me that quite a few would not be considered virtues by modern secular society*, e.g. patience, chastity, humility, and temperance.

      This is a little odd coming from me (a theist social conservative), but I’m not sure that statement is fair to modern secular society. I think patience (i.e., forbearance under provocation) and humility (i.e., non-boastfulness) are uncontroversially popular in most quarters.

      Chastity and temperance seem to me to have strong secular equivalents, even if their traditional understandings are no longer in favor. Thus, a modern secular version of “chastity” would include being responsible and careful about STDs and unplanned pregnancy (particularly while still in high school or college) and about being scrupulously observant of social norms around consent. It would also strongly distinguish between open polyamory and adultery, with temptation to the latter calling for exactly the virtue of chastity, whatever you’d call it.

      Likewise, although secular moderns don’t worry about employing the virtue of temperance to avoid the sin of gluttony, they very much do worry about both healthy eating (including not eating to excess and avoiding eating disorders) and about ethical eating (including concern about animal welfare, organics, fair trade, and the climate impacts and global economic sustainability of meat-based diets). As for temperance w/r/t alcohol and other drugs, although secular moderns are relatively indulgent about people getting buzzed at parties, secular moderns are very concerned about drunk driving, drug and alcohol addiction, the effects of second-hand tobacco smoke, etc.

      So although the old names might not have quite the right associations, I think the ideas that mature adults should be responsible about their sex lives and their eating/drinking/drugging habits are still pretty ubiquitously held (although of course ideas about what constitutes acceptable behavior in these areas is very different between modern seculars and the old religious traditions). Maybe “responsible sexuality” and “ethical eating” are better modern secular names? But the constellation of concerns seems pretty related to the field covered by the old virtues, even if modern seculars aren’t worried about pre-marital sex or anything like that per se.

      • Odoacer says:

        Your reply is interesting and has caused me to rethink things. People are still very concerned with sexual and eating behaviors (their own and others’). However, e.g. for temperance, there are large and celebrated displays of gluttony (Man vs. Food, eating contests, excessive enjoyment of food by foodies). So, I think your new term of “ethical eating” would be a better fit for a modern virtue.

        • Tamar says:

          On the contrary, I think the “large and celebrated displays of gluttony” indicate that gluttony remains a modern vice. It’s interesting, engaging, titillating because it plays with the temptation to overindulgence and in some respects crosses a line – it indicates that the line is still there. There are probably good examples of line-crossing indicating the boundary existing for the modern values of chastity as well (probably, alas to bring it up, Fifty Shades of Grey? I haven’t read much outside of chapter-by-chapter snippets, but it toys with dangerous/dubious sex where good consent practices are not employed, but where the protagonist eventually is vindicated in her desire for love and connection in the context of that sex? And also the series that controversially was softcore porn that you might see read by someone at the train station or some equally public venue).

      • Leo says:

        You value the virtue of chastity, and agree that having promiscuous sex but only as permitted by the rules of your relationships constitutes an attempt to practice that virtue. I am going to read your entire blog.

    • Adam says:

      Irenist just said it, but those are all still secular virtues. In fact, they’re largely Aristotelian virtues originating from secular thought in the first place.

      • Nick says:

        Yeah. It’s probably worth taking a look at Brandon’s Virtues and Vices series, which usually if not always discusses ancient or medieval consideration of them. Aristotle’s, Cicero’s, and Aquinas’s treatments, each if any, are mostly commonly discussed. Aquinas, of course, has the only religious treatment of the three.

  30. NZ says:

    Has anyone looked more into the concept of Amistics (which anyone who’s read Neal Stephenson’s “Seveneves” should be familiar with) beyond the definition someone put up at Urban Dictionary? Does Amistics relate to any already-studied concept?

    David Friedman, if you’re alive, maybe you’ve got something…?

    • brad says:

      Aside from the obvious Amish connection, the other historical incidents that concept tickled in my head was the Ming Dynasty turning away from its preeminent navy, and 17th century Japan turning its back on the gun.

      • NZ says:

        Yeah, similar examples are easy to find. Actually, every culture, even every household, rejects some technologies and embraces others.

        I’m wondering if the study of how these decisions get made has been articulated further than just the few mentions it gets in “Seveneves.” For example, has anyone really studied whether the Amish have a consistent method for making these decisions? I know they meet and decide what technologies they will embrace or reject, but how are those meetings structured? What kind of questions do they always make sure to ask about a given technology? That kind of thing.

        • Seth says:

          A while back, I was looking into the way the Amish make those decisions, from wondering about steelmanning the weak argument of “If you don’t like modern technology, go live in a cave and wear animal skins”. Apparently the Amish have factions and huge debates over the topic.

          Here’s the best explanation I found:

          “Amish churches regulate use of technology through a set of oral guidelines known as the Ordnung. Amish leaders aim to slow or prevent change if a given technology is seen to be a threat. …”

          “Amish use technology selectively, thus not completely ruling it out. Some Amish disagree on whether—and how much—use of certain technologies is actually necessary, however. One sees these disagreements reflected in both individual opinion and across church districts and affiliations, resulting in a variety of practices. But the idea that the Amish shun technology and live as people did hundreds of years ago is incorrect.”

  31. walpolo says:

    Cryonics fans, has anyone thought much about ways to create incentives for future people to revive you?

    I’m thinking it’s not a great idea to be the first human being they attempt to revive. Odds of a fatal mistake will be much reduced if they’ve already successfully performed the revival procedure (whatever it turns out to be) on someone else.

    Maybe one could set up a perpetual trust designed to automatically pay out a reward to someone who attempts to revive you if at least one (or several?) other people have already been revived successfully by the same procedure, with a higher reward if they succeed in your case, obviously.

    Has anyone ever blogged about this question? It seems to me like an important one for the cryonics crowd.

    • Samuel Skinner says:

      “There is a monster out there. Someone will have to distract it so we can kill it. Let me roll around in BBQ sause before the meeting where we decide this question”.

      Sorry. Who should be sacrificed has an unfortunate shelling point in the person posing the question. More seriously, we will probably experiment on animals first.

      • walpolo says:

        Sure, but you’ll never know whether the memories of an animal subject are properly retrieved in a revival, for example.

        • Pku says:

          You could try, say, training a dog to do some tricks and see if he remembers them after unfreezing (not perfect, but you could probably assure at least partial memory retention that way).

          • Emile says:

            They tried that on worms:

            Can memory be retained after cryopreservation? Our research has attempted to answer this long-standing question by using the nematode worm Caenorhabditis elegans (C. elegans), a well-known model organism for biological research that has generated revolutionary findings but has not been tested for memory retention after cryopreservation. Our study’s goal was to test C. elegans’ memory recall after vitrification and reviving. Using a method of sensory imprinting in the young C. elegans we establish that learning acquired through olfactory cues shapes the animal’s behavior and the learning is retained at the adult stage after vitrification.

            It could probably also be tried with the frogs that survive freezing.

    • stillnotking says:

      One of the reasons I’m skeptical about cryonics is that I can much more easily imagine us being revived as slave labor or research subjects than as fully-empowered citizens of the future. Corpsicles will have neither a political lobby nor a particularly compelling moral claim — they’re already dead, after all. Human nature being what it is, an ordinary dead guy’s chances of being revived from beneficent motives don’t seem good. And if human nature will have changed significantly for the better, why conjure up a horde of atavistic barbarians to make life difficult? If you think immigration is a thorny question now

      If our descendants just want to know what life was like in the 21st century, well, all those Facebook pages will still be around somewhere.

      • Emile says:

        One of the reasons I’m skeptical about cryonics is that I can much more easily imagine us being revived as slave labor or research subjects than as fully-empowered citizens of the future.

        I’d find the “slave labour” thing quite unlikely, or at least for what we would think as slave labour: it’s much cheaper to produce new humans than it is to thaw and revive corpsicles. And if uploading exist, just copy-paste an existing one.

        Research subjects, or “novelty pets”, that’s more likely, but on the other hand, it beats rotting to little bits, especially since you’d still be pretty likely to have standards of living better than today, even as a research subject locked up in a lab (or in a simulation).

        The status of revived corpsicles shouldn’t be compared to those of “fully-empowered citizens of the future” (I agree there are some chances it will be a special lower status), but to death.

    • tgb says:

      You’d probably be OK. I suspect there’d be a large incentive to use a recently cryonified subject rather than one of the earliest ones, due to technical details. This leads to the idea of intentionally sabotaging your own preservation in a small way just so that it would be a poor choice to experiment early techniques upon.

  32. Mark says:

    I have discovered the following fact, but I’m not sure if it has any wider implications. Does it?

    The experience of time as events occur is a universal constant (it is the same for everything experiencing time).

    When someone wishes to represent life as it appears to a fly, they show you a video in which events are slowed. If I were to suddenly experience life as the fly sees it, I might say that “time slowed down”, and what I would mean by this is that external movements appeared to slow – for each beat of my internal clock, things moved less of a distance ( since *all* external movements have changed to the same degree, it is only with reference to some internal clock that anything could have changed at all (and something clearly has)).

    If you measure things in comparison to external events, it is possible to say that the internal clock has sped up – “15,000 perception unit flows per hand wave” – but from the perspective of … perception itself… the person experiencing those perception units – there is only ever a *number* of internal clock ticks (that feels as it feels) and then some corresponding external events.

    I have 6 perception units and you have 6 perception units – no matter what those units have enabled us to see, or the speed of the things going on around us, we have experienced the same amount of internal time.

    Things slow – time is constant.

    • Nate Gabriel says:

      Seems unlikely. Trivially, if we live the same length and have the same amount of internal time, and then one of us lives another year, the perception of that year isn’t zero.

      • Mark says:

        No, I don’t mean to say that everyone will always have the same amount of perceived time in their life, rather that the rate (this is a bad word here, because the whole point is that there is no “time” to the perception of time) at which people perceive time must be the same. If I have six units of perception and you have six units of perception, from our perspectives we have experienced the same amount of time even if you are a fly and I am a human. Nothing happens to our *internal experience* of time – external events just get faster or slower.

        • Adam says:

          Have you looked into tachypsychia at all? It’s related to the way people claim that time slows down when they are faced with extreme fight or flight responses, with the proposal here being that the response focuses perception in such a way that peripheral perception is reduced but focus increases tremendously, allowing you to process more input per unit time within the main field of your attention.

          I don’t know if this is related to the observation of a pretty striking linear relationship between metabolic rate and life expectancy in mammals.

        • AnObfuscator says:

          Mark, have you read much on Special Relativity?

          • Marc Whipple says:

            Yep. He’s rediscovered the reference frame.

          • Adam says:

            He actually seems to be saying something close to the observation of the strongly linear relationship between metabolic rate and life expectancy in mammals, or the not quite true but roughly reality-based statement that we get all get a different number of years but roughly the same number of heartbeats.

            Slightly tangential but related is the phenomenon of tachypsychia, whereby your perception of time is affected by the rate at which you can process units of sensory input within your primary frame of attention focus.

          • Mark says:

            Nope… do you think it is related to this?

          • AnObfuscator says:

            Yeah, see the comment by Marc Whipple.

            What you are describing is rather reminiscent of Special Relativity, except without the weird (no offense meant) philosophical parts you posted earlier.

            Also, SR describes this effect with the spatial dimensions as well, and other aspects of reality we normally take for granted (such as simultaneity).

            The time dilation and length contraction aspects are fairly straightforward to understand, IMHO. The simultaneity aspects can be a mind-bender, and I would recommend getting assistance in wrapping your head around that part.

          • Mark says:

            I don’t know the details of the science, but I do know that the principles of observed “external” reality must be determined by the capacities and nature of the observing equipment.
            One might expect to find explanations (predictions?) for the most general principles of science by considering the mind, if there was a sufficiently good understanding of what it meant to be conscious.
            Perhaps that is what is happening here… (though I seriously doubt it)… if so it’s quite a significant break-through.

        • Adam says:

          I don’t seem able to reply to this.

          Okay, now that that finally worked and I am able to reply, I tried to say an hour ago that what you’re saying sounds similar to the not-quite-true-but-close billion heartbeats idea, or the observation of a strong linear relationship between metabolic rate and life expectancy in mammals. In light of something like tachypsychia, you might speculate this means all mammals end up with roughly the same number of discretely identifiable memories even though they spend different numbers of sun cycles on the planet, but that’s kind of out-there speculative territory.

        • Adam says:

          Okay, fuck it, never mind. I guess wordpress doesn’t accept links to Stack Exchange? It could have just told me that instead of wasting an hour of my life trying to figure out why a reply wouldn’t post.

  33. Irenist says:

    I’ve been wanting to ask about nuclear war X-risk for a while in one of these threads, at least since having the issue brought to my renewed attention by this Max Fischer piece in Vox:

    I see that … someone (Topher Hallquist actually, but PLEASE let’s ignore that?) brought this up over on the EA Forum, with a link to the Vox piece, asking why GiveWell, and the rest of the EA community doesn’t highlight nuclear war prevention more:

    With his famous knack for concise incisiveness, Carl Shulman replied in part:

    I know GiveWell is aware of these articles, and has looked more into nukes. Probably more conversation notes will be coming out. There is broad agreement (and good object-level evidence) that NATO-Russia nuclear risk is the highest it’s been in the post Cold War period. One reason GiveWell has cited for not putting resources into nukes (although it was perhaps runner-up to the GCRs they have invested more in) is the existence of a large established community working on the problem that seemed fairly competent.

    Now, that might be all there is to say about it. But maybe not? So I’m raising the issue here b/c I want to see what the SSC community has to say about the relative prioritization of nuclear war vs. other X-risks (UFAI, asteroids, megavolcanoes, climate, etc.) in terms of both EA and activism. Thoughts, SSC folk?

    • walpolo says:

      The expansion of NATO to include Eastern Europe wasn’t a very good idea. I wonder if there’s any way out for the Western NATO countries to abandon the Eastern ones in case of a Russian attack on somewhere like Estonia? That seems like it would be the obvious right move.

      • Jiro says:

        By that reasoning, NATO shouldn’t include Western Europe if it happens to be a time when Western Europe is under threat.

        The whole point of NATO is to be a defense against this sort of threat. Having it include only countries that are not under threat defeats the purpose.

        • Saint_Fiasco says:

          I think the problem is that there is no buffer zone between NATO and Russia, which makes them nervous.

          • walpolo says:

            Exactly. Also, it’s strongly in the interests of the US to defend Western allies. What interest do we have in whether Estonia gets re-absorbed by Russia?

      • Irenist says:

        NATO’s posture in E. Europe seems to me to be a geostrategic issue that the various national security “deep state” institutions are going to run however they want. Since that sector is very, very resistant to public influence, I generally don’t think that there’s much of an EA angle: where we’re powerless to influence, it’s of limited use proposing ideas.

      • LHN says:

        Whether it was a good idea or not, I don’t see a way to abandon eastern European NATO countries while maintaining the alliance’s credibility at all.

        It might be replaced with a different defense arrangement among a smaller group of countries, or bilateral agreements between the US and key European allies. But the ex-NATO countries would have to expect those to at least be tested after a wholesale abandonment of Article 5, so the risk of war wouldn’t go away by doing so.

      • brad says:

        NATO should have been disbanded at the end of the Cold War. We won the war, but instead of disbanding the military alliance whose reason for existence was opposing the Soviet Empire we choose to expand the alliance to include its former client states. That doing so would cause resentment in Russia and an eventual rekindling of the conflict should have been obvious to anyone and everyone.

        Second best would have been to include Russia in NATO. NATO would then have been a silly waste of money instead of a dangerous provocation.

        • Earthly Knight says:

          It is also possible that failing to expand NATO would have resulted in Russia absorbing eastern Europe back into their sphere of influence the second they regained their footing, with conflicts breaking out in the early 2010s over Poland and Hungary instead of Armenia and the Ukraine. Counter-factuals are tricky.

    • Adam says:

      Are there nonprofits dedicated to reducing nuke risk? You can’t donate money to the State Department. Anything you give to the government goes straight to the Treasury and sits there until Congress appropriates it, with no input from whoever gives it. I guess you could give to the RAND Corporation or similar IR think tanks.

      • Irenist says:

        Are there nonprofits dedicated to reducing nuke risk?

        Great question. Here’s what GiveWell’s “cursory” research has turned up:

        Of the groups listed by GiveWell, the Nuclear Threat Initiative seems the most interesting. (Unsuprisingly, NTI was brought to GiveWell’s attention by Carl Shulman; I am all about reinventing his wheels today apparently). Very, VERY loosely, NTI sort of reminds me of a “nuke risk MIRI.” Here’s their site:

        • Adam says:

          Thanks. Looks pretty good and that’s a legit donor list. I’m personally way more concerned about nuke risk than AI risk.

          • walpolo says:

            Way, way more concerned. There are lots of actual nukes in the world right now.

          • Deiseach says:

            Yeah, to me worrying about AI risk as the big threat to human existence looks like worrying that “If we consume a rich diet, and keep consuming it, we run the risk of eventually developing gout” which is a perfectly reasonable worry except can we please wait to worry about it later AFTER I GET OVER THIS BOUT OF THE BLACK DEATH FIRST?

          • Irenist says:

            That’s pretty much where I am. I don’t think there’s anything at all silly about worrying about AI risk, and I disapprove of those who reduce such concerns to an occasion for snark. But in terms of urgency, I think nukes are a far bigger deal. Not just intentional deployment in war, but far more likely IMHO, a less well-handled version of the famous Petrov incident.

      • Troy says:

        Are there nonprofits dedicated to reducing nuke risk?

        The Catholic Church?

      • walpolo says:

        Union of Concerned Scientist is another such nonprofit.

    • John Schilling says:

      I don’t like the way the Vox piece hides, “and then NATO invades Russia”, down in the fine print. Because, WTF? That would certainly be a good way to start a nuclear war, but it’s a remarkably bad way to accomplish anything else, and far from the only way to e.g. defend Estonia. So, while I agree that the risk of nuclear war with Russia is rather higher than it was fifteen years ago, Vox has not given us a terribly impressive analysis.

      The major risk of nuclear war today is Pakistan v. India, with North Korea vs. the civilized world as a close second.

      More generally, nuclear war is not a plausible X-risk, not even with Russia. Not for the extinction of humanity, or even human civilization. It’s really, really, really damn hard to extinctify a species that has invented canned food, or totally destroy a civilization that has libraries. Harder still to do that with the civilization’s own weapons, because a necessary part of the process will be “kill everyone / break everything you can find, wait a decade, and see what has crawled out of the woodwork that needs re-killing”. Since the list of things you broke the first time around includes the civilization’s civilization-destroying superweapons…

      H.G. Wells forecast the end state of such conflicts, IMHO fairly accurately, all the way back in 1908’s “The War in the Air”. OK, wrong about aircraft with conventional bombs being the civilization-destroying superweapons, but the result is a world whose reduced nation-states can at best concentrate the resources to build a bare handful of superweapons with which to knock each other back to the not-quite-able-to-build-superweapons stage. W/re nukes, that means a balkanized world of roughly 1950s level scientific and industrial development, and occasional wars in which a few nukes get tossed about. Also libraries full of advanced science that you mostly don’t have the infrastructure to use. Mostly.

      And now I want to see works of fiction in that setting, because it sounds even cooler than Steampunk.

      • walpolo says:

        How would you get back to 1950s-level infrastructure with all the easily-extracted fossil fuels used up, along with a large portion of the easily-extracted nitrogen and phosphorus for fertilizer?

        You’re right that an all-out US-Russia war wouldn’t be extinction, but it would very likely be the permanent end of industrial civilization.

        • LtWigglesworth says:

          Meh, its fairly likely that in a nuclear exchange between the US and Russia that the majority of warheads would be targeted at counter-force targets, with a minority aimed at economic targets like oil refineries.

          Both sides would be attempting to prevent the launch of as many of the opponents missiles as possibles.

          I think that any major counter-value strikes (against places like LA, or NY) would most likely occur in the second wave launched from SSBNs.

          You will have a bad day if you live near missile fields in Montana, North Dakota and Wyoming, near major airbases, or nuclear submarine bases.

          Living near the major oil refineries in the south will also be unpleasant.
          If you live a large city like New York, there is a significant chance you will get nuked.

          However large swathes of the rest of the US would be untouched, as would many other countries around the world. Why hit Rio when that warhead (plus the others on the same bus) could be aimed at North Dakota to hopefully prevent some ICBMs from coming back at you?

          • walpolo says:

            Counterforce attacks would also be ground bursts, causing way more fallout and contributing much more to the likelihood of a long nuclear winter.

            >I think that any major counter-value strikes (against places like LA, or NY) would most likely occur in the second wave launched from SSBNs.

            That wouldn’t make them any less damaging.

            Also, keep in mind that any predictions about what would be targeted are contingent on the command chains staying in place during a war.

          • John Schilling says:

            Fallout is not an extinction-level threat outside of bad fiction; it isn’t even the dominant cause of death in nuclear war. Just the spookiest.

            And nuclear winter is driven by the soot from burning cities and oil depots. Dust from groundbursts is too heavy and, well, falls out.

            We put our missiles in underground fortresses in North Dakota for good reason, and part of the process was having smart people verify that this really does reduce the collateral damage to civilian population and infrastructure by a significant degree.

          • walpolo says:

            Huh, you are right about the soot vs dust issue. Don’t know how I ended up with the wrong impression about that one.

            Still, I’d really like a good citation for the claim that a US-Russia war would mostly consist of counterforce attacks.

          • Nornagest says:

            Still, I’d really like a good citation for the claim that a US-Russia war would mostly consist of counterforce attacks.

            I don’t have a cite on hand, but counterforce/countervalue strategy during the Cold War was dictated largely by the accuracy of the weapons being used. Silos and command bunkers are very tough, and essentially immune to the thermal effects that dominate the damage done by megaton-scale weapons; you need a close hit to neutralize them. That necessitates an accurate weapon: you don’t necessarily need a direct hit, but even a CEP in the hundreds of meters is difficult to achieve when you’re tossing nukes from the other side of the world.

            Naval weapons, which have to be relatively small and also have to be tough enough to spend months in a moving submarine with all the attendant problems, had a hard time being that accurate until relatively recently. It’s much easier for silo-based weapons. Since you are trying to win a war and not just to kill a bunch of people, it follows that it makes sense to dedicate your silo-based arsenal to a counterforce role, aimed at destroying or degrading your opponents’ ability to make war. The less accurate naval weapons then are kept as a deterrent.

            (I could be wrong about this, but I seem to remember that air-launched options such as cruise missiles are intermediate between the two. This is post-ICBM, I mean; before ICBMs were perfected, Cold War strategy was all about bombers.)

        • John Schilling says:

          You can run a 1950s-style industrial civilization on hydroelectric power, biofuels, and other renewables. Wood-burning automobiles are a thing, and have been so since before the 1930s. Diesel engines run just fine on straight vegetable oil, if they haven’t been too narrowly tuned for kerosene.

          It wouldn’t be as materially wealthy an industrial civilization as we had in the 1950s; V-8 “muscle cars” will not be toys for the middle class. Look at, e.g., the domestic economies of Germany and Japan during WWII for example, where scarce petroleum was largely reserved for combat vehicles.

          And this is probably a good thing after all; the line is being drawn such that a large nation-state can barely afford a small enrichment plant or breeder reactor; relative poverty in the material sense would allow for greater scientific and technical ability at the margins while still staying at that limit. And I want at least 1970s level dentistry in my post-quasi-apocalyptic future; maybe even a few Altair 8800s for the hardcore geeks.

          I don’t think this is actually our future, but I increasingly want to read its stories.

          • Hedonic Treader says:

            I think astronomical waste is a crucial consideration for many x-risk worriers. If nuclear war means a permanent 1950s-level cap on total wealth and technology, it would mean locking humanity out of space.

            What happens on earth is less important than what happens to billions of other star systems and galaxies.

            Of course, such a cap would also prevent astronomical amounts of additional violence and suffering. Either way, this seems important to model correctly.

          • houseboatonstyx says:

            @Hedonic Treader
            > 1950s-level cap on total wealth and technology, it would mean locking humanity out of space.
            > What happens on earth is less important than what happens to billions of other star systems and galaxies.

            Focusing on particular and perhaps incompatible risks (nuclear war, biological or political disasters of one kind or another, paperclip AI, asteriod strike, etc) is inefficient. You’ve added another very important risk: letting space continue lifeless rather than spreading our own life (human, Gaian, whatever), ie ‘colonizing’.

            But one of these things is not like the others. Even colonies extending about as far as our current probes, would prevent extinction by disasters on Earth. Terraforming Mars is nowhere near even pie in the sky — but orbital stations are in the sky already, and can be expanded by current or reachable technology, with spinoff tech helpful for dealing with climate change or asteroid threats to Earth.

          • Hedonic Treader says:

            I think the “eggs in more baskets” approach of off-world colonies or stations only works if they are self-sufficient to bootstrap an interplanetary civilization on their own. But that would probably mean we can do it on earth too, despite volcanos or nuclear winters.

            You are right that leaving space lifeless is a risk for those who want to maximize a positive quality that relies on life.

            But I want to reiterate that this implies maximizing all the negatives as well, unless you can identify good filters against them. David Pearce suggests phasing out suffering from the living world through biotechnology and hedonic enhancement and then spreading to the stars. Others just assume that the future will contain more good than bad aspects of life, or they just declare life a net-good by fiat.

            I myself am not convinced more life is better, as I see it as a torture trap without any optimization process – including human benevolence and intelligence – that would sufficiently filter against the negatives. Unfortunately, this is low status while “yay life” is high status and conforms to social desirability bias. So even if pessimism is objectively right humanity will act as though it’s obviously wrong.

          • houseboatonstyx says:

            @ Hedonic Treader

            > I think the “eggs in more baskets” approach of off-world colonies or stations only works if they are self-sufficient to bootstrap an interplanetary civilization on their own. But that would probably mean we can do it on earth too, despite volcanos or nuclear winters.

            I disagree with several practical points there, but on to more interesting things.

            >You are right that leaving space lifeless is a risk for those who want to maximize a positive quality that relies on life.

            > But I want to reiterate that this implies maximizing all the negatives as well, unless you can identify good filters against them. David Pearce suggests phasing out suffering from the living world through biotechnology and hedonic enhancement and then spreading to the stars.

            Artificial habitats can be safe baskets long before the stars would be possible, and would give Pearce a laboratory for Edens.

            > Others just assume that the future will contain more good than bad aspects of life, or they just declare life a net-good by fiat.

            > I myself am not convinced more life is better, as I see it as a torture trap without any optimization process – including human benevolence and intelligence – that would sufficiently filter against the negatives. Unfortunately, this is low status while “yay life” is high status and conforms to social desirability bias.

            I very much enjoy arguing, er, steelmanning from both sides at once. The immediate “yay life” reply would be, “Spread within Sol’s system, and let future generations think about that tomorrow.” (With my System 1 telling me to go ahead and capitalize Life or Evolution or something like that, which is probably in some Jaina dictionary).

          • Hedonic Treader says:

            The risk is that future generations might not worry enough about, say, large-scale involuntary suffering and that the Eden experiments don’t work.

            So altruists who use their own money and energy today to enable this future may be pushing in the wrong direction now, if pessimism turns out to be correct.

          • houseboatonstyx says:

            I’m kind of boggled here. Getting back to the current x-risks … assuming an asteroid strike that wipes out all life on earth before Pearce’s ideal could be realized, would you prefer there be no starter baskets of life anywhere else? The whole universe lifeless, not just extra-terrestial space ?

            if pessimism turns out to be correct

            Do you mean pessimism about some particular outcome (eg Pearce’s idea proving impossible)?

          • Hedonic Treader says:

            Pessimism about the net utility of space colonization, e.g. slightly more negative than positive experiential value in aggregate would suffice for me, but you could also enter other more complex values like injustice, victimization, violence which might constitute a negativity surplus.

            This could easily be true if, for instance, it turns out that certain efficient learning algorithms map onto our concept of suffering, or that r-selected species dominate most of sentient life forever and have a suffering surplus (hard to empathize for humans), or if the game theory of group hatred or blackmail winds up incentivizing large-scale violence.

            >assuming an asteroid strike that wipes out all life on earth before Pearce’s ideal could be realized, would you prefer there be no starter baskets of life surviving anywhere? The whole universe lifeless?

            Yes, if pessimism is correct and there is an ethical negativity surplus in an expanding sphere of life, not having a backup would obviously be better.

            People tend to think, but life has positive option value! The problem lies in who gets the option. If unFriendly AI (or immorally acting humans) gets the option value, it can well be negative from the perspective of those who are now considering x-risk donations for altruistic reasons.

      • Irenist says:

        @John Schilling:

        I think of nuclear war as an X-risk mostly because I assume there’s a non-zero possibility that nuclear winter could actually be a thing. I mean, I’m aware that some have claimed to have “debunked” it, but I’m assuming it’s not a definitive disproof of the worry. So sort of like AI, I figure it’s a matter of non-zero probability multiplied by infinitely bad outcome (from a human perspective) = X-risk. I’d be relieved to have those assumptions corrected, if you’d care to tell me that nuclear winter is just a total theoretical non-starter.

        • John Schilling says:

          The human race survived an actual Ice Age, back in the day. The bulk of our evolutionary history has been about how to survive ice ages. And then we went and invented canned food and mine shafts.

          A worst-case nuclear winter would be massively disruptive and perhaps massively lethal. It might kill most human beings living at the time, though I rather doubt it. It is not a plausible extinction risk. To our future civilized descendants of 5000 AD, it would be a historical footnote on the order of the Bronze Age Collapse.

          • James Picone says:

            That was prior to agriculture, wasn’t it?

            I’m not sure that’s convincing evidence that advanced civilisation would survive a nuclear war + nuclear winter.

            I agree that it’s unlikely to lead to a literal extinction of human beings. Is there room in the term ‘x-risk’ for something that’s merely extremely awful? It’s in the same broad category, even if there isn’t an actual risk-of-x.

          • John Schilling says:

            Agriculture – that’s that thing that lets the Mormons each stockpile a year’s worth of food against possible calamities, isn’t it? 🙂

            Even the old worst-case models of nuclear winter didn’t postulate a global cessation of agriculture, particularly at lower latitudes and in the southern hemisphere. Nor did most of the effects persist beyond a single growing season, IIRC. And the worst-case models aren’t
            particularly credible any more, for several reasons.

            In terms of the acute catastrophe, using agriculture to fill the planet to close to its carrying capacity, means creating on the order of five billion more casualties for the first substantially agriculture-disrupting calamity to come about. But if the question isn’t what might be lost in this generation but what will be saved in future generations, agriculture is in the “can’t hurt, will probably help” category.

            Between the accumulated food reserves of agricultural civilization and the continuation of agriculture as possible in less affected regions, the human race is better positioned to survive a hard winter than it was ten thousand years ago. Billions of people, and technological civilization, would come through a nuclear winter intact. Worst case is, the number of billions might not be seven.

          • Irenist says:

            These are helpful responses. Thanks!

  34. Nuño says:

    I´ve recently stumbled upon the concept of “Rational ignorance”. I understand the concept, and for some reason (see, my extensive collection of trivia), I really don´t like it. Are there any incorrect assumptions or inferences, or anything in, the wikipedia page explaining it (

    • Urstoff says:

      Seems pretty straightforward, and a love of trivia isn’t a counterexample (as the benefit is the pleasure you get from knowing the trivia).

      I think the real question is whether voters are ignorant because of rational ignorance or are ignorant because of unjustified confidence in the beliefs that they already posses (will vary per individual voter, of course).

    • roystgnr says:

      If you go out of your way to find and remember an extensive collection of trivia, then presumably knowing trivial things has a high terminal value for you, and so being ignorant of those things instead would not be rational.

      If I’m wrong and you actually expected your trivia collection to have significant instrumental value, then you may be disappointed. 😉

      • Deiseach says:

        If you’re setting out to learn trivia with the expectation “Some day this will come in useful!” you may indeed be disappointed.

        On the other hand, it’s constantly astounding to me how often I know the answer to some question, not because I’ve deliberately studied up on it, but out of a headful of magpie-acquired facts. Often I go “How did I know that?” (particularly when it’s something unsavoury) 🙂

        Odd nuggets of facts can come in surprisingly useful when making connections and correlations between disparate fields.

    • Richard says:

      I’ve always thought the concept applied to depth of knowledge. At some point it stops being beneficial to know more and more about a specific topic. Most trivia is rather shallow knowledge about a ton of different topics which I quite often find useful in my day to day life.

      So unless you collect an extensive collection of trivia about, say, the 1912 olympics, you should be OK 🙂

  35. Guy says:

    This was originally going to be in reply to desiach, but then I decided it wasn’t really related to that.

    Anyhoo, when I saw the pictures in this article about the #ilooklikeanengineer thing, I didn’t think the actual advertising image was the same woman as the one in the first shot (or even a real engineer). Somehow the glasses, tech-related-t-shirt, and advertising-speak combined to punt me straight to “person paid to model geek girl” rather than “woman discussing her job”. It was also eerily reminiscent of those ads for “funny” tshirts that inexplicably feature half-naked women, which seem to crop up on sites like TvTropes. Though obviously Ms Wegner is fully clothed. That I’m putting down to “attractive woman on a white background”.

    Anyone else have a similar reaction (now that I’ve primed you) or am I The Reason There Are Too Few Women In Tech / in possession of some subconscious biases that need fighting?

    • Adam says:

      Well, I think at bare minimum they didn’t select an ugly person wearing ill-fitting clothing who didn’t comb her hair that day for a reason. This brings me back to being an undergrad when I met a girl in a writing class who was speaking that night about being comfortable with your looks and I outright asked her how she was supposed to be an effective spokesperson for that kind of thing when she was maybe the third or fourth most attractive person on campus. She didn’t have an answer.

      • Guy says:

        Honestly, in this case, I think they could have saved the ad (and made it vastly more effective) by placing her anywhere other than somewhere-in-adspace. That’s the difference between the two photos – in the one for twitter, she’s in a tech environment. I have no reason to believe someone just took a picture of a model and then stuck the words of the real Isis Wagner next to her because I can see the environment she is part of.

        If you can find some other images from the original campaign, many of them suffer from this. The only one I saw that didn’t managed to succeed because the guy wore a crazy looking hat and some steampunk goggles, probably at his own insistence.

        I don’t think her looks eliminate the possibility that she’s an engineer in the ad, just that the design (or lack thereof) in the ad points me to “model” way faster than “actual engineer with this company”.

        • HeelBearCub says:

          Mmmm, I don’t think you quite hit the mark, but I think you are close.

          My guess is, she looks like a model in the ad because she is being treated as a model. My guess is they had an ad-creation team that probably did things like pick the clothing, tailor them, professional makeup, chose a set of “clears” for her, professionally lit the shot, etc.

          Broadly, in-tech, I think we have a bias against people who seem like they put effort into looking good, especially if they are competent at it, and that probably cuts across genders. I also think there is some sort of bias in tech that sorts out females, but I can’t quite say what it is.

          I know some very good female developers, just as good as any male developer I know. They each have their own strengths and weaknesses and programming traits, but none of those traits are attributable to their gender AFAIK. The all happen to be foreign nationals, which may not say much at all, but is a piece of information.

          • Guy says:

            I think even if they had professionally lit the shot and such, putting in a background of her workplace would have helped. Look at the MIT engineering professors in the NPR article: they may look like fairly generic “women in science” photos, but they’re pretty clearly STEM women who do STEM things, even if you don’t necessarily recognize them. If you added a quote and attributed it to a likely name, It will look like the statement is related to the image provided, rather than just “here’s a picture of a woman, here’s a statement by an employee”.

            I think in STEM generally we have a bit of a bias against those who appear to try (ie, exert effort rather than simply excelling naturally or failing to), especially if they appear to try at non-STEM things. This is not a good thing.

          • HeelBearCub says:


            Take this picture from MIT as an example.

            I feel fairly confident that if an Ad firm had taken those shots, they would look very different, even with the same people in them. That first shot (top left), could easily be made to look like this.

            I’m not sure how much that would really have helped.

        • Adam says:

          Her workplace looks pretty friendly beside, and I don’t get the focus on web developers. That’s already the highest female ratio of all computing occupations, third-most behind statistician and operations research analyst in ‘computer and mathematical occupations’ as classified by BLS, and why the focus on computing in general when actual engineering is way lower?

    • Deiseach says:

      I think the problem is that we expect “real people” in advertisements to actually be portrayed by actors and/or models, because any time real people are used in ads, they are hilariously stilted (you should see some of the Irish cattle drench ads using GAA players).

      So naturally actors/models are used to play ‘real people’ so as to deliver lines in a smooth and convincing fashion, but the trouble is that when you’ve heard the same handful of voice actors over a period of time, you recognise them the next time ‘real people’ Joe or Maisie pop up in an ad to talk about product X or service Y or working for company Z.

      So, as I said, you automatically expect people in ads to be fakes. We’ve got female engineers and technical staff in my workplace and they don’t look like the woman in the ad, either (even though they’re all fine-looking women, let me defend my colleagues!)

      Obviously for the purposes of the ad, the woman was made up and styled professionally, and that didn’t help make her look ‘real’ like a ‘real person’ who wouldn’t be professionally styled and photographed.

    • sam says:

      I wouldn’t worry about it. OKCupid has done research on their huge dataset of user photos that suggest a host of factors about the image affects people’s judgement of the attractiveness of the person in the photo. Camera quality has a pretty big impact on attractiveness ratings. Even whether or not the person is smiling and which direction they’re looking has an effect on ratings. So it’s not really surprising that a twitter selfie makes you think “average person” and a polished, professional ad makes you think “model”.

  36. FedeV says:

    I saw this article getting linked around a bunch on twitter:

    Which cites this interview, from 2013, ( where Gena Davis mentions this:

    DAVIS: We just heard a fascinating and disturbing study, where they looked at the ratio of men and women in groups. And they found that if there’s 17 percent women, the men in the group think it’s 50-50. And if there’s 33 percent women, the men perceive that as there being more women in the room than men.

    This is basically all over twitter, yet I’ve been completely unable to find the original source. Has anyone heard of it? Jezebel, Geek Feminism, etc all went gaga over it but nobody seems to find the primary source.

    • Emily says:

      My suggestion would be to e-mail the researchers who work for her organization. (Or their research assistants.) I think this is a misreading of the 17% statistic from this , which seems to have morphed into some more-ambitious claims over the years.

      • FedeV says:

        I thought about it, then backed down.

        I have no issue e-mailing from my academic account to request clarification when a paper is in my area of expertise, but I was worried I’d come across as a random gamergater with a chip on their shoulder if I wrote to an organization that’s dedicated to gender research. I’m probably being unreasonably paranoid, I realize.

        Politicization ruins everything :/

      • Who wouldn't want to be anonymous says:

        Hypersexuality refers to an overemphasis on attractiveness and sexuality by way of clothing (i.e., alluring attire) and body proportions (i.e. uncharacteristically small waist, hourglass figure, thinness).

        Females were over five times as likely as males to be shown in sexually revealing clothing, which was defined as
        attire that enhances, exaggerates, or calls attention to any part of the body from neck to knees.

        This is almost completely unrelated, but…

        Unless they would count Alfred as hypersexualized by that definition, I am pretty sure they are wallowing in confirmation bias. Or they just don’t understand how cloths work.

        Edit: Added the second half of the quote.

        • HeelBearCub says:

          I think this is a much better example of something that actually draws attention to the persons body (as opposed to drawing attention away from the actual body and into the accessory, as the bowtie does).

          • Who wouldn't want to be anonymous says:

            Haha, awesome.

            I was trying for an example that isn’t an obvious sex symbol that is nonetheless wearing clothing that is specifically designed to accentuate the body. That was the example that popped into my mind (I watched Batman recently), but I couldn’t find a good full body picture. And Micheal Gough is actually a lot better looking than I remembered, which doesn’t help the not-a-sex-symbol thing.

            Nevertheless, I was actually thinking about everything *but* the bow tie.

            Standing collars are torture devices designed to lengthen the neck and adjust posture (kind of like how high heels are torture devices designed to lengthen the legs/height and modify posture).

            Tail coats (which Alfred wears most of the time), along with frock coats and morning coats, are designed to create an unrealistic hourglass figure. They are actually constructed completely different than a suit coat so that the waist can be suppressed to an unnatural degree, sometime coupled with large amounts of padding around the chest and shoulders. Suit coats don’t have as much waist suppression but they generally have the shoulder and chest padding. In both cases the size and shape of the lapels (and the pocket and hanky), are designed to accentuate the chest.

            High waisted and high rise trousers are, of course, cut to accentuate the (long) legs, (narrow) waist, hips, ass and… package. (Or, eg, in Frank Sinatra’s case, to hide how freakishly large the package is.)

            Dress shoes are invariably heeled which, while not as torturous as high heels, still aim for the same ends; high heels having fallen out of favor with men as soon as women adopted them.

            Ooh yeah, and hats! Especially those ridiculously top hats. Because height is a really big deal. You know, like that wage discrimination per inch and stuff.

            But my real point is that all clothing is intended to modify how the body looks; there are very, very, very few people that literally only wear clothing because they would get arrested if they didn’t. Anyone who seriously thinks virtually all men’s clothing isn’t sexualized to a very high degree has never talked to a tailor or read a men’s magazine.

            Unless someone is willing to grant that Alfred is hypersexualized, any claims about differences in degree are automatically suspect.

          • HeelBearCub says:

            @Who wouldn’t want to be anonymous:

            I don’t know about all clothing (right now I am wearing cargo shorts and a T-Shirt and I recently read an article that this was “unacceptable” for men to wear) but perhaps all “style” though.

            Even then, it seems that it might be useful to consider a difference between hyper-sexualization, hyper- masculinization and hyper-feminization?

            There are many modes of dress that amplify (a specific culture’s) gender difference markers. Some of them accentuate the body, some distract from it, some substitute for it.

            Are Victorian fashions in general considered hyper-sexualized? I don’t think so. The women’s fashion seemed to both cover everything and emphasize the existence of that which was covered. But I’m certainly no expert in how academic feminism views Victorian dress.

          • Nornagest says:

            Are Victorian fashions in general considered hyper-sexualized? I don’t think so.

            “Hyper-sexualized” today tends to imply “showing a lot of [usually female] skin”, and Victorian fashion didn’t, but it was very much concerned with projecting an idealized silhouette. Anonymous has covered menswear; women’s wear involved crinolines (semi-rigid frames worn under a skirt to produce a bell shape), bustles (frames or thick padding worn on the upper buttocks to help produce an hourglass figure), and a depth and variety of corsetry that’s almost indescribable. I only know about it because I spent a year dating a girl who was deeply into the steampunk/Dickens Fair scene.

            Hyper-aestheticized might be the best way to put it now, but it would definitely have had sexual overtones in its era. The late Victorian period saw a number of dress reform movements, often linked with the feminism of the day, which after a couple decades of recombination helped produce the much lighter styles of the 1920s.

          • HeelBearCub says:


            I think we are in broad agreement on Victorian women’s dress. That’s what I was trying to get at with my statement that it seemed to “cover everything and emphasize the existence of that which was covered”.

            I’m sure these aren’t period accurate, but compare:
            this picture of Audrey Hepburn in “My Fair Lady” with this one.

            To the extent that either one is “hyper-sexualized” it is the first one, with it’s skin-tight fit. Compare that to the gentlemen in the picture, who’s bodies can scarcely be found.

            I assume that the argument could be made that a bustle is hyper-sexualized. It does literally increase the apparent size of a part of the anatomy that functions in sexual signalling. But if video games, comic books, TV shows and movies were currently slapping bustles on all their female characters, I don’t think the accusation would be of hyper-sexualization. (Other things, sure, but not that).

          • Jaskologist says:

            The only thing more alluring than a nipple is the promise of a nipple.

          • Nornagest says:

            @HBC — Certainly that’s all true by modern standards, and the norms concerning public sexuality (private’s another matter) in the Victorian era were such that it might even be true by Victorian standards. But I think that’s a distinction without a difference in every way that matters.

            Victorian people (of both sexes) spent far more on average than we do, in terms of money and time and discomfort, in pursuit of their society’s ideal of public beauty. Hands down. Now, their version of that ideal is very different than ours; theirs spent a lot of effort shaping the body and adorning it but did not directly display it, while we do the opposite. But both approaches are aimed at public display, theirs is probably more gendered than ours, and I don’t see a good argument that there’s something uniquely evil about ours that spans the gap in effort.

            Yeah, you can construct a reading of “hyper-sexualization” that includes 2015 fashion but not 1890. What does that buy us, though?

  37. Nate Gabriel says:

    The India-Bangladesh border is being cleaned up. This is apparently a good thing, at the cost of making maps of Earth that much less interesting.

    • Anthony says:

      Well, maps of Earth don’t generally have the detail that would show the craziness of Cooch Behar, but it does make cartography less interesting. For an example of craxy borders that probably won’t get rectified, look up Baarle-Hertog/Baarle-Nassau.

  38. Kavec says:

    aw crap

  39. Hoyticus says:

    As a long time reader I’ve noticed SSC has an excellent and eclectic base of posters and readers. My question for this quirky group is this, how to defeat insurgencies post-1945? Especially foreign armies trying quell insurgencies beyond their own borders. Also, both examples are from Martin van Creveld’s excellent book “The Changing Face of War: Combat From the Marne to Iraq”. For the uninitiated Martin van Creveld is likely the most famous living military historian.

    So far the only solutions that I’ve encountered if one wants to call them that are either the Hama Method or Northern Ireland Method. Early 1982 in Hama, Syria Hafez al-Assad ordered his brother to gather a few divisions and more or less shell the city into the ground in under a month killing anywhere from 10,000 to 40,000 as Syrian intelligence gathered the city was the home base for the Muslim Brotherhood (Assad’s only likely threat to usurp power back then, which had also been clashing with security forces for a couple years). Syria had a depressing peace for roughly 30 years.

    In Northern Ireland The British struggled for decades to quell The Troubles and started having their greatest successes after the Bloody Sunday Massacre once they quit using tanks and other heavy equipment to maintain peace. Overall. the British military suffered about as many casualties as the entire civilian population, they more or less won by “killing them with kindness” or more accurately breaking their will to fight by appearing stolid, stoic, and unyielding.

    So the only successful methods I’ve seen are either using the swift terrible sword to quickly smash the head of the insurgency and mercilessly crush the opposition civilians and all, or to use what amounts to morale attrition and breaking the insurgents by making them think you’ll last longer than they will. The issue is both of these examples are governments defeating local insurgecies (unless you disagree with me and believe the Baa’thist regime in Syria run by native Alawites and the British are sufficiently distinct from the people of Northern Ireland). So my problem of how can foreigners quell an insurgency in another land still stands.

    I’m eager to hear the thoughts, musings, and opinions of the SSC crew.

    • Adam says:

      The RAND Corporation produced a pretty comprehensive study of How Terrorist Groups End, focused on the period 1968 – 2006.

    • Troy says:

      Not a direct answer to your question, but Erica Chenoweth has done interesting research on the effectiveness of nonviolent vs. violent insurgencies, finding that the former tend to be more successful. If the former tend to evoke a different response than the latter, then that could be one causal mechanism explaining their greater success. (On the other hand, if nonviolent tactics tend to more effective than violent tactics for more intrinsic reasons, then this might tell in favor of the second approach you mention by governments.)

    • brad says:

      So my problem of how can foreigners quell an insurgency in another land still stands.

      We are positing an insurgency which by definition means that there’s a government being insurged against. So where is that government? Why isn’t it taking the lead in quelling this insurgency, perhaps with the assistance of outsiders? Is it because the government in question is a government in name only? Is it because the government in question doesn’t really want to quell the insurgency? Something else?

      I think if you answer that question you’ll figure out why these attempts are failing spectacularly.

      • Hoyticus says:

        @ brad

        I believe we’re in agreement that most of the governments that have battled local insurgencies since 1945 are generally awful and ineffective. I should have been more explicit in wording my real question. Why do great powers like the U.S. fail in Vietnam, and Iraq and seemingly so again in Afghanistan, why did France lose in Vietnam and again in Algeria, Why did the USSR fail in Afghanistan. The post-1945 trend of insurgents besting some of the best trained and equipped militaries humanity has ever marshaled is a red flag that something has changed about war. Martin van Creveld has hypothesized what he calls the Power of Weakness.

        My unoriginal idea is simply that local insurgents care more about where they live than we do. The FLN was willing to fight to the death over Algeria, the French were not. The NVA and Viet Cong had the same love of country that couldn’t be broken.

        • Adam says:

          The new Army/Marine Corps Field Manual on Counterinsurgency came out in 2006. The official doctrine at this point is that successful insurgencies work by undermining the ability of the government to provide basic security and infrastructure for the people, and successful counterinsurgency has to work by focusing on the security of the local population. This is notoriously hard to do when the locals you’re protecting and the insurgents look exactly alike and your primary concern as a soldier is the safety and survival or yourself and your unit.

          It’s been difficult to get operational units to actually use the new doctrine, as opposed to the counter-guerilla doctrine of focusing on killing the enemy. There was a very high-profile case a few years back of a brigade commander openly ignoring the orders to use counter-insurgency doctrine rather than counter-guerilla. He was never relieved, but I actually saw him at Fort Knox a few years later running an Internet survey program, so I’m guessing it derailed his career.

          Actually, it would appear that he quit and has decided to pursue IT consulting and project management instead.

        • brad says:

          It sounds old fashioned given that after 2001 we forgot everything we thought we knew, but I’d say it comes down to “nation building”. Pre-2001 the consensus (though not universal) was that militaries aren’t good at it and probably shouldn’t try.

          But what exactly is the mission in these counter-insurgency-in-another-country wars? I’d argue it is to build up our ally/puppet government to the point where they can either drain the swamp which creates the insurgency — using either the carrot or the stick. That in turn is required because a credible government is a necessary end state. Without that your foreign army can never leave, which means it can’t win.

          I don’t think we are necessarily disagreeing. If the foreign government cared enough that it didn’t want to ever leave than it wouldn’t be necessary to build up a local government, it would just be the local government forever. Think China in Tibet.

          Of the examples you gave, Algeria is in some ways the most interesting, because it plausible to believe that at the beginning many or most Frenchmen, really did want to stay forever. But the insurgents sapped the will of enough of them that they changed their minds. It may make sense to think of that as more losing to an internal insurgency, than it does to lump it in with Vietnam/Afghanistan/Iraq (or not, I haven’t’ fully thought it through).

    • ddreytes says:

      Don’t want to speak too assertively here because (a) there’s obviously a lot of really contentious political issues tied up with this stuff (b) it’s been a long time since I read anything extensively about these issues.

      But basically, in my mind, there are a few basic methods here.

      First, you can make it clear that you will outlast them one way or the other – I think both of the strategies you outline fall into this category: the swift terrible sword method seeks to increase the pain to the insurgent group, while the solid stoic unyielding approach attempts to demonstrate the superior willpower and commitment of the government group.

      Second, you can try to disrupt the functioning of the insurgency group, eliminate as much of the reasons for its support as possible, and in general take the ground out from under it – this is the basis of classical counterinsurgency warfare, I think. You try to prevent the insurgency from functioning.

      Third, you can give in and negotiate a political compromise with the insurgent group (and, in passing and without wishing to involve myself in an obvious grounds for political controversy, I think there’s an argument to be made that you should classify the end of the Troubles in this category).

      The question of whether the insurgency is ‘native’ or ‘foreign’ doesn’t really affect any of those. A ‘foreign’ insurgency will be harder to deal with, of course.

    • Tracy W says:

      The British defeated a Communist insurgency in Malaysia in the 1960s, from memory. There was a hearts-and-minds thing i think, and also some British-caused atrocities. But they did it. And that’s the sum of my knowledge on it.

      • AlphaGamma says:

        I’m not entirely sure how much of the leadership of the campaign was British and how much was Malaysian- Malaya* gained independence partway through, but I think Malaysians were involved in decision making well before that.

        Of course, independence meant that the insurgents could no longer cast themselves as fighting for independence!

      • John Schilling says:

        The Malaysian insurgency was also a notably easy case in that the insurgents were part of an easily-recognizable minority ethnic group. Helps if you know at a glance whose hearts and minds you want to win and who just needs to be behind barbed wire for the duration, and if the numbers in the latter class are not intractably large.

        Also helps if you’re willing to throw someone behind barbed wire for the duration because the look Chinese. Than can be a problem for some people.

    • Deiseach says:

      In Northern Ireland The British struggled for decades to quell The Troubles and started having their greatest successes after the Bloody Sunday Massacre once they quit using tanks and other heavy equipment to maintain peace. Overall. the British military suffered about as many casualties as the entire civilian population, they more or less won by “killing them with kindness” or more accurately breaking their will to fight by appearing stolid, stoic, and unyielding.

      *deep breath*

      This is a sore topic and this is my completely uninformed gut-level reaction. (There’s a reason I think Iraq is not “Vietnam Part II” but “America’s Ulster”).

      Who ‘won’ in the North? Well, that depends on several elements. Margaret Thatcher was adamant that there was no war going on, that terrorism was merely criminality and should be treated as such. Now, this was part propaganda and part political in order to refuse political prisoner/prisoner of war status to IRA members (and the other paramilitaries). Secondly, there wasn’t one insurgency or one umbrella insurgent grouping; the IRA and its offshoots and associates (do we consider Sinn Féin as a political party? as the political wing of the IRA? as something inbetween?) wasn’t the only game in town, there were also the Loyalist paramilitaries and political parties.

      The problem there was that the Loyalists were ‘on their side’ for the British, or needed to be kept sweet by being treated as such, and so the will to crack down on Protestant murder-gangs was not perceived to be there (whether it was or was not in actuality is another matter for dispute).

      Thirdly, the police force was militarised (even more than it already had been, you could argue: see the B-Specials). Bloody Sunday was a PR disaster for the British, whatever else; it may or may not have been militarily effective, but when the whole sales pitch of the British government was that they were going in as neutral peacekeepers, having your soldiers enforce martial law on a civilian population by going in and shooting everyone rather undercut that in the global perspective. It also massively revived the IRA which, ironically, had been losing ground during the period of the 30s-50s (the 50s bombing campaign was an effort to revitalise and make themselves relevant again) particularly to the Civil Rights movement which was inspired by the American Civil Rights movement and the French student protests. After Bloody Sunday, the physical force strain in Irish politics predominated once again. The IRA was seen as the only element willing or able to protect the Catholics (the police were not so viewed) and even down South, where for certain elements in the political establishment our revolutionary past was an embarrassment, it galvanised public opinion very strongly pro-IRA (see the burning of the British Embassy).

      The slack was then taken up by the RUC and UDR. Things like Diplock courts as well as internment without trial (which backfired spectacularly by inducing such resentment in the civilian Catholic population that it pretty much acted as recruitment drive for the IRA and other Republican paramilitaries).

      British intelligence ‘black-ops’ and the whole morass of secret service meddling, along with the duelling going on between the British Army and the IRA in the Bandit Country were more focal points of the ‘war’, post Bloody Sunday, than out-and-out direct conflict between the British Army and paramilitaries.

      Well, let’s fast-forward over forty years of The Troubles and get to the Good Friday peace agreement, which so far seems to be holding. Talking about Who Won The War is not really helpful; casting it in terms of winners and losers perpetuates divisions which, when everyone has to live in the same place, really don’t help create a unified community.

      Can you say the British government “won”? They still retain dominion over the North, yes, and that appears to be psychologically important to the English (as witness the campaign on Scottish independence where many who held disdainful views on Scotland still urged the Scots to remain part of the United Kingdom). But they still have to pay billions per year out of their national budget to prop up the statelet and they are increasingly reluctant to do so. What will happen further down the road? Who knows?

      Sinn Féin can claim it “won” since it has achieved recognition as a legitimate political party and power-sharer from a status of being censored (under Section 31 of the Broadcasting Act, our national radio and television station could not broadcast any interviews with Sinn Féin representatives on any matter at all; in Britain, the national public and commercial stations were permitted to dub an actor’s voice over footage of, for example, Gerry Adams in order to get around the embargo on reporting).

      Moderates can say they “won” because people like John Hume persisted in talking with terrorists and extremists and pushing the peace process forward (and at the time, they were excoriated by the right-thinking in the Irish media for doing so, as well as some of our politicians).

      Both sides have had to compromise. Both sides, and their allies (whether in Britain or the Republic) have had to accept unpalatable measures (e.g. the ‘on the run’ guarantees) in order to get progress. We’ve (tentatively) managed to put aside the Armalite and concentrate on the ballot box – for now, at least.

    • sourcreamus says:

      The Battle of Algiers. The french took a city that was run by the insurgency and systematically destroyed the insurgency in that city. What they did was make a census so that they knew who lived there and where they lived. Then they arrested known insurgents and tortured them until they gave up all the names of their superiors. Then they arrested the superiors and tortured them until they gave up the name of their superiors. They did this until all of the insurgent leaders had been arrested, killed, or both.
      Algeria ended up becoming independent because democracy is fundamentally incompatible with colonialism.

    • John Schilling says:

      On the military side, I think drone warfare is likely to be the decisive counterinsurgency technique of this era (and possibly of all future eras, including the one where hostile AIs are suppressing human insurgencies). To date, the United States has been doing this only on a small scale and focused more on international terrorism than local insurgencies, but the results have been generally favorable.

      The knee-jerk response to drone warfare is generally, “But look at all the dead civilians! Obviously this is just going to drive the populace into the arms of the insurgents!”, but that’s usually not backed up by even anecdotal evidence. Empirically, drone warfare is associated with fewer civilian casualties than any other form of land warfare the human race has ever come up with, and the casualties are more closely associated with the actual enemy, e.g. civilians attending a social gathering at the home of the local terrorist or insurgent leader. The ability to conduct persistent surveillance at low risk before deciding on a strike, is particularly valuable against insurgents hiding within a civilian population, and you avoid all of the problems that occur when ground troops come under fire and insist on shooting back without spending an hour or three carefully assessing the targets.

      But more important than that, I think, is that drone warfare denies the enemy meaningful victories. Nothing is more destructive to an army’s morale than the persistent impossibility of victory, and an insurgency has little hope of keeping its combatants around when their morale goes south.

      Consider classic hearts-and-minds counterinsurgency warfare. The government’s pitch is basically,
      “Support us and we’ll build you schools and hospitals and we’ll protect you as best we can because we’re the Good Guys (it won’t be perfect, but we’ll try). Support the insurgents and, we’ll still build you schools and hospitals to show why you should support us instead of the murdering insurgents”. The insurgent’s pitch is, “Support us and you can help Throw Off The Shackles of Colonial Oppression, which is way more fun than being a farmer. Support the government, and we’ll murder your family, because the government can’t protect them 24/7 and nothing else is good enough”.

      I don’t care how much you value schools and hospitals, or how much you privately agree that the insurgents are murdering bastards, that’s a pretty lopsided argument. Very likely you’ll wind up believing that Throwing Off the Shackles of Colonial Oppression is indeed a worthy endeavor, for lack of any good alternative. And in the endgame, the insurgency can start setting up its own schools and hospitals, and they’ve won.

      With drone warfare, there don’t need to be schools and hospitals in contested areas, which would just be targets for the insurgents to attack. The government doesn’t need to ask for local support in the places where that would be too dangerous. And “Come Throw Off the Shackles of Colonial Oppression” means “Come Hide in a Cave and Hope to Die Later Rather Than Sooner”. Because the Shackles of Colonial Oppression are untouchably thousands of meters overhead, and watching to destroy any concentration of insurgent force that might take the war to the enemy.

      The insurgents don’t have a recruiting pitch any more. And the local population is faced with “colonial oppressors” who very occasionally kill them by mistake, and the insurgents who don’t have anyone to kill but the local population. In the endgame, when the insurgents have completely worn out their welcome, the government can start moving in to clear territory and build schools and hospitals, and they’ve won.

      The most viable strategy for the insurgents, in this case, is to become a nonviolent civil disobedience movement, in which case we aren’t in the realm of counterinsurgency warfare any more.

      • ThirteenthLetter says:

        I’m skeptical due to the seemingly ineffectual results from American drone campaigns so far, but on the other hand: the angry, reflexive opposition to drone weapons from all the usual suspects — the UN, “human rights” organizations, and so forth — does suggest they are worried this could actually work.

        • Pku says:

          Even in the more focused Israeli assassination policy (which focuses just on leaders rather than foot soldiers), there’s a tendency for intelligence analysts (Most former heads of shin bet, for instance) to be against it as a long-term strategy (at least according to my brothers, who follow politics a lot more than I do). Not an exact equivalent, but if anything I’d expect the focus on leaders to make it have a better effect/cost ratio (and Israel’s been doing it for a longer time, so there’s more data on long-term effects).

      • FacelessCraven says:

        @John Schilling – “The most viable strategy for the insurgents, in this case, is to become a nonviolent civil disobedience movement, in which case we aren’t in the realm of counterinsurgency warfare any more.”

        That’s actually a really interesting picture of drone warfare, and one my previous political views largely blinded me to, so thanks for that.

        At the same time, I wouldn’t expect Drone Warfare to be anything like a long-term solution. Insurgency was the answer to our perfection of conventional warfighting, and I’d expect some other method of killing people and breaking things will answer Drone Warfare just as thoroughly as it may answer insurgency. Maybe drone terrorism?

        • John Schilling says:

          Drone terrorism is constrained by the fact that drones can’t operate if the enemy has air supremacy and/or electronic supremacy.

          At the moment, the air forces of the world aren’t set up to use their supremacy against drones, so the first few groups to implement the “drone terrorist assassin” strategy will win some cheap victories. Within ten years of those cheap victories, the USAF at least will be able track every drone operating over any populated area in its domain, and override or destroy at the touch of a button the ones that are not fully compliant with Drone Traffic Control. I don’t see any likely way to trump that without more and/or better jet fighters than the USAF has, and that requires way more in the way of heavy industry than any insurgency is likely to command. This is a case where the asymmetry in “asymmetric warfare” is 100% in favor of the government.

          Other sorts of terrorism are a possibility, but one thing drone warfare has already proven good at is preventing terrorists from getting together to organize their efforts. Civil disobedience will I am fairly certain work better, at least against most governments.

          The nightmare scenario, and the reason I am not entirely happy with my observation that drone warfare trumps classic insurgency, is the government that deploys armed drones against nonviolent civil disobedience movements. I’m not sure that won’t work, and I haven’t yet found a good counter to it.

          • FacelessCraven says:

            @John Schilling – “Drone terrorism is constrained by the fact that drones can’t operate if the enemy has air supremacy and/or electronic supremacy.”

            I’m thinking less something like a predator, and more like a flying hand grenade or machine pistol. There’s nothing particularly novel about that sort of violence, other than it being possibly safer for the terrorist than deploying the gun or grenade themselves. Then again, one of the things that always mystified me about the War on Terror is why an implacable terror organization with global reach didn’t simply move gunmen over our borders and start an organized campaign of public shootings all over America. It always seemed like an obvious strategy to me, and one that would have done a hell of a lot of economic and morale damage to America than killing a few of our working-class soldiers and a bunch of local people we don’t give a damn about in some dusty place we’ve never heard of.

            “Within ten years of those cheap victories, the USAF at least will be able track every drone operating over any populated area in its domain, and override or destroy at the touch of a button the ones that are not fully compliant with Drone Traffic Control.”

            Would this be viable for small drones as well?

            I think you have a pretty good argument that the classic model of insurgency taking the countryside and besieging the towns and cities might be no longer viable. Does that make insurgency within the cities impossible?

            “The nightmare scenario, and the reason I am not entirely happy with my observation that drone warfare trumps classic insurgency, is the government that deploys armed drones against nonviolent civil disobedience movements. I’m not sure that won’t work, and I haven’t yet found a good counter to it.”

            Part of what makes drone warfare so effective is that the drone is ten thousand feet up, the pilot is a thousand miles away, and the country employing him is on the other side of the world. The pilot still has to live somewhere, though, and the country is a big place full of people who can be shot or stabbed or set on fire. In that context, I’m not sure how drones solve the terrorism problem any more than widespread CCTV deployment does.

          • John Schilling says:

            Without going into the details (I can if you want), yes, a competent first-world air force with reasonable support can keep its skies clear of even small drones that aren’t fully compliant with the National Drone Traffic Control System (a wholly-owned subsidiary of Amazon Hyperprime). There is some new technology that would need to be deployed, but mostly it would be a matter of changing the concept of operations and rules of engagement.

            Other sorts of terrorism are still on the table, of course, but most nations have been able to keep purely domestic terrorism down to a tolerable level. Terrorists with foreign support can be a major nuisance, but that mostly goes away when the relevant foreign skies are filled with counterterrorist drones. Terrorists supported by a foreign nation-state with an air force of its own would be a serious threat, but it’s the sort of threat that would justify outright war.

          • FacelessCraven says:

            @john schilling “Without going into the details (I can if you want), yes, a competent first-world air force with reasonable support can keep its skies clear of even small drones that aren’t fully compliant with the National Drone Traffic Control System (a wholly-owned subsidiary of Amazon Hyperprime).”

            I’d be very interested in the details.

            Someone passed me the youtube video earlier this week of a drone-mounted handgun being test-fired, and while clunky, it looked small enough to be disassembled and carried in a backpack. The VF40 minigrenade weighs a bit less than 5 ounces, and has a claimed lethal radius of 5 meters. It seems to me that a “flying grenade” drone using one of those could probably be made small enough to be concealed like a handgun. Minidrones like that, especially if they’re autonomous rather than RC and can be manufactured easily, seem like they’d be hell to defend against, and still seem like they’d give the insurgent a whole lot more standoff and increased chance of survival/escape than actually doing the shooting and grenade-throwing himself.

            “Other sorts of terrorism are still on the table, of course, but most nations have been able to keep purely domestic terrorism down to a tolerable level.”

            This is true, but I wonder whether it’s a tactics solution or simply a case of a larger political reality steering tactics a certain way. I guess I just find it hard to believe that a general solution to armed resistance is that close. If you’re right, that’s a seriously scary thought.

          • Glen Raphael says:

            yes, a competent first-world air force with reasonable support can keep its skies clear of even small drones that aren’t fully compliant with the National Drone Traffic Control System

            This sounds like a recipe for an arms race that ends with “small drones” that look and act most of the time like local bats, birds, and insects.

          • John Schilling says:

            Countering drones: First, if terror-drones are more than a nuisance, drone traffic control will be comprehensive and mandatory. Consider that every commercially available GPS chipset already includes hardware lockouts to prevent it from being used for missile guidance (and to the great inconvenience of amateur rocket builders). This will be extended to drone control chipsets generally. If your quadcopter isn’t talking to traffic control, it’s not going to fly. USAF cyber command will own everyone’s drones the way the NSA owns cellphones and servers, except control rather than just surveillance and with about zero controversy.

            Dedicated geeks will still be able to get unlocked drones in the air, and may be allowed to do so in remote areas. Anywhere else, they will be extremely conspicuous. And it will require enough technical expertise and specialized hardware that they won’t be very numerous.

            Detection: The radar on an F-35 can nominally detect an eight-pound bird at just under fifty kilometers distance. Obviously, it filters out birds. Drones, unlike birds, have shiny metal parts – even if the skin is plastic or composite, the wiring harness is a nice bunch of dipoles. And drones have rapidly spinning rotors that give a doppler signal like nothing in nature. Almost certainly going to leak RF from the motors and electronics as well. The filters necessary to sort drones from birds and ground clutter will reduce the range quite a bit, but even an order of magnitude won’t save the drone.

            When you spot a drone that traffic control doesn’t recognize, a high-power microwave beam will usually fry the electronics. If that doesn’t work, a kilowatt-ish laser will fry the optics. If that doesn’t work, dial the laser up to ten or a hundred kilowatts.

            We’ve already got the beginnings of this. The latest fighter-aircraft radars can digitally reconfigure themselves to a high-power jammer, and autotracking lasers to blind missile guidance systems are becoming common. Not ideal for drone-sniping, but it will a start.

            The F-35 is explicitly designed to take the dedicated microwave beam and the hundred-kilowatt laser when they become available; other aircraft can carry them in pods. The digital interfaces and onboard computing are flexible enough that this can be mostly automated. Hopefully it will still require the pilot push a button, and a boresighted FLIR will be giving him a picture of the target zone.

            And really, an F-35 is overkill for this. A militarized King Air would be about right, or an airship parked over every threatened city. You could even use high buildings as a vantage point.

            Or, of course, a bigger drone.

          • FacelessCraven says:

            @John Schilling – so, essentially a scaled-up version of the anti-mosquito laser system? That’s a pretty damn interesting idea. Thanks much for the write-up!

  40. onyomi says:

    Was longevity really as simple as sucking the blood out of young people all along?

    • Deiseach says:


      Yes, I know from the article he’s tired of vampire jokes, but if he’s involved with setting up a business that names itself after an alchemical universal solvent, he has to be prepared for this kind of response.

  41. irrational says:

    I know there are many economists here, so I am very curious what you think about of the work of prof. Keen. Here’s a link to a lecture of his that I am particularly interested in. TLDR version: he argues that the basic idea that having more companies in the market increases competition is wrong (firms in competitive markets mimic the behavior of the monopoly) . He presents both a simulation (which is very persuasive) and math (which is fairly persuasive – I can verify that the math itself is correct – though his assumptions may not be). According to him, if his argument is true, it would basically completely kill classical idea of supply curve and most of modern economics. Any takers to explain why he is wrong?

    • bluto says:

      By summing all the marginal cost curves to the monopolist’s marginal cost curve he’s making his 100 firms effectively independent divisions of a monopolist.

    • Adam says:

      I don’t really feel like watching that entire video, but just from viewing the beginning and reading the basic outline of his critique that empirical pricing behavior shows that firms charge cost plus markup, not a price that sets marginal revenue equal to marginal cost, seems to be making a basic error about what constitutes an economic cost. Firms set a price sufficient to earn a certain amount of accounting profit, but the argument of the perfect competition model is that this accounting profit should be roughly equal to the normal return to any other investment asset with a comparable risk profile, resulting in zero economic profit.

      Note also that it’s just an intro-level model. I guess I don’t know any working economists, but I can’t imagine that are very many, if any, that think the perfect competition model is a very accurate description of any real firm. It just builds intuition about how micro behavior creates aggregate effects in an idealized setting, but get past intro micro and you quickly move on to models involving cross-market effects, principle-agent problems, game theory, and even behavioral finance more recently.

      There has to be more to his argument than this, given how basic of an error that is, so if you know what it is, please tell.

      Update: Here are some blog posts claiming he is wrong. Okay, yeah, he had more of a point than indicated at a quick glance, which is usually the case, but he seems to be wrong about finding a basic error in the math of standard models.

    • Trevor says:

      His claim about killing the idea of the supply curve is very silly since economists have long known, outside of completely 100% perfect competition with 100% flat firm demand curves, that the relationship between price and supply is ambiguous. Empirically no firm faces a flat demand curve and so you might assume that the idea of the supply curve is just useless; however, also empirically the relationship between price and output is usually positive, so this fact doesn’t really matter much in practice and supply and demand is a good approximation in most situations. Neoclassical economics also already has the concepts of fixed costs and economies of scale, so it’s already well understood by neoclassical economists that output could increase and prices decrease as an industry becomes more concentrated in the presence of these factors.

      His claim that competitive profit maximizing firms would act like they were always colluding is simply based on a misunderstanding of the math. Really this guy is just full of bluster and doesn’t even have a basic understanding of mainstream economic theory.

      • irrational says:

        The simulation is still quite interesting though, don’t you think? If real firms simply can follow the gradient like he has them do in the simulation, and end up effectively colluding (and therefore making more profit), that would seem to undermine a lot of economic theory (which assumes that businesses uses an entirely different decision rule – which obviously most of them don’t, since they aren’t economists).

        • Trevor says:

          I’m not sure why his simulation gives that result, but if what he where saying where true then OPEC or any other cartel would have no business existing because trying to raise prices above the competitive equilibrium would only lower industry profits, if as he says, in the competitive equilibrium firms will always set prices and output at the level that maximizes industry wide profits. Also the prices people pay for food would be massively high as the demand curve for food is highly inelastic. His conclusions don’t match reality, whereas, Neoclassical economics at least broadly matches reality even if the assumptions can sometimes seem absurd.

    • Tracy W says:

      We do see monopolies in the real world, and they do things like withholding goods to drive up prices (eg De Beers in diamonds) which we don’t see in markets with lots of competition so either there’s an error in his maths or his starting assumptions are wrong.

  42. Esquire says:

    I think the banning of Dr. Beat is sad. Sure he is at fault for letting himself be trolled, but his counterparts were pretty clearly bad-faith trolling him, it seems to me.

    • Anonymous says:

      In my defense, you don’t have solid evidence that “male brain” and “female brain” are meaningful categories. If you did, we could take it over to the queer theorists and demonstrate to them how gender essentialism is true. It’s not really my fault that people go all apoplectic when pressed to provide a position that is internally consistent and reflects the scientific evidence, and I don’t think I should be called a troll for making such demands. I genuinely want to understand these issues from a scientific perspective, and you’re not helping when you just label my inquiry “bad-faith” just so that you can ignore it.

      • Esquire says:

        Lol I am not going to let you troll me too.

        I don’t have any particular view on this issue and I wholeheartedly support you trying to genuinely understand it from a scientific perspective. But, on the off chance you don’t know what you are doing, I’ll point out for your edification that:

        1. you are combatively (and incorrectly) assuming that I have a contrary view to yours, and
        2. you are characterizing my unrelated statement as being made “in order to ignore” you.

        That is what bad-faith trolling looks like. A non-trolling version might look like: “I really think what you observed was a substantive factual disagreement and disagree with your characterization of my statements as trolling. Could you point to something specific that you considered bad-faith so that I know you are not just being a jerk?”

  43. Esquire says:

    I have an economics question.


    My basic understanding of the macro policy recommendation of mainstream Keynesianism is: slack in labor market -> fiscal and/or monetary stimulus.

    My recollection is that during the crisis, Krugman (synecdoche for mainstream Keynesianism) was constantly calling for massive fiscal stimulus when monetary stimulus was tapped out at the zero bound.

    My recollection is that the response to folks who were moralists or long-term worriers about the aggregate national debt was: don’t worry because rates are low, we borrow in our own currency so can’t really be forced to default since we can print it, and debt debt levels are not that high historically.


    None of my recalled responses provided a real limiting principle.

    What if debt levels were high and interest rates were high? Should a country in recession go into an arbitrarily high level of debt and let interest become an arbitrarily high level of the budget to provide the required level of fiscal stimulus? What if you need to print money to pay the debt but then the slack leaves the labor market and printing money is no longer called for?

    More generally: is there indeed a limiting principal for how much to borrow and spend when there is slack in the economy? What are the implications if yes or no?

    This is fairly relevant for Japan today, not purely academic. So far acting as though there is no limiting principle (contra previous decades) seems to be actually correct?

    • brad says:

      The real constraint isn’t debt load. It’s not necessary to issue in the first place, you can just print and spend money without going through the debt dance. The real constraint is inflation. There’s a lot of cargo cult thinking around inflation and how to keep into from getting out of control. Some people claim the debt ritual is needed to keep the inflation monster in check. To me it makes the most sense to treat inflation as an empirical fact to be observed. So long as inflation is in check and the labor market is slack keep the firehose open.

      N.B. Krugman wouldn’t subscribe to quite this analysis as it goes past mainstream Keynesianism to a school of thought called MMT.

      • Adam says:

        I’ve had semi-serious discussions with just my wife that I’m starting to take more and more seriously about just financing all government operations this way. Stop issuing debt. Stop taxing. Just print money. Inflation will go up quite a bit, but say it goes up to 25%. Okay, that’s roughly the percentage of GDP eaten up by taxes right now. We just created a tax that doesn’t privilege any form of income-generation over another and is impossible to evade. Businesses no longer need to drain billions a year paying tax attorneys and accountants to consider the tax implications of their decisions. You can sell yourself to a holding company based in Bermuda, but as long as you hold dollars, you still pay the tax.

        The obvious fly in the ointment is inflation is tremendously sensitive to exogenous variables we have no direct internal control over, like the desirability of the USD as a world reserve currency, which would almost certainly go out the window. It all comes crashing down as soon as ‘as long as you hold dollars’ is no longer a slam dunk yes answer. It only works if every country does it.

        • Trevor says:

          The deadweight costs of inflation are actually extremely high and almost any other tax will be more efficient and have more equal distributional consequences as poor people hold more of their wealth in cash than rich people who hold interest bearing assets. This is even more true in the USA where the capital gains tax is not indexed to inflation so you will also increase capital taxes massively and those are some of the worst taxes.

          • Adam says:

            Well, obviously the distributional consequences would be horrible. It’s already regressive just as a flat-tax, but also kills anyone whose primary holdings are cash. It’s totally a fantasy proposal, not a real idea.

          • walpolo says:

            What’s the definition of “deadweight costs”? Not a term I’m familiar with.

          • Adam says:

            Efficiency loss from economic incentive distortion. Consider a taxation example that imposes a 30% tax on working for McDonald’s and a 20% tax on working for Burger King. You’re going to cause a certain number of people who would have derived greater utility working at Burger King at the relative wages each offers if they were taxed equally, deciding to work at McDonald’s instead.

            In the real world, think of whatever actual resources get poured into structuring takeover deals so American medical device manufacturers can be owned by a firm in Ireland, for no reason other than to avoid tax. Whatever the economy might have gotten if they hadn’t done that – more R&D, a share buyback, whatever – is a deadweight cost.

          • Tracy W says:

            @walpolo: deadweight losses are those losses from the transactions that didn’t happen because of the tax or other policy. Eg I’m willing to hire a uni student to mow lawns for any price up to $10 an hour, the student is willing to work for any price above $8 an hour, we find an agreement somewhere in between.
            Now introduce a tax of $3 an hour on labour income. I still only value the lawn mowing at $10 an hour but that means the most the student will get in hand is $7 an hour so no transaction takes place. (Numbers picked for ease of illustration, not realism.)

            The key point is that everyone loses: the government doesn’t get paid any taxes from their tax, I don’t get my lawns mowed, the student doesn’t get a part-time job.

            The higher the tax, the greater the number of transactions that don’t take place and thus the higher the deadweight losses.

            Inflation has deadweight losses from people not being willing to hold money and probably other causes I’m not thinking of right now.

          • walpolo says:


          • bluto says:

            The easiest way I’ve come to think about deadweight losses is a decision that changes due to a government action. They’re more complicated than that, but that’s a quick approximation.

            So each time someone spends money (or spends more money) in a high inflation environment that they would save in a low inflation environment that’s adding to the sum of the dead weight loss.

        • onyomi says:

          Also, I think this would give an even greater unfair advantage to any business which deals with the government than they already have: prices go up only after new money has filtered into the economy. Therefore, the people who get new money first have the advantage of being able to buy stuff with it before prices go up.

          So basically Lockheed Martin gets more value for its money while Joe Schmo’s dollar doesn’t go as far at the grocery.

        • brad says:

          You can do the get rid of debt part without the get rid of taxes part. Under the MMT way of thinking taxes are one tool for controlling inflation by destroying money.

          That said, you may want to keep issuing at least some debt for use as a risk free asset. The main idea is to stop conceptualizing the fiat currency issuer as a giant household with a “salary”, “bills” and a “credit card”.

        • Adam says:

          To be clear, Mark, I have no say in U.S. tax or monetary policy.

        • Tracy W says:

          Also, in the 1960s and 70s people started writing contracts which included inflation clauses. So the prices of the government’s inputs will start rising at 25% per year automatically.

        • Mark says:

          Seems to me… problem is… we’ve got a system where only money created by big banks can be used to pay government tax demands, and that every single transaction does create such a demand.
          For anyone wishing to engage in any kind of economic activity, there is no escape from the government-private financial nexus – as such, it really makes no sense to talk of private money (or perhaps we should talk of privately owned government). The system dictates that we must have this money before we can do anything – to me it is deeply mean-spirited of the financial institutions to demand payment for providing us with that which they insist (violently, through force of law) we must have (payment of interest on government debt before fiscal policy can be pursued).
          Obviously, this combination of private finance with government power has been an extraordinarily successful way of creating powerful nation states, but we should never forget that this is exactly the purpose of our financial system: creating a broad based powerful state by co-opting business into government (or alternatively, by giving business control of government).
          Anyway, if the objection to government creation of money is that it destroys the value of private money, I have a suggestion: only money created by government can be used to pay tax – lets separate out the private and the public. No more government debt! Hooray!
          I’m guessing that the financial institutions won’t be too keen on that one.

          • HeelBearCub says:

            Do you really think the existence of fiat currency is the result of a system that is designed to “creat[e] a broad based powerful state by co-opting business into government”?

            My understanding of the existence of current fiat currency is that it flows logically from the first creation of currency at all. Currency is desirable because absent currency it is very hard to trade labor. Private currency is undesirable because it is inherently less universal, and therefore less useful in trading labor. Commodity currency is inherently less useful in trading labor because it is not backed by labor itself, but depends on the commodity.

          • Mark says:

            The important thing to remember about fiat money is that it is only successful to the extent that it *is* a commodity.
            Let’s say that there are two types of money – commodity and debt. The difference between them is that the value of debt is determined by our trust in the person holding the liability – commodity money can be used more extensively because there is an expectation that the money itself will be widely viewed as having some value.
            Whether government money has value because it is made of gold, because the government has said it is money and there is an expectation that others will view it as such, or because it can be used to extinguish tax liabilities (3rd is likely most important) it is still a commodity.
            Our current system is a combination of government fiat/tax money with private debt.
            Yes, this system was (more or less) explicitly established as a means of creating a powerful state backed by business.
            17th century England: English mercantile class rejects absolutism -> glorious revolution -> William II established as a constitutional monarch with limited ability to tax -> establishment of the Bank of England allowing financiers to both receive interest payments from the state and issue currency on the basis of the state debt -> England is able to finance its navy, and becomes a major world power.
            This financial system proved so effective that it was adopted throughout the world – financiers graduated from ruling city states to ruling nation states – and the nations with the closest integration of the business class into government proved to be the most effective.

          • HeelBearCub says:


            A) That’s a chain of events, not an intentional reasoned strategy. The merchant class didn’t reject absolutism because they wanted to establish a powerful navy.

            B) Money represents labor. In its simplest, two party, private form, money is simple a promise to provide a good or service in the future. I need a chicken now and promise to give you grain in the fall when the harvest comes in. You need shoes now and give my promise of grain to the cobbler.

            You can say this is a commodity, but it is not. My IOU is only good to the extent that you and the cobbler believe I will do the work necessary to produce the grain.

            And tying that promise of labor to a specific commodity represents a problem. In the event of a bumper year you are now holding something much less valuable. Conversely, in the event of a drought I have great risk (and you risk a default on my obligation).

            Far better to make a generic currency. Once that occurs you need a mechanism to make sure it isn’t counterfeited. First you do this with a rare commodity (cowrie shells, very large milled stones, silver, gold). Eventually the problems of universality and counterfeiting lead you inexorably to a government printed currency. Printed on gold, then backed by gold.

            But eventually, the underlying value of your national labor starts to outstrip the supply of gold. Gold is a fixed resource but labor is growing, now you have deflation, which is extremely painful for anyone who has not already made a pile of money.

            So you get rid of the gold standard and issue fiat currency.

          • Mark says:

            Well… I suppose to some extent, everything that happens is an expression of underlying historical processes – but at the same time, these institutions were designed expressly to both strengthen the state and enrich financiers (who, along with their rich friends, were in control of the state). Is there anything else that needs to be said?

            “You can say this is a commodity, but it is not. My IOU is only good to the extent that you and the cobbler believe I will do the work necessary to produce the grain.”
            I wouldn’t say that that was commodity money – I would say that it was debt money as in “I owe you” ~ (“the value of debt is determined by our trust in the person holding the liability”)

            “Far better to make a generic currency. Once that occurs you need a mechanism to make sure it isn’t counterfeited…. First you do this with a rare commodity… inexorably [leading] to a government printed currency. ”
            I don’t really disagree with you when you say that these different technologies had various advantages – I just don’t think that (historically speaking) they developed as some kind of inevitable result of those advantages, and not necessarily for the convenience of the people using the money.
            I mean… as far as I’m aware, money wasn’t established to enable two guys on a village to better exchange produce – it was established as a means of accurately measuring the tax liability of peasants to despotic governments.
            An example: the present-day institution that is closest to the village is probably the family. Well, is it really a concern for me to have some impossible to counterfeit token for the dinner that my wife owes me in return for me doing the washing up? It just doesn’t make any sense.
            But yes, when you have the mass society,when transactions take place with people you don’t know, you need some means of exchange – hence commodity money.
            (Also – the process hasn’t been a straight line – coins have come and gone and come and now gone again… we’ve gone from silver to virtual currency to tally sticks to gold to paper… )

            Anyway, the main point is that what is unique about our monetary system is that it is essentially debt based, with all the advantages of debt based money, but that this debt has value even to those who know nothing of the people involved because it can be used to extinguish tax liabilities (it has a real value as a “pass” enabling us to do business/live in society without getting sent to prison).

          • Mark says:

            Also, I would say that your story makes no sense (at least in the order given) because the IOU you start off with *is already perfectly abstract* – we are apparently going through these deeper levels of abstraction in order to find the perfect money – but we essentially end up where we’ve started from (this is actually a key point in that we have managed to make an entirely abstract mass currency).
            But it couldn’t have been a need for greater abstraction alone that led to this process – the two villagers could have just denominated their debts in “giggles” or something to start with.

          • HeelBearCub says:

            Do you have some citation that says money was invented to allow pay,net of tax. My understanding is that cowrie shells as a means of money was simply an invention of those people, not some “government”.

            And no, my idea is not “perfectly abstract”. How many bushels of grain did your iPhone cost? Trying to take a chit promising 4 bushels of grain in the future and trade it for shoes is not a pricing strategy that allows merchants to declare the price of their goods. What’s the conversion rate between live chickens, bacon, a pair of pants, shoes, a flask of mead and bushels of grain? People carrying around obligations for all of these items will (did) want some way of using a common unit of measure, rather than trying to convert between all these various items.

        • Anthony says:

          Just print money. Inflation will go up quite a bit, but say it goes up to 25%. Okay, that’s roughly the percentage of GDP eaten up by taxes right now. We just created a tax that doesn’t privilege any form of income-generation over another and is impossible to evade.

          No, you’ve created a “tax” which privileges a whole different set of recipients – in this case government employees and contractors and welfare recipients. That may actually be an improvement over the current set of economic actors who receive inflated money first – banks and people in the financial markets – as the money may be less likely to create financial bubbles. But that’s hard to guarantee in advance.

          Inflation is actually less subject to exogenous variables than you think. Oil price shocks, which are about the only external shocks which signifcantly impact the measured rate of inflation, do so by reducing the quantity of goods available to be purchased with the existing amount of money, since damn near everything involves oil directly or indirectly as a major cost component. Almost everything else that’s considered an “external” driver of inflation is actually driven by the initial mechanism of inflation – the increased supply of money, and how that increased supply works its way through the economy.

          However, creating inflationary expectations of about 20-25% per year is going to wreak havoc on all sorts of markets, and the consequences Mark Atwood predicts will eventually prove correct.

        • FJ says:

          @Adam: people have tried financing government deficits by printing money before, although I don’t know of anyone who tried to finance a deficit of 100% of government spending by inflation before. You’re wrong about exactly what happens, though. You get accelerating inflation, i.e., hyperinflation, not a constant inflation rate. This has nothing to do with exogenous variables and everything to do with the fact that people will try to stay ahead of a high inflation rate.

    • Trevor says:

      Optimal policy in the New Keynesian model is to stabilize aggregate demand with monetary policy, as long as interest rates will one day not be zero then you can always do this in the model. As long as aggregate demand is stabilized through monetary policy fiscal policy will not effect inflation or the output gap and borrowing should only be done for microeconomic concerns.

      If interest rates will always be zero then a government could borrow and much as it wants and never have to pay any interest on that debt, so yes in this highly unrealistic case then there is no limit to the amount they could borrow.

      However, that case is highly unrealistic because land at the very least pays positive rents and of course private capital exists that will probably pay out positive rents when the economy returns to its long run equilibrium. Therefore a limit does exist as interest rates will one day rise and if you have too much debt you can’t pay off the interest and that raises the risks of default or hyperinflation further raising interest rates and you end up like Greece.

    • John Schilling says:

      Japan does have the highest public-debt-to-GDP ratio in the world, at 226%. This seems to have resulted in twenty years of economic stagnation, but no collapse. So that’s semi-promising.

      The second-highest, at 203%, is Zimbabwe. Coming in at number three, Greece with 161%. Portugal, at 129%, is high on the list of European nations considered likely to fall if the Greek collapse spills over.

      There does come a point where you can’t ignore debt any more. If you default, people will stop lending to you and you’ve already become addicted to deficit-financed social welfare spending. If you print money to pay your debt, hyperinflation eventually sets in. Raise taxes to pay debt, and you’ll find that the Laffer curve actually does have a back side. Cut spending to pay the debt, and your people will elect an outright socialist.

      People disagree with exactly where the edge of the cliff is. Japan certainly benefits from strong, stable institutions in this regard. But there is a limit, and for most nations it’s probably shy of 200% of GDP.
      Even for Japan it’s probably less than 400%.

      Believing otherwise is like believing that your’s stock price will rise every year forever even though your business model absolutely precludes profit.

      • HeelBearCub says:

        Greece is a bad example of the question at hand. Greece is in the Euro. The don’t control their own monetary policy. I’m fairly certain Krugman has inveighed against the Eurozone’s separate fiscal policies with a single currency on many occasions.

        • Adam says:

          I feel like 161% is pretty high even if you’re borrowing in your own currency, but to be fair to Greece, that isn’t all debt growth. A lot of the debt-to-GDP ratio growth has been driven by the fact that its GDP has dropped 34% in the last five years. Tanking your income is just as good a way of becoming unable to pay back debt as taking on too much debt.

          • HeelBearCub says:

            If you borrow in your own currency, you can always pay it back, because you can print money. Given a fundamentally sound economy, that simply has suffered a demand shock, this does no harm internally and merely devalues your currency vs. the rest of the world.

            That devaluation is a further benefit, as it discourages holders of your currency from purchasing foreign goods and encourages foreign purchases of your goods.

            The key phrase in all that is “fundamentally sound economy”. Monetary stimulus won’t work if, say, a huge disaster (like a decade long drought) idles your workforce because work can literally not be done. It’s possible that Greece has fundamental problems that go beyond just sudden retraction in demand.

            Regardless, Greece cannot afford itself of that lever, and monetary stimulus in the Eurozone overall doesn’t provide the kind of differential benefits to Greece that drachma stimulus would. Just as one small example, think of the benefit a weak Drachma compared to a Deutschemark would provide to Greek tourism, which is one of their major economic drivers.

          • Adam says:

            You can always pay it back, sure, but outright default isn’t the only undesirable outcome. Having debt service as 50% of our federal budget is still a real constraint keeping you from doing other things governments are supposed to do in the absence of a legislature that will just appropriate 50% more money, which for the most part, they won’t do.

          • HeelBearCub says:


            That’s a fair point, but I think it just begs the question of whether the economy is fundamentally sound.

            If Greece had a fundamentally sound economy and control of their fiscal and monetary policies, they could grow their way out of that bad debt to GDP ratio. As you said, their GDP has dropped 34%. Recover that and the long term debt service numbers are fine.

        • sourcreamus says:

          If fiscal stimulus only works when monetary policy is right, doesn’t that lead to the implication that the monetary policy is what is doing the work and the fiscal stimulus is superfluous?

          • Brock says:

            Yes, but conventional monetary policy (targeting interest rates) is ineffective when interest rates hit the zero lower bound.

            At that point, you have to resort to unconventional monetary policy (i.e. just buying stuff), which differs from fiscal policy only in who decides what to spend the money on.

          • HeelBearCub says:

            I think that Keynesians would contend that monetary and fiscal policy should work in conjunction.

            In a recession of a fundamentally sound economy, you have a demand shock. Demand for labor is greatly lessened. This can go downward in a vicious spiral.

            Monetary policy puts more money in the hands of the private market, making it easier or encouraging them to spend it. One way to describe this is that the credit market is loosened. More money available makes more available to lend. If money is lent, it will be spent, which becomes a demand for labor.

            Fiscal policy works by demanding goods or labor directly, thus it can be a more efficient stimulus (in some senses). Or it provides money directly to the idled laborer which bypasses the inefficiency of demanding a good or service and directly addresses the spiral of “lower demand -> lost wages -> even lower demand”

            In conjunction, when the government demands a good directly, the money they spend should encourage investment in new ventures somewhere down the line, which increases the effectiveness of the easing of credit that the monetary policy provided. Both monetary and fiscal stimulus amplify each other to increase demand.

    • Adam says:

      This is kind of an open question given the somewhat unorthodox behavior of the last seven years. Honest economists will tell you their prior models and beliefs have largely not held up. Actual living economists will bend the facts to fit their pet theory.

      On just a bare accounting level, obviously there is a difference between having 10% of your budget eaten up by debt service and being faced with a 0.03% ten-year rate, versus having 90% of your budget eaten up by debt service and being faced with a 13% ten-year rate. So sure, in the former case, if you otherwise have a good case for debt-financed stimulus spending (that is, you have good reason to believe stimulus will work), go for it.

      On the other hand, if you’re doing it because you believe monetary policy can no longer be effective at a zero risk-free rate, then the Fed just spent much of the last seven years proving you wrong because, after ARRA, we didn’t do any fiscal stimulus, yet recovered faster than much of the rest of the world that doesn’t just have intrinsically faster growth because they’re less mature economies with more low-hanging fruit, partly because the Fed proved that just dumping a crap-ton of money into an economy works if the structural elements necessary for inflation aren’t present and your currency is still more desirable than everyone else’s.

      At the far end of the contrarian spectrum you get modern monetary theory, and frankly, they can explain themselves better than I can.

      Edit: I just wanted to add that I think it’s worth noting that at many times in the last seven years the US Treasury has been able to borrow using short-maturity instruments at a negative rate, essentially making money by borrowing, and to the extent this is borrowing from the Fed, this functionally is just ‘printing money.’ It’s been somewhat of a unique situation.

      • Trevor says:

        Why do you think the last 7 years have proven everyone’s models and priors wrong? The only false prior I can see is the widespread belief economists and business men had that their central banks would stabilize inflation and employment.

        Also I am not sure why you think the Fed buying treasuries with negative returns is functionally ‘printing money’ when the Fed buys all assets with printed money, so any open market operation they do irrelevant of the return the assets bought provide is printing money.

        • Adam says:

          The Fed is printing money either way. If the Treasury has to pay the Fed back a positive interest rate, it’s borrowing from another party that prints money. But if the Treasury doesn’t have to pay for the money at all and can sometimes even earn money by borrowing, it’s borrowing in name but functionally equivalent to the Treasury printing its own money.

          • Trevor says:

            The Fed gives all it’s profits back to the Treasury. The Fed is a part of the federal government, so whether the Fed or the Treasury actually prints the money or has the bonds on its balance sheet doesn’t really matter except to government accountants who have to keep track of all this stuff.

          • Adam says:

            I was actually under the impression that Fed profits went to its member banks, but I guess not, so you see the source of my error.

          • Trevor says:

            It doesn’t usually matter what the Fed does with its profits since they are fairly low compared to US gdp and money printing mostly has its effect on the economy through the monetary effects of increasing the money supply and not the fiscal effects of giving the federal government more revenue.

            Some people though forget all monetary policy is already fiscal policy since the Fed is part of the government. They start saying that we need helicopter drops of money (fiscal policy financed through money printing) to increase aggregate demand and that the Fed should stop buying bonds and just give money to the Treasury.

          • Adam says:

            Yes. It also makes no difference to the economy whether the borrowing rate is 0.03% or -0.03%. I just thought it gave me the chance to edit in a last-minute throwaway quip.

            They start saying that we need helicopter drops of money (fiscal policy financed through money printing) to increase aggregate demand and that the Fed should stop buying bonds and just give money to the Treasury.

            This was mostly what I was referring to with the earlier bad models comment, the belief that monetary policy won’t work at the zero bound and we need fiscal stimulus.

          • Trevor says:

            OK it seems like our only disagreement is what the state of macroeconomics was like before 2008. To me it seems like most economists didn’t believe in liquidity traps prior to the recession. Even Paul Krugman was critical of Japan for its unsustainable deficit spending and not doing enough on monetary policy in the 1990s. Papers like this one by Lars Svensson show how in the New Keynesian model there is no liquidity trap.

            I think the whole liquidity trap fad in the profession is an irrational and emotional reaction to a huge crisis the same kind of thing you saw during the Great Depression with the rise of Keynesianism.

          • Adam says:

            To be fair, you might know a lot more about this than I do. My impression of economics comes largely from economists who happen to maintain popular blogs. I don’t exactly keep up with the actual state of the field.

  44. onyomi says:

    I’m getting quite upset looking at Facebook right now, not because I’m strongly pro-life, but because I’m seeing a lot of glib, callous, and illogical arguments defending Planned Parenthood, including “they do other things besides abortion,” and “the world is overpopulated anyway.” And this got me thinking about something I mentioned briefly in the previous thread, but which might be worth exploring a bit more here if people are interested:

    Namely, what is the view of people here on world population? Is it too big? Too small? Irrelevant? It seems to be an unreflective true-ism among the general public now that the world is overpopulated–this despite the total failure of doomsday predictions like “the population bomb” and the sorts of demographic crises places like China with an enforced low birth rate, or Japan, with an unenforced low birth rate, are experiencing.

    It seems to me that Julian Simon has been proven totally right that human beings are the most precious resource. The more human beings we have (and especially human being with access to first-world level education–which is getting much easier with the internet, though I am also strongly in favor of free immigration), the more the economy grows and the more likely are revolutionary innovations. With today’s technology already we have long past the point where each additional person, on average, contribute’s more than one person’s worth of goods and services, so more people are a net plus just generally, to say nothing of the innovations that might be possible if 10,000 super geniuses who wouldn’t have been born are born, along with another 10,000 who would have languished in desperate poverty but now have a chance to access the world’s information.

    But it’s so expensive to live in New York, San Francisco, Tokyo… you cry! I look around and there’s sooo many people it’s already uncomfortable! And what about natural resources? I have lived in China. Believe me, I know what crowded is like. But I also know that even in China, 99% of the land is just… empty. Have you flown over the United States? What do you see? A whole lot of nothing! This is true of even the most populous countries! There is tons of usable space, especially now that we don’t need to be right next to a river to get our crops, etc., only, in many cases, there are restrictions on freedom of movement, freedom to build, etc. etc. If the world population were to double that wouldn’t mean twice as many people living in New York and San Francisco. It would mean five new Chicagos and Atlantas, most likely, and three more Shanghais. Would that be a bad thing?

    And as for resources: read Simon. Every time we think we’re running out, we find a way to access more: consider fracking, etc. Plus, what if we have 10 more Elon Musks inventing new ways to power things?

    • LtWigglesworth says:

      The way I see it is that there are 3 things that we have in balance: Global equality and quality of life, Environmental sustainability (in all the forms that that encompasses), and Population size.

      We can pick two.

      We could support the entire world at a lower quality of life, (but more equal) than that which we in the west enjoy now, assuming a slow rate of population growth. However, this would require international co-operation and people in the west accepting a cut in living standards so that someone in Laos can live at a higher standard.

      If we were to try and raise the standards of living in the rest of the world to the same as in the west then we will very quickly bugger then environment.

      We could sustain a smaller population at a high standard of living and prevent environmental degradation.
      But then you have to choose the populations you want to remove/ prevent from breeding/ whatever.

      • onyomi says:

        “If we were to try and raise the standards of living in the rest of the world to the same as in the west then we will very quickly bugger then environment.”

        Why is that? The United States is cleaner today than 100 years ago. Yeah, new industrial revolutions are generally messy, but having a vastly increased number of smart people with access to first world education means more clean energy innovations.

        • LtWigglesworth says:

          Because it would be a global increase in energy usage of about 380% (based on kg oil per annum equivalent. Global average s about 1800, US in 6800).

          To do that, and quickly, would required either:
          A massive increase in oil, gas, and coal . Which would cause massive increases in GHG emissions (No carbon capture technology is current commercialised on a large scale).

          A massive increase in production of advanced battery and solar tech, which will quickly run into limits on rare earth metal.

          A massive increase in the deployment of nuclear power.

          A rapid deployment of fusion power (something thats not planned to break even in terms of of energy production until the mid 2020’s and the first commercial plant is planned for the 2050’s).

          There is no good source of all that required energy that is both cheap, and environmentally sustainable.

          The environmentally sustainable ones are uneconomical without global collaboration that results in environmentally unsustainable power generation methods paying fairly for the negative externalities that they cause.

          • LtWigglesworth says:

            Also, one reason that the US use of energy could be even higher than the figures given is due to the importation of a lot of manufactured goods from china.
            Which counts towards China’s energy usage, not America’s. Despite contributing towards the standard of living more in the US than in China

          • onyomi says:

            Well, how fast is “quickly”? Even if we could convince much of the developed world (where birth rates have fallen) that they were wrong about overpopulation and the desirability of big families to the extent that each family had, on average, one more child than they would have otherwise, the population isn’t going to double overnight.

            Rather, it would gradually get bigger to the point that it would be much bigger by say, 2050. At which point maybe we will have workable fusion. Or something else invented by some kid who might never have been born. Either way, the more people in the world, especially the more people with access to first world education, the faster we’re going to get that better technology (and/or new technology for tapping previously unknown sources of rare earth metals–as Simon points out, production of all natural resources has continuously gone UP over time, not down).

          • ThirteenthLetter says:

            “To do that, and quickly, would require… A massive increase in the deployment of nuclear power.”

            So what you’re saying is that there’s no problem, then.

        • James Picone says:

          US uses 20,700,000 barrels per day in 2005, according to wikipedia, here. Official US population clock gives a population of 318,900,000ish in 2005. 0.065 barrels/person-day. Multiply by 7 billion, we’d expect usage of ~450 million barrels per day. Some googling around the largest sensible oil reserves number I’ve seen is 1200 billion barrels of oil, so ~7.3 years at 450 million barrels per day. 900 billion seems to be the low-range Serious Number, which gives ~5.5 years before exhaustion. That’s actually more time than I expected, to be honest.

          Caveats: I know nothing about fossil fuel reserve estimates, and I haven’t attempted to take into account new discoveries. I think the reserve number I’m using is the appropriate one – ‘probable’ – but again, I don’t know much here. Obviously people would use alternate forms of energy in this example, because the oil runout would be pretty obvious. I don’t know whether things like shale oil go into this. Etc. etc.

          But broadly speaking, I think that suggests that resource exhaustion problems would pop up if everyone on Earth lived like a first-world citizen.

          • onyomi says:

            I mean, it seems counter-intuitive but production of oil and gas has only gone up in recent decades. If we were anywhere close to running out one would expect diminishing returns to have set in already.

            Yes, it might be burdensome or just downright impossible for 6 billion people to enjoy US living standards *with the current level of technology* (not even sure that is the case, but for the sake of argument), but let’s say the two possible futures we’re looking at are:

            A. In 2050 the population of the world is 15 billion

            B. In 2050 the population of the world is 10 billion

            I think, all things equal, A is likely to be a better future, all told, than B, especially if we accept a utilitarian calculation that 15 billion people of happiness level x is, in some sense, “better” than 10 billion people at happiness level x. (But I think A would actually be 15 billion people at happiness level x+some positive value).

            Can I imagine scenarios in which A is worse? Yeah, maybe if only the rural poor have a ton of children while people in the developed world have ever fewer children but clamp down their borders to keep the unwashed masses out. But that seems unlikely, and is all the more reason to encourage first world people to have more children (or at least not have fewer children than they might like to on the theory that “the world’s already overcrowded”).

            But given that we will probably arrive at 15 billion people eventually, and given that A is probably a better world than B (and more technologically advanced–technological advancement which every precedent proves outstrips the depletion of resources using current methods), why not get there faster? Of course, population is 15 billion *tomorrow* due to genie would be a disaster, but that is not an option, anwyay.

      • Deiseach says:

        My cynicism here is that people who argue “I don’t want to have children/people should not be having children because that is using up too much of the planet’s resources” generally, when they free up the money/time that would otherwise be spent on those children, don’t tend to “sell all they have and give to the poor”, they up their standard of living.

        So tech toys, better goods, holidays, holidays for longer and in more exotic places, clothes, dining out, etc. etc. etc. take over from “spending money on raising and educating children”.

        In other words, as I’ve seen said, it’s a case of “Just enough of us, way too many of you” and we’re the ones with the enviable and desirable standard of living telling the rest of the world “No, you are taking up too much room and too much resources and the world can’t sustain it!” We won’t give up our current standard of living and (under the appearance of worrying about the state of the planet) we protect it by discouraging others from seeking to attain it – at least, not if it’s the acceptably limited number of new people aspiring to it, a number that will enable us to keep our standard of living without stinting.

        “If these shadows remain unaltered by the Future, none other of my race,” returned the Ghost, “will find him here. What then? If he be like to die, he had better do it, and decrease the surplus population.”

        Scrooge hung his head to hear his own words quoted by the Spirit, and was overcome with penitence and grief.

        “Man,” said the Ghost, “if man you be in heart, not adamant, forbear that wicked cant until you have discovered What the surplus is, and Where it is. Will you decide what men shall live, what men shall die? It may be, that in the sight of Heaven, you are more worthless and less fit to live than millions like this poor man’s child. Oh God! to hear the Insect on the leaf pronouncing on the too much life among his hungry brothers in the dust!”

      • DES3264 says:

        I’d love to read some good studies of the first option. If we aim for a sustainable population of around 10 Billion, and roughly equal standards of living everywhere, is that standard of living Uganda? Vietnam? Brazil? Spain?

        Of course, it is easy to take the world GDP and divide by population, but I am more interested seeing estimates for the world capacity if we remove terrible governments and other blatant inefficiencies.

        • Adam says:

          The ratio of world-average ecological footprint to U.S. footprint is 2.7:8, so if we take that as a decent starting point of production capacity (not necessarily replacement capacity, but at least a number we know is achievable with current resources and technology), we can say somewhere around 24.5% to 35% of U.S. consumption is a first-order approximate interval of what an average world citizen could expect to consume if we distributed everything roughly equally between countries, with the lower bound assuming no capacity increase and the upper bound assuming a capacity increase exactly equal to population increase.

          I’m not sure that’s actually a measure more robust to failed-state inefficiencies than economic output, though. Trying to figure that out would take some legitimate model-building well beyond what can be estimated on the back of an envelope.

    • Adam says:

      I don’t feel like this is a meaningful question to ask about the entire world. Texas is underpopulated. India is overpopulated, at least in the cities.

      • onyomi says:

        Yeah, in the cities. So maybe the question to ask is how to get populations to spread out more? I think improved communications, internet, cell phones, etc. are certainly already helping.

        And I do think it is meaningful to ask with respect to the world in general: sure, some places are overcrowded and many places are empty, but, in general, is it a good idea for people to be having more children or fewer (setting aside the question of individual family’s emotional satisfaction, though I think even that might favor big families). Is the *planet* overcrowded, or is it just a few tiny niches near historical ports and fertile rivers?

        That is, when people commonly say (as they very commonly do, in my experience) “there are too many people in the world, anyway,” or “don’t want to bring one more mouth to feed into this crummy world,” etc. etc., is it justified?

        My contention is no. The world is a better place to live now than it has been at any time in history (perhaps not coincidentally, there are also more people in the world now than there have been at any time in history), and each additional child you have will, on average be a net benefit to humanity, not a net cost.

        • Adam says:

          For the world, sure, but I think most people’s calculus assumes their children are going to live where they live, not move to Montana. The problems we face are largely distribution, but I think the LT’s concerns above become real if we assume a world that consumes at American rates. That’s the other important variable. We’re not overpopulated as a world because most people in the world have and consume virtually nothing. It’s the amount of resource consumption that matters, not the number of people.

          My own argument for not having kids, aside from not wanting any in the first place, is the world is going to grow perfectly fine without me. Your concern about low growth seems to be limited to a small number of places. The world as a whole has still yet to hit the inflection point to downward concavity.

          • onyomi says:

            “most people’s calculus assumes their children are going to live where they live” Really? In 21st century America? This seems increasingly untrue. I mean, I’m sure it’s still true for a lot of people, maybe even most people, but rapidly less and less true.

          • Adam says:

            Not the same place necessarily, but an equally densely populated place. I’ve known plenty of city-dwellers who move to other cities, including myself. I haven’t known as many that move to the middle of nowhere (though more than zero).

          • onyomi says:

            I mean, I grew up in a city and am currently living in a very small college town. And my current rural lifestyle is much better than it would have been 30 years ago thanks to the internet, Amazon Prime, Netflix… and my rent is ridiculously cheap.

            The past few centuries have been the age of the city. I think as jobs and lifestyles get more portable there will be a gradual spreading back out to take advantage of cheap cost of living.

          • Adam says:

            Well, you’re currently a student and planning to become a professor, right? It stands to reason you’d end up in a college town. I personally moved from LA to Dallas and that tremendously cut my cost of living, but I didn’t have to give up being in a city at all.

          • onyomi says:

            Well, I am a professor… and yes, this town is more interesting than other towns of comparable size because the college brings in some cultural activities. There are definitely still disadvantages to living in a small town, but when I compare it to say, Boston, where it routinely took 30 mins to an hour door to door to take the T, driving 40 mins to a mall doesn’t seem so bad.

          • Urstoff says:

            Not sure why density is a concern; after all, cities are much more efficient at consuming resources than non-urban areas. If people are concerned about resource usage, they should want increasing urbanization.

            And, of course, the price system exists to signal scarcity, which is Julian Simon’s original point. Of course there isn’t enough oil to last indefinitely if the whole world consumed the same amount of resources as the US, but that’s not a remotely plausible scenario.

          • Adam says:

            It depends where. It’s not a concern if the infrastructure grows as fast as the population. It’s a concern where it does not. In the long run, it tends to catch up. New York City was a pretty terrible place 120 years ago. Now it’s not. Hopefully Bombay can say the same at some point in the near future.

            I guess it might become a concern for California if they really do run out of water, but that’s just putting people in the wrong place generally, regardless of how packed or spread out they are.

            It’s really a function of number of people, the ability to deliver sufficient energy and clean water, and the ability to remove the waste they generate. So partly infrastructure, but also partly the locality of the resource. It’s less of an engineering problem delivering water to Philadelphia and New York than it is LA, Las Vegas, and Phoenix.

          • onyomi says:

            I am agnostic about what configuration a bigger population would or should take: existing cities become more dense? existing cities spread out? new cities sprout up in formerly sparsely populated areas? I think any of these are possible, and it would, of course, most likely be some combination of the above.

            It’s just that, of people who tend to feel the world is already overpopulated, I find they are often city-dwelling blue tribe members. Living in a crowded city, it subjectively seems like the world is overcrowded. But when you live in fly-over country, you realize there is plenty of unused space.

            People have different tolerance levels for the nuisances which come along with the benefits of high population density. I predict that, as living away from cities gets more convenient, that benefit/hassle ratio may appear less and less attractive, resulting in a more spread out population. This may also allay the concerns of city dwellers that a bigger future population will necessarily mean that your current city becomes even more crowded and unaffordable.

            It could instead mean that new, interesting cities come into existence where there was once only boring farmland, to say nothing of the new technologies which will probably come into existence faster as a result of a bigger population.

    • John Schilling says:

      The next trillion people to live in this Earth, will each have on average the accumulated intellectual capital of half a trillion people behind them. However many Elon Musks that turns out to be, that’s what they get. And however many Hitlers for that matter.

      Since thought alone produces nothing, those half a trillion people will have all of the Earth’s non-renewable resources to work with. Including, on average, all of the non-renewable resources which are discovered or otherwise enabled by a quarter of a trillion people.

      In this calculation, it makes essentially no difference whether the next trillion lives are spread over a thousand years or a million years.

      But, when it come to the Earth’s renewable resources, and its finite but nigh-eternal resources like real estate, spreading the trillion lives over a million years, gives each potential Elon Musk a thousand times more wealth to back his genius. That’s got to help some.

      Might also help that the Hitlers will have plenty of Lebensraum from the start.

      It is possible for the population to be too small. There are economies of scale that we want to exploit, activities that aren’t viable at very small scales. If you want high-end computers you need a market that can support a high-end chip fab, and that’s a tens-of-gigabucks piece of infrastructure that needs to be embedded in an even larger economy for support.

      But if you’ve got people farming mediocre land because the alternative is to starve, you’ve maybe got too many people. If you’ve got people fighting wars over that mediocre land, you’ve almost certainly got too many people. Check to make sure these people aren’t part of some great and massive enterprise that will be diminished by their absence, and if not, acknowledge that your world is overpopulated.

      I think we’re there. I don’t think the world is hugely overpopulated, and I don’t think this overpopulation will have catastrophic consequences, but I think a few billion people back we passed the point of diminishing returns for marginal population growth and I’d prefer the next few billion be spaced out a bit more.

      • onyomi says:

        “spreading the trillion lives over a million years”

        Assuming we don’t get hit by a comet or something for the next million years. The faster technological development moves, the more likely we’ll be prepared for something like that, or have colonized Mars, or whatever.

        “think a few billion people back we passed the point of diminishing returns for marginal population growth”

        I mean, what makes you think that? This seems far from obvious to me. In fact, it seems obvious to me that the opposite is true, especially in a world where the quality of farm land is less and less important. In the developed world most people don’t need to live on or near good farmland. They just need to live somewhere to which farm products can easily be shipped, which is almost everywhere now. Unless you are using some other definition of “good” land, like scenic? But I think we’ve got plenty of empty, beautiful land too.

        Now with people who are currently subsistence farmers, yes, they probably don’t need to be having more children than they are already having until those children have some better prospects. But open borders would greatly improve those prospects. But for people in the developed world, each marginal person seems to me to be an obvious net positive. And yes, I’d much rather get the next 200 years’ worth of innovations in the next 50 years… that means I get to see them! Or maybe even live 200 years if some of those innovations involve longevity research or improved medical care, which they surely will.

        Can you really claim, even with Hitler, Stalin, Mao, Pol Pot, et al, that the 20th century was a net negative for humanity?

    • Hedonic Treader says:

      Toby Ord gave a good talk in the 2011 Oxford seminar series “Overpopulation or underpopulation? – Seminar Series 2011 – Is the Planet Full?”

      What I would add is that people who believe in overpopulation should oppose all bans on voluntary euthanasia, unless they think pre-existing lives are strictly more valuable than potential lives.

      • Adam says:

        I don’t think I’ve ever actually known a person under the age of 50 that opposed voluntary euthanasia. Clearly, they exist since this seems to still be the majority policy position, but who the heck are these people?

      • suntzuanime says:

        I’m fine with voluntary euthanasia, but I don’t think there’s any coherent utilitarianism that’s remotely palatable that doesn’t consider pre-existing lives more valuable than potential ones.

    • Deiseach says:

      What I would like to know, and what I haven’t seen covered, is this: PP claims that women want, or agree to have, their donated foetal tissue disposed of in this manner. How do they get consent for this? Is this a question routinely asked when setting up appointments? How do you approach this in a sensitive manner (given the uproar that requiring transvaginal ultrasounds was a horrible disgusting invasive degrading procedure)? Is it merely in the ‘fine print’, so to speak, and it’s assumed that unless you specify otherwise, the clinic can dispose of “medical waste” in whatever manner it sees as best?

      I’m not going to get stuck into another fight over abortion because I’m sufficiently cynical about Planned Parenthood (I’ve had my Tumblr dash crowded with Elizabeth Warren bloviating about how this is all part of an attack on women to take away their health provision) that they can mobilise enough support to keep themselves cast as the victims of horrible unscrupulous lying deceitful pro-lifers anti-abortion anti-abortion rights anti- reproductive justice oppressors of women.

      • onyomi says:

        I have no idea how it works in this particular box, but it could be as simple as “check this box” or even, “DON’T check this box” (the latter is far more effective, it has been found with organ donation at the DMV).

        • Deiseach says:

          If it’s “DON’T check this box”, that rather puts the kibosh on their defence that “But women want us to donate their foetal tissue for research!”

          Then again, I’m not inclined to believe PP if they said grass was green. I’m fairly sure the guy with the undercover videos has an axe to grind and did some selective editing, but he didn’t arm-twist the doctor into saying things like “Yeah, we can arrange to crush the foetus different ways to get you the organs you want without damage”.

    • jaimeastorga2000 says:

      The more human beings we have (and especially human being with access to first-world level education–which is getting much easier with the internet, though I am also strongly in favor of free immigration), the more the economy grows and the more likely are revolutionary innovations.

      I think this is an incredibly inefficient way to increase the rate of technological advancement. To quote Eliezer Yudkowsky’s “Inexploitability”, “the number of geniuses and the speed of progress in a civilization does not seem to scale anything like the total population size – maybe for the same reason that small startups can be as creative on average as entire large companies (whatever that reason is).”

      There is tons of usable space, especially now that we don’t need to be right next to a river to get our crops, etc., only, in many cases, there are restrictions on freedom of movement, freedom to build, etc. etc. If the world population were to double that wouldn’t mean twice as many people living in New York and San Francisco. It would mean five new Chicagos and Atlantas, most likely, and three more Shanghais. Would that be a bad thing?

      Unless and until someone figures out why people live in already existing cities and how to make them stop, I think it is perfectly reasonable to treat overpopulated cities as essentially equivalent to an overpopulated country or an overpopulated planet (at least some of the reasons are that land has to be near other people in order to be useful, that cities are where the jobs are, and that it is much cheaper to provide utilities and distribute goods to tightly-packed clusters of people).

      • onyomi says:

        It’s definitely true that just sheer numbers isn’t enough to ensure good technological progress. Obviously there are times and places–Renaissance Italy, Classical Athens–when a relatively small number of people came up with innovations far in excess of what we’d expect based on just population numbers. And of course there are places–often war-torn–which seem to perennially underperform relative to their population numbers. The major difference, of course, is that something is different about the environment: a better legal regime, a culture of encouraging innovation, better access to trade and communication routes, etc. etc.

        And this is why I said it’s especially important that 1st world people have as many children as they want and not prevent other people from coming to the 1st world, at least not for the purpose of work. Because the 1st world right now is a much more conducive environment for flourishing than the 3rd world or most places throughout most of history. It’s not ideal–in terms of the legal regime the 19th c. US was probably more conducive to innovation than the 21st c. US–but it’s still way better than living in rural India or Africa.

        If the goal were just “speed up innovation,” then yes, increasing sheer world population numbers would not be a very efficient way to do it. But if we add to that: increase population of *developed world* and make it easier for the growing populations of non-developed countries to come here, then suddenly it starts to look a lot more effective.

        Even then, I’d agree that, if the goal were just “increase speed of innovation,” then improving governments around the world would get the job done faster than increased population numbers. But everybody’s already in favor of better government around the world, even if they disagree about what constitutes “better.”

        But most people in the developed world right now don’t even seem to view increased population in the developed world as a good thing *at all.* If anything, there is a widespread perception that the world is *already* overpopulated and that falling birth rates are a good thing. I think this is a very wrong and harmful view.

        And, of course, having more children is already within the power of most people living in the first world today, as is voting for people who want to ease immigration restrictions. Improving the governments of 3rd world countries, by contrast, helpful as it could theoretically be, is not something most of us can do much about.

        • houseboatonstyx says:

          If the goal were just “speed up innovation,” then yes, increasing sheer world population numbers would not be a very efficient way to do it. But if we add to that: increase population of *developed world* and make it easier for the growing populations of non-developed countries to come here, then suddenly it starts to look a lot more effective.

          If you mean “increasing sheer world population numbers BY increasing the population of developed world”, that sounds more acceptable than your earlier statements. Could you also unpack the time factor that you hinted at in saying “subsistence farmers” should not increase their birth rate till better provision can be made for the children? Allow another two or three decades for that to improve?

          Back to “the population of developed world”, could you make that “the educated, productive classes of the developed world”? And add something like “as soon as provision can be made for the adults already in innovative work, to get their children cared for without handicapping their own careers”?

    • Anthony says:

      Derailing the topic a bit: I’m moderately pro-choice – I don’t think early abortion should be illegal at all, but I think there should be more limits on abortion than the Supreme Court allows, and this is not just because I believe that it’s difficult to enforce laws against early abortion. But the more I read pro-choice arguments, the less pro-choice I become. The pro-choice people really, really don’t know how to persuade people in the middle; and reading the arguments over the videos about selling organs, it’s clear that for some pro-choicers, abortion is a sacred value, which really offends me.

      • onyomi says:

        I feel very ambivalent, because I’m very libertarian and very much of the school of thought which says that people should be free to do what they want with their own bodies and that making things illegal which people are going to do anyway (drugs) only makes them dirty and dangerous. Yet, I can’t get behind the view that fetuses are just a part of the mother’s body or are just a bunch of cells or something.

        At a certain point, and I’m not sure when exactly, the fetus pretty clearly ceases to be just a possible human being and becomes in fact, a developing human being. It becomes not a part of the mother’s body, but a small, distinct person inside of the mother’s body. At that point, though I’m not sure when it is, I think abortion definitely becomes murder and not just the exercise of a right of control over one’s body.

        I have heard such supposedly science-based guidelines as 20 weeks–that is, after 20 weeks it is, at least theoretically with intense neonatal care, a viable person. Personally, I think society is going to have to find some seemingly arbitrary dividing line like that and just settle on it. Because very late-term abortions, other than in cases of life threatening health concerns, seen clearly wrong, but it also seems unreasonable to say you can’t give pills to a rape victim to prevent implantation, etc. Personally, I think if you make the decision early on, I may think it is often not a good decision, but it is your decision. I think if you’ve waited 20 weeks you can wait 15 more and give the baby up for adoption.

        And yes, I think the hard line pro-choicers are losing the argument as it becomes clear that they are obfuscating in order to prevent peoples’ natural moral horror at the reality of some of what goes on. And for that reason, sensational as they may be, I’m glad that the pro-life organization is putting out these videos because it shows the inhumanity of the more extreme pro-choice position.

        • Adam says:

          It’s definitely going to get weird when the technology exists to keep a fetus alive and incubate it outside the womb at 20 weeks, or perhaps even earlier, because there seemingly isn’t much further argument for killing it if it can be removed intact. You get bodily integrity and life preservation all for the price of one. It seems like there should be widespread support for this, but I can’t help but feel there won’t be.

          • Deiseach says:

            because it shows the inhumanity of the more extreme pro-choice position

            You will excuse my hollow laughter at this, because I’m still seeing the “it’s only a clump of cells” line being trotted out in response to these videos.

            Once again, on Tumblr, an approving reblog of a selection of Twitter tweets from a clearly very upset pro-choice person in relation to the PP controversy, and one of the tweets is “They [I’m going to presume pro-lifers in general are meant] care more about a clump of cells than the woman and family”.

            Well, let’s pull some figures from the CDC on 2011 abortion statistics in the USA:

            In 2011, most (64.5%) abortions were performed by ≤8 weeks’ gestation, and nearly all (91.4%) were performed by ≤13 weeks’ gestation. Few abortions (7.3%) were performed between 14–20 weeks’ gestation or at ≥21 weeks’ gestation (1.4%). From 2002 to 2011, the percentage of all abortions performed at ≤8 weeks’ gestation increased 6%.

            In 2011, among reporting areas that included medical (nonsurgical) abortion on their reporting form, a total of 71.0% of abortions were performed by curettage at ≤13 weeks’ gestation, 19.1% were performed by early medical abortion (a nonsurgical abortion at ≤8 weeks’ gestation), and 8.6% were performed by curettage at >13 weeks’ gestation; all other methods were uncommon. Among abortions performed at ≤8 weeks’ gestation that were eligible for early medical abortion on the basis of gestational age, 28.5% were completed by this method. The percentage of abortions reported as early medical abortions increased 3% from 2010 to 2011.

            I’m going to assume the abortions referred to in the controversial videos are the ones done by curettage (because of the doctor discussing how they can accommodate the needs of tissue purchasers, full transcript here):

            PP: Exactly. So then you’re just kind of cognizant of where you put your graspers, you try to intentionally go above and below the thorax, so that, you know, we’ve been very good at getting heart, lung, liver, because we know that, so I’m not gonna crush that part, I’m going to basically crush below, I’m gonna crush above, and I’m gonna see if I can get it all intact.

            So: 3.5 week foetus – probably fits the “clump of cells” description, may not be as developed as needed for research purposes, anyone more informed tell me?

            Going for the 71% of abortions performed before or at 13 weeks’ gestation, we have 7.5 week foetus, still on the “doesn’t look quite human” side of the line, but a bit more than “just a clump of cells”; 10 week foetus – probably usable for tissue donation purposes? – and 12 week foetus which definitely hits the “looks like baby!” target, and is still well within the permissible limits of abortion legislation.

            If we push abortion back to the earliest permissible date, I think we’ll still see protests about how this is a war on women and it’s privileging a clump of cells over real women and their real families. I don’t see any change in attitudes; people for whom abortion is permissible and even an important part of their values, coupled with the fear of pregnancy as this absolute life-destroying disaster and horror of their body being taken over by a parasite, will react with extreme psychological distress to any talk of restricting abortion and will use any argument to attack the opponents.

            As I’ve said, I’ve seen Elizabeth Warren etc. on my Tumblr dash and nobody addressing the PP issue; it’s been very clearly positioned as “this is all a stunt by cis white het Christian males to control women and force them to be pregnant by rapists and incest-abusers and take all their hard-won freedoms as equal individuals with the right of choice away from them so as to regain the patriarchal power and control in society they see slipping away”.

            PP, for instance, does not do any mammograms, but it’s touted as a free health service for men, women and families that only incidentally does contraception and abortion services as part of a broader range.

            I’ve seen someone proudly boast online “Even if Planned Parenthood only performed abortions I would still support them”. With an attitude like that, I don’t see any changes no matter what future technology or viability outside the womb at earlier stages may bring.

        • blacktrance says:

          You call it showing inhumanity, I call it triggering people’s disgust reflexes and exploiting their irrationality.

    • Troy says:

      I’m getting quite upset looking at Facebook right now, not because I’m strongly pro-life, but because I’m seeing a lot of glib, callous, and illogical arguments defending Planned Parenthood,

      While I’ve seen some dumb progressive arguments on Facebook in this vein, I’ve actually been struck by how many people in my newsfeed are opposing PP. I don’t think I’ve ever seen a “conservative” issue dominate my newsfeed in the same way before.

  45. John Schilling says:

    More fun with “hydrologically-induced smartness”:

    Variance of GDP with state access to waterways is now corrected to 0.446, a few typos in the table of which states accessed which rivers. That’s the strongest correlation I can find in the raw dataset. And the slope is 1.007, almost exactly one dollar per kilometer. Interesting coincidence, but kind of neat.

    GDP vs “smartness”, adjusted NAEP scores, holding at at 0.365, important, but not as important as the waterways.

    Smartness vs. waterway access, variance is 0.119, slope is 1.54 extra points on the NAEP (roughly a third of an IQ point?) per thousand kilometers of waterway access.

    And here’s the fun bit: graphing it up, the relation looks weak but solid for the reasonably populous states with any navigable waterways at all. Indeed, if we look only at river/coastal states, the slope increases to 2.4 NAEP points, almost half an IQ point, per thousand kilometers, and the variance increases to a respectable 0.218.

    But every single completely landlocked state without even a navigable river to call their own, comes in above the trendline – by an average of a full standard deviation. All but two (NM and OK) are above the national average, though only half a standard deviation.

    With the exception of Colorado and Wyoming, these states are of below-average wealth, and they make up only 8.7% of the nation’s population. Barring vast mineral wealth, landlocked regions are poor and desolate and almost nobody wants to live there. But the few who do, at least in the United States, are noticeably smart
    and well-educated.

    Throw in a mediocre bit of coastline or river access, and your test scores drop almost twelve percent. Still trying to figure that one out.

    • Steve Sailer says:

      The potential for canal-building was an obsession of America’s Founding Fathers. For example,

      “Few ventures were dearer to George Washington than his plan to make the Potomac River navigable as far as the Ohio River Valley. In the uncertain period after the Revolutionary War, Washington believed that better transportation and trade would draw lands west of the Allegheny Mountains into the United States and “…bind those people to us by a chain which never can be broken.””

      • Adam says:

        Not only that, but there’s a decent argument that the construction of the Erie Canal was the most important development in New York City becoming such a global power as a city, enabling it to capture all the bounty of the new American midwest that would have had to ship through New Orleans if not for a waterway connecting the Hudson to the Great Lakes. It doesn’t matter as much now, but back in the days when it was ship or pack animal, it was pretty damn important to win that race.

      • John Schilling says:

        No argument with that. There’s still the question of whether the IQ-vs-waterways thing is a leading or a lagging indicator. Do the rivers and harbors generate wealth which attracts smart people, or do the smart people see the rivers and natural harbors and say, “That’s where I need to be”?

        And then there’s the cluster of smart, well-educated people who run off to e.g. Arizona. Any ideas?

        • Adam says:

          I don’t think you’re asking me, but if I had to hazard a guess anyway, cities close to navigable waterways were probably tremendously attractive to smart people 100 years ago, and even moreso 200 years ago. That might still have an enduring effect. I doubt it matters much at this point and Arizona and Colorado are benefiting from being very cheap places to live that are attracting all of California’s castoffs who are still smart but also smart enough to realize they can earn ten grand a year less moving away from the coast while paying thirty less on their mortgage.

          At least, that was part of my own reasoning moving from California to Texas. Sure, I could go to the bay and cluster with a bunch of other smart people, but I honestly don’t care that much about revolutionizing the world by making it possible for people to get a cab five minutes faster on average. I do care about having a decent amount of land and a hundred feet of space separating the outer wall of my house from my nearest neighbor.

          • John Schilling says:

            I was asking anyone with a good idea, and you have one. Technological improvements render navigable waterways less important for future economic development; this makes landlocked sites relatively undervalued, and smart people are first to notice this and take advantage of it.

            I think that would drive a flattening of the curve, via regression to the mean, rather than the noticeable upturn I am seeing at the far left of the graph. I note that your move from CA to TX took you from the top right of all three smartness vs. wealth vs. waterways graphs to a state just above the midpoint, not all the way to the left on any axis.

            But if this isn’t the whole story, it could well be half of it.

      • It was an obsession of the smart people in Olde Englande as well…they were the start-ups of the day.

    • Who wouldn't want to be Anonymous says:

      Not all waterways are created equal.

      In the US, because the federal government has virtually unlimited authority over navigable waters, what is or isn’t navigable is/was a surprisingly political question. We will set aside, as Steve mentioned, Washington’s delirious visions of making the Potomac navigable, and talk about the Susquehanna River (which coincidentally also empties into the Chesapeake bay).

      The Susquehanna is currently the 16th largest river in the US by volume and (IIRC) almost 500 miles in length along its main branch (with at least one other large branch that I know of), passing through three States. I think it was second largest Colonial river, and maybe the 10th-ish largest around the canal building bonanza. Its size and location made it an important geographical feature. If there was any risk of the British being able to navigate it, West Point would have been built on its banks instead of the Hudson. It neatly bisected the colonies and, as an added bonus, was a formidable barrier to land transportation. Actually, scratch that remark about West Point. The British didn’t need to give the river any help. It was already more inhospitable than any warship could hope to be.

      It is a rocky nightmare. It is so shallow and rocky that it is only moderately passable during floods and then only in the downstream direction. It could be claimed with no fear of contradiction that you would be hard pressed to get so much as a canoe down it in low water. Passing upstream is an impossibility at any time. Even getting just a ferry across its waters requires the construction of a dam to raise the water level. Historically, the preferred method of trade along the river involved building a log raft and running it downstream with whatever goods you wanted to sell during flood season, hoping you don’t smash up on somebody’s dam or in one of the rapids, dismantling the raft to sell the timbers along with your goods, and walking home.

      Of course there were attempts to improve the situation while the US was having a love affair with canal building. The coal and timber in PA were valuable commodities, after all. But all the attempts to improve navigation were ruinously expensive. Having to rebuild everything every year or so tends to have that effect. They all failed on their own accord, or immediately after someone had their first fevered dream about building a railroad. In fact, I am under the impression that is was considered profitable to haul cargo upstream and then overland (and later by canal or rail) into the waterways of New York. Someone told me once that more Penn coal was shipped through NYC than anywhere downstream. You can take that anecdote with a whole block of salt, but it is undeniable that a few cities in NY owe their existence to their location on the antistreamwise trade route.

      The Susquehanna is so useless that building a dam less than ten miles from its mouth on the Chesapeake completely sealing the river from the bay (because there is no bypass canal or w/e) resulted in absolutely zero loss of utility for anyone that lives on it. They didn’t even bother putting in a fish elevator until a couple years ago.

      This might be because the only city worth mentioning on the banks of the Susquehanna is Harrisburg. Aside from being the capital of PA it is an unremarkable armpit of nowhere. It barely cracks the top ten largest cities in the state. (I hope nobody lives there….)

      I… don’t remember were I was going with this.

      Oh yeah! Being “navigable” does not mean much. We’ll consider a foil to the Susquehanna. One of those waterways of NY, like the Genesee river.

      You can be forgiven for thinking “the what river?” It is barely 150 miles long; has a discharge about an eighth of the Susquehanna; never leaves NY; and, most importantly, has a very inconvenient impediment to navigation: a series of falls a few miles from its mouth. A couple hundred feet of elevation change in about two miles does not for a very navigable river make.

      Upstream from the falls, however, the river meanders lazily through the region’s famously picturesque and arable hills. I think they make wine now, but historically it was grain. This grain was easily shipped downstream to the falls, were a boom town was powered by the super-ultra-abundance of power the falls provided to a preindustrial city. The grain needed to be unloaded in order to mill it, anyway, so rolling it downhill a couple miles to the port was certainly no hassle. Rochester was The Flour City. Having the Erie canal (which I am pretty sure is also not considered navigable) pass through the city perked things up even more, and the falls provided a super-abundance of power when it transitioned to an industrial city. Xerox, Kodak, and Bausch&Lomb, for example, were founded there. They’ve all moved their headquarters or gone bankrupt since then, though, and I understand the city is kind of a shithole now. That hasn’t hurt the brewery, though! (I hope nobody lives there….)

      Harrisburg has Hershey! Because it is the capital, Harrisburg also has several state and federal instillation there (including the Army War College, which I didn’t know until recently). In fact, the state and federal governments are the two largest employers in Harrisburg, which together account for almost as many jobs as the population of the city-proper. The largest private employer is Giant Food Stores. Being close to Philly is certainly also a plus. Oh yeah, and Harrisburg has recently filed for bankruptcy, which doesn’t seem very auspicious.

      Coincidentally, Rochester is the home of Wegmans, also a grocery chain, which is its second largest private employer… after the University. And the second university is the ninth largest employer.

      Despite being on non-navigable water ways, despite not being near another major city, despite not being the state capital, despite all the major industrial employers moving their headquarters or going bankrupt… Rochester has twice the population in its metro area (and four times in the city itself), apparently has an active science sector, and… isn’t bankrupt.

      And Harrisburg is the only city worth talking about on the Susquehanna, which is a major navigable waterway. So…. yeah. Don’t rely on navigability to tell you anything about usefulness.

      Was I rambling? I think I was rambling… Well, at least I had fun writing it.

      • John Schilling says:

        As it turns out, the NOAA dataset I am using does count Rochester as being on a navigable waterway, and makes no mention of Harrisburg or the Susquehanna river. So whoever is behind that, seems to have a fair understanding of actually useful navigability.

  46. FeepingCreature says:

    Why is it that most people who publically criticize LessWrong also dislike HPMOR as a work of fiction?

    • rsaarelm says:

      Maybe because it has a lot of actual shortcomings as fiction if you want to view it as something more ambitious than some fun entertainment. It’s pretty hit-and-miss for people, and quite likely to be a miss if you don’t buy into the fandom at all. It also has the fans with on average pretty low literary standards who keeps going on about it belonging to the canon of best fiction ever.

    • Deiseach says:

      I’m not a critic of LessWrong, but the excerpts of HPMOR I have read I dislike intensely, and this is down to Harry James Potter-Evans-Verres being so eminently slappable (starting with his triple-barrelled name).

      As a primer for rationality, luring people in with your hero is not working if a reader’s dearest wish after five minutes’ exposure to the little smartarse is to sign up for the Death Eaters so they can have the glory of dumping him at Lord Voldemort’s feet (with all his bits still attached optional). It really doesn’t help to be told “But three-quarters of the way through it gets better!” and “He’s meant to make you want to punch him in the face, he gets a sharp lesson in humility and improves afterwards!” because you are not going to keep slogging through three-quarters to get to the good bits if every second paragraph you are going “My God, I want him to be eaten by a crocodile. Slowly. Feet-first, so we get the maximum slow, agonising, screaming death”.

      That’s why JK Rowling made him an orphan living in a cupboard under the stairs with the standard Cinderella wicked step-family when she introduced him to us: otherwise, with his superpowers, his Epic Destiny, and his ability to be a brat and get away with it, we’d all hate his guts too much to keep on reading.

      EDIT: I wonder if, for those who do dislike LessWrong/Eliezer Yudkowsky, there is the inclination to see HJPEV as an authorial self-insert of EY and so dislike HPMOR for those reasons, the same way the blogger about “Ender’s Game” read Ender as Scott Card’s Christ-figure and heartily disliked the book for its Christianity (whether that was the author’s intent or not) and so decried it as geek wish-fulfilment revenge porn.

      • Nornagest says:

        I think the argument for Ender’s Game as geek wish-fulfillment porn is far stronger than for Ender of that book as a Christ figure, the quasi-religious Speaker for the Dead angle in the sequel and the last twenty pages aside. Revenge porn, not necessarily, though there’s some revenge there: the main thrust of that book is all about holding up Ender as smarter than everyone else, victimized because he is smarter, and morally purer because he is victimized. A view, incidentally, that should be nauseatingly familiar to anyone that’s ever hung out with nerds in an American high school.

        (The Demosthenes/Locke subplot approaches the same points from a different angle, and ends with some dude taking over the world by being a smart guy with a blog. That should be all I need to say about it.)

    • I’ve never read any of it, and have no intention to. Does that count as disliking it?

      • Deiseach says:

        Personally I’d say that counts as good taste, but we’re in the territory here of those who love the blinkin’ thing, so let’s be careful 🙂

    • moridinamael says:

      So, Eliezer has this theory that (paraphrasing and possibly simplifying) some people are much more sensitive to arrogance than others. Here arrogance generally means any signalling of competence or confidence above one’s socially designated station. I think his insight is worth something here because he’s probably thought about this a lot.

      Eliezer is often accused of this type of arrogance – Eliezer’s alleged overconfidence is a main pillar of Halquist’s criticism. HJPEV is typically accused of exactly the same type of arrogance. Whether or not HJPEV is a literal self-insert, both Eliezer and his character both exhibit this characteristic of not caring whether they’re allowed to be as smart as they are.

      In my own life I’ve observed that there are people for home the aphorism “ask for forgiveness, not permission” is a guiding tenet of behavior, and then there are some people who absolutely hate that mentality.

      I further surmise/suspect/project-my-biases-onto-reality that people for whom EA makes sense are more other-oriented and thus less likely to be don’t-give-a-fuck-what-you-think iconoclasts like Eliezer.

      • Emile says:

        In my own life I’ve observed that there are people for home the aphorism “ask for forgiveness, not permission” is a guiding tenet of behavior, and then there are some people who absolutely hate that mentality.

        That’s a very interesting observation – I myself tend to live by that aphorism, and have indeed met people who dislike it; and I am also perfectly fine with Eliezer’s kind of “arrogance”, I wonder how much of a correlation there is…

        • Deiseach says:

          I’m more inclined to the attitude of “Ask for permission first, and if it is unreasonably refused, then go ahead and do the thing” – as long as what you want to do is reasonable, e.g. it’s not “Parents, will you give me the money in my college fund so I can play poker in Vegas because I have an absolute winning system? No? But I’ll triple the money! Fine, I’ll just take it anyway!”

          What annoys me about the style of “Ask for forgiveness, not permission” is the attitude of “Oops, naughty me! Sah-ry!” when caught out, when obviously the person is not sorry they did it, doesn’t care that you’re upset, and has every intention of doing that or something like it again in the future, combined with the expectation of automatic forgiveness and then they get upset and indignant that you didn’t forgive them, so that’s because it’s you being mean, not them having done something wrong.

          Now, if you want to do dumb things that affect you and you alone, go right ahead. But when it comes to your actions having an effect on other people, ask first. At least, that’s my view 🙂

          • FeepingCreature says:

            Well, if you’re looking at a job, asking for permission first is iffy because it shifts the blame to your boss if your idea turns out stupid. It’s not that problematic if a random employee tries to do something, but if somebody in management actually signs off on it, they’re on the hook if it fails – more so than you-the-employee if you fail by yourself! So it’s not a zero-sum; asking for forgiveness is actually more defensible in sum than asking for permission.

            (This depends on how much creativity your job asks for, of course. Also: always keep the option open to abort the thing in case your boss actually objects.)

          • Deiseach says:

            Well, in a work context, it’s generally better to get permission because of the principle of Cover Your Ass.

            I’ve seen examples where people have been instructed to do Stupid Thing and precisely because if (when) it goes off the rails they’re going to be left to carry the can, they get the boss to sign off on it or at the very least leave a clear and definite paper trail that This Ain’t My Idea.

            Random employee deciding to do something stupid off their own bat – there’s pros and cons there. If they really screw up, the first question that’s going to be asked is “Who the hell let Joe blow five million on artificial banana substitutes?” and if you’re not adequately managing your staff so things like this don’t happen, then remind me again why you’re a manager?

            On the other hand, the private sector may be different. The majority of my work experience is in the public service sector, and that’s a whole other ball of wax: we regularly get Do This Stupid Thing* directives so we’re very sensitive about making it abundantly clear This Ain’t My Idea 🙂

            *Also dealing with the public is not all beer and skittles. There’s a heart-rending front page cover story this week in the big local newspaper about a family living in public housing that’s too small for their needs and their kids with medical needs are suffering and they can’t even have a birthday party in the house for the little ones, boo-hoo, sob sob, oh the cruel, hard-hearted, red-tape pen-pushing bureaucrats in the public service.

            The omissions and flat-out lies in that story, of course, are not known to the general public. Cases like this, where Father of Three says “We only ran up arrears of rent because the council never notified us about the rent increase” require you to have a very clear paper trail, because that’s a lie. We always make the greatest effort to get in touch with people to pay their rent; we phone, we write, we don’t just let it happen. And people very conveniently change phones and don’t tell us their new number, never answer calls, never answer letters, never turn up for appointments to discuss why they’re falling behind in rent and how we can arrange a payment plan – until it suits them to use local councillors or local media to push their side of the story.

      • Deiseach says:

        Here arrogance generally means any signalling of competence or confidence above one’s socially designated station.

        Lay not that flattering unction to your soul! Sometimes it’s more a case of “A modest man, with much to be modest about” 🙂

        Arrogance is perceived as unjustified when the arrogant person acts like a jerk to others. It’s one thing to be smart, it’s even one thing to be smart and aware of how smart you are and not be bashful about how smart you are. It’s another thing to rub people’s noses in how dumb by comparison to you they are, to belittle them, to make them the butt of your mockery.

        I think saying that “Well, people only call me arrogant because they want to confine me to my assigned social station – they don’t like it when the lower orders get uppity” can be a way of excusing yourself: I’m not a jerk, they’re wannabe oppressors. Unfortunately, the way the world works, even if that’s true, showing off how much smarter than the boss you are and how dumb their pet idea is will probably get you fired, or at the very least treated with disfavour; you have to indulge in some element of boot-licking when dealing with the higher echelon of superiors (until you get into a position that’s unassailable, either because you make so much money for the company they can’t afford to fire you, or you’re one of the higher echelon yourself).

        “That’s really stupid and you’re really stupid for saying that and I’m so much smarter than you I can tell you to your face you’re really stupid and you just have to take it” may not get you a punch in the nose (e.g. Prince Philip gets away with dreadful gaffes because, well, he’s the Queen’s husband so the recipient just has to smile and take it) but it will make people want to punch you in the nose, and not simply because they think you are trying to rise above your allotted station.

        With HJPEV, it’s not that he’s smart or even that he’s a show-off about it, it’s that his attitude is nobody else can possibly be smart at all (unless they’re agreeing with something he’s said). He’s got the only functional brain in the room and boy is he going to make you aware of that.

      • Anon says:

        Your assumption about EAs disagrees strongly with my experience of EAs irl (n=~30). They care what other people think, but are also, basically definitionally, of the opinion that almost everyone is confused/wrong about what they should be doing with their lives. Also they tend to really like HPMOR.

      • Its important to remember that there are two sides to the overconfudence issue. One is as a potentially annoying behaviour in personal interactions, the other is as a potential stumbling block to rationality.

        I don’t care about the first kind so long as the person is actually good…that is, if they are re well calibrated. You meet people like that in the IT world. Y. Thinks he’s one of those hotshot types, but he’s more like a bedroom programmer that never had any code review, and is therefore poorly calibrated.

    • Held in Escrow says:

      I actually see a lot of it come the other way around. Basically, HPMOR is kind of shit for anything but self insertion power tripping. That’s fine from time to time, but as a work of fiction it’s bad in a