Open threads at the Open Thread tab every Sunday and Wednesday

Before You Get Too Excited About That GitHub Study…

Another day, another study purporting to find that Tech Is Sexist. Since it’s showing up here, you probably already guessed how this is going to end. Most of this analysis is not original to me – Hacker News had figured a lot of it out before I even woke up this morning – but I think it’ll at least be helpful to collect all the information in one easily linkable place.

The study is Gender Bias In Open Source: Pull Request Acceptance Of Women Vs. Men. It’s a pretty neat idea: “pull requests” are discrete units of contribution to an open source project which are either accepted or rejected by the community, so just check which ones are submitted by men vs. women and whether one gender gets a higher acceptance rate than the other. This is a little harder than it sounds – people on GitHub use nicks that don’t always give gender cues – but the researchers wrote a program to automatically link contributor emails to Google Plus pages so they could figure out users’ genders.

This alone can’t rule out that one gender is genuinely doing something differently than another, so they had another neat trick: they wrote another program that automatically scored accounts on obvious gender cues: for example, somebody whose nickname was JaneSmith01, or somebody who had a photo of themselves on their profile. By comparing obviously gendered participants with non-obviously gendered participants whom the researchers had nevertheless been able to find the gender of, they should be able to tell whether there’s gender bias in request acceptances.

Because GitHub is big and their study is automated, they manage to get a really nice sample size – about 2.5 million pull requests by men and 150,000 by women.

They find that women get more (!) requests accepted than men for all of the top ten programming languages. They check some possible confounders – whether women make smaller changes (easier to get accepted) or whether their changes are more likely to serve an immediate project need (again, easier to get accepted) and in fact find the opposite – women’s changes are larger and less likely to serve project needs. That makes their better performance extra impressive.

So the big question is whether this changes based on obviousness of gender. The paper doesn’t give a lot of the analyses I want to see, and doesn’t make its data public, so we’ll have to go with the limited information they provide. They do not provide an analysis of the population as a whole (!) but they do give us a subgroup analysis by “insider status”, ie whether the person has contributed to that project before.

Among insiders, women do the same as men when gender is hidden, but better than men when gender is revealed. In other words, if you know somebody’s a woman, you’re more likely to approve her request than you would be on the merits alone. We can’t quantify exactly how much this is, because the paper doesn’t provide numbers, just graphs. Eyeballing the graph, it looks like being a woman gives you about a 1% advantage. I don’t see any discussion of this result, even though it’s half the study, and as far as I can tell the more statistically significant half.

Among outsiders, women do the same as/better than men when gender is hidden, and the same as/worse than men when gender is revealed. I can’t be more specific than this because the study doesn’t give numbers and I’m trying to eyeball confidence intervals on graphs. The study itself say that women do worse than men when gender is revealed, so since the researchers presumably have access to their real numbers data, that might mean the confidence intervals don’t overlap. From eyeballing the graph, it looks like the difference is 1% – ie, men get their requests approved 64% of the time, and women 63% of the time. Once again, it’s hard to tell by graph-eyeballing whether these two numbers are within each other’s confidence intervals.

The paper concludes that “for insiders…we see little evidence of bias…for outsiders, we see evidence of gender bias: women’s acceptance rates are 71.8% when they use gender neutral profiles, but drop to 62.5% when their gender is identifiable. There is a similar drop for men, but the effect is not as strong.”

In other words, they conclude there is gender bias among outsiders because obvious-women do worse than gender-anonymized-women. They admit that obvious-men also do worse than gender-anonymized men, but they ignore this effect because it’s smaller. They do not report doing a test of statistical significance on whether it is really smaller or not.

So:

1. Among insiders, women get more requests accepted than men.

2. Among insiders, people are biased towards women, that is, revealing genders gives women an advantage over men above and beyond the case where genders are hidden.

3. Among outsiders, women still get more requests accepted than men.

4. Among outsiders, revealing genders appears to show a bias against women. It’s not clear if this is statistically significant.

5. When all genders are revealed among outsiders, men appear to have their requests accepted at a rate of 64%, and women of 63%. The study does not provide enough information to determine whether this is statistically significant. Eyeballing it it looks like it might be, just barely.

6. The study describes its main finding as being that women have fewer requests approved when their gender is known. It hides on page 16 that men also have fewer requests approved when their gender is known. It describes the effect for women as larger, but does not report the size of the male effects, nor whether the difference is statistically significant. Eyeballing it, it looks about 2/3 the size of the female effect, and maybe?

7. The study has no hypothesis for why both sexes have fewer requests approved when their gender is known, without which it seems kind of hard to speculate about the significance of the phenomenon for one gender in particular. For example, suppose that the reason revealing gender decreases acceptance rates is because corporate contributors tend to use their (gendered) real names and non-corporate contributors tend to use handles like 133T_HAXX0R. And suppose that the best people of all genders go to work at corporations, but a bigger percent of men go there than women. Then being non-gendered would be a higher sign of quality in a man than in a woman. This is obviously a silly just-so story, but my point is that without knowing why all genders show a decline after unblinding, it’s premature to speculate about why their declines are of different magnitudes – and it doesn’t take much to get so small a difference.

8. There’s no study-wide analysis, and no description of how many different subgroup analyses the study tried before settling on Insiders vs. Outsiders (nor how many different definitions of Insider vs. Outsider they tried). Remember, for every subgroup you try, you need to do a Bonferroni correction. This study does not do any Bonferroni corrections; given its already ambiguous confidence intervals, a proper correction would almost certainly destroy the finding.

9. We still have that result from before that women’s changes are larger and less likely to serve immediate needs, both of which make them less likely to be accepted. No attempt was made to control for this.

“Science” “journalism”, care to give a completely proportionate and reasonable response to this study?

Here’s Business Insider: Sexism Is Rampant Among Programmers On GitHub, Research Finds. “A new research report shows just how ridiculously tough it can be to be a woman programmer, especially in the very male-dominated world of open-source software….it also shows that women face a giant hurdle of “gender bias” when others assess their work. This research also helps explain the bigger problem: why so many women who do enter tech don’t stick around in it, and often move on to other industries within 10 years. Why bang your head against the wall for longer than a decade?” [EDIT: the title has since been changed]

Here’s Tech Times: Women Code Better Than Men But Only If They Hide Their Gender: “Interestingly enough, among users who were not well known in the coding community, coding suggestions from those whose profiles clearly stated that the users were women had a far lower acceptance rate than suggestions from those who did not make their gender known. What this means is that there is a bias against women in the coding world.” (Note the proportionate and reasonable use of the term “far lower acceptance rate” to refer to a female vs. male acceptance rate of, in the worst case, 63% vs. 64%.)

Here’s Vice.com: Women Are Better At Coding Than Men: “If feminism has taught us anything, it’s that almost all men are sexist. As this GitHub data shows, whether or not bros think that they view women as equals, women’s work is not being judged impartially. On the web, a vile male hive mind is running an assault mission against women in tech.”

This is normally the part at which I would question how a study got through peer review, but luckily this time there is a very simple answer: it didn’t. If you read the study, you may notice the giant red “NOT PEER-REVIEWED” sign on the top of every page. The paper was uploaded to a pre-peer-review site asking for comments. The authors appear to be undergraduate students.

I don’t blame the authors for doing a neat study and uploading it to a website. I do blame the entire world media up to and including the BBC for swallowing it uncritically. Note that two of the three news sources above failed to report that it is not peer-reviewed.

Oh, one more thing. A commenter on the paper’s pre-print asked for a breakdown by approver gender, and the authors mentioned that “Our analysis (not in this paper — we’ve cut a lot out to keep it crisp) shows that women are harder on other women than they are on men. Men are harder on other men than they are on women.”

Depending on what this means – since it was cut out of the paper to “keep it crisp”, we can’t be sure – it sounds like the effect is mainly from women rejecting other women’s contributions, and men being pretty accepting of them. Given the way the media predictably spun this paper, it is hard for me to conceive of a level of crispness which justifies not providing this information.

So, let’s review. A non-peer-reviewed paper shows that women get more requests accepted than men. In one subgroup, unblinding gender gives women a bigger advantage; in another subgroup, unblinding gender gives men a bigger advantage. When gender is unblinded, both men and women do worse; it’s unclear if there are statistically significant differences in this regard. Only one of the study’s subgroups showed lower acceptance for women than men, and the size of the difference was 63% vs. 64%, which may or may not be statistically significant. This may or may not be related to the fact, demonstrated in the study, that women propose bigger and less-immediately-useful changes on average; no attempt was made to control for this. This tiny amount of discrimination against women seems to be mostly from other women, not from men.

The media uses this to conclude that “a vile male hive mind is running an assault mission against women in tech.”

Every time I say I’m nervous about the institutionalized social justice movement, people tell me that I’m crazy, that I’m just sexist and privileged, and that feminism is merely the belief that women are people so any discomfort with it is totally beyond the pale. I would nevertheless like to re-emphasize my concerns at this point.

[EDIT: I don’t have much of a quarrel with the authors, who seem to have done an interesting study and are doing the correct thing by submitting it for peer review. I have a big quarrel with “science” “journalists” for the way they reported it. If any of the authors read this and want my peer review suggestions, I would recommend:

1. Report gender-unblinding results for the entire population before you get into the insiders-vs.-outsiders dichotomy.
2. Give all numbers represented on graphs as actual numbers too.
3. Declare how many different subgroup groupings you tried, and do appropriate Bonferroni corrections.
4. Report the magnitude of the male drop vs. the female drop after gender-unblinding, test if they’re different, and report the test results.
5. Add the part about men being harder on men and vice versa, give numbers, and do significance tests.
6. Try to find an explanation for why both groups’ rates dropped with gender-unblinding. If you can’t, at least say so in the Discussion and propose some possibilities.
7. Fix the way you present “Women’s acceptance rates are 71.8% when they use gender neutral profiles, but drop to 62.5% when their gender is identifiable”, at the very least by adding the comparable numbers about the similar drop for men in the same sentence. Otherwise this will be the heading for every single news article about the study and nobody will acknowledge that the drop for men exists at all. This will happen anyway no matter what you do, but at least it won’t be your fault.
8. If possible, control for your finding that women’s changes are larger and less-needed and see how that affects results. If this sounds complicated, I bet you could find people here who are willing to help you.
9. Please release an anonymized version of the data; it should be okay if you delete all identifiable information.]

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

886 Responses to Before You Get Too Excited About That GitHub Study…

  1. E. Harding says:

    Because GitHub is big and their study is automated, they manage to get a really nice sample size – about 2.5 million pull requests by men and 150,000 by women.

    -This is the impressive part. Only 6% of pull requests are by women. I bet there’s some selection effect going on here, with the result being that women in tech on average make better programmers than men in tech, but would be much worse programmers were there as many women as men in tech.

    Report comment

    • Scott Alexander says:

      Yes, the authors agree that selection effects are probably important.

      Report comment

    • Nadja says:

      Wait, why would women in tech be much worse programmers if there were as many women as men in tech?

      Report comment

      • Pku says:

        Because you’re assuming that only the top women are posting on GitHub, and just about have even skill with the men; more women would be weaker and drop their level.

        Report comment

        • Nadja says:

          Is there any support for that assumption?

          Also, are we saying that A) top women from among the pool of women programmers are posting on github or B) top women from the pool of all women go into programming.

          I interpreted the original comment to mean something closer to B: only top women go into programming. Otherwise how could we think that if there were as many women as men in tech, then women would be much worse programmers?

          Either way, is there any support for either a or b?

          Report comment

          • E. Harding says:

            I was assuming B. Evidence for such a thing should exist, but I have no clue where to find it.

            Report comment

          • Max says:

            Bell curves for gender IQ. Programming is one those jobs where being smarter helps a lot (just like STEM in general). Assuming you need 2std dev + to be good programmer pool of available women is smaller than mens.
            And that before even considering that intelligent women has many more options than programming.

            Report comment

          • Nadja says:

            Thanks for bringing that up, Max. I agree that the gender/IQ effect will likely be there. In fact, if the original commenter’s claim had been that we could expect male programmers to be slightly more intelligent, which might translate into male programmers being somewhat better than female programmers on average, then I wouldn’t have questioned it. But the actual statement was that “women in tech would be much worse programmers if there were as many women as men in tech.” Much worse? I think that’s a strong statement that needs some better evidence than what we have seen here so far.

            First of all, how much of an IQ difference would there be? If the majority of programmers were in fact in the 2+ sigma IQ range, then perhaps we’d have a little bit more of an argument here. But that doesn’t seem to be true. Googling for IQ by profession yields a graph showing the 90th percentile IQ for “computer professions” to be just under 130. (If anyone has better IQ data for programmers, that might help.) Also, I’m not really familiar with the gender/IQ research, but I always thought there will many more men than women in the 2+ sigma range, and even more in the 3+ sigma range, but that’s not where the vast majority of programmers seem to be on the curve. So, given that, how much of an average IQ advantage could we actually expect?

            Then, there’s the question of how much of a job performance advantage would the (slight?) IQ advantage translate into. Non-cognitive skills are often as important as (or even more important than) IQ in determining various types of performance. SAT scores aren’t as good at predicting college success as GPAs are, for example. Isn’t it plausible that women might have other skills that would compensate for their slightly lower IQ and make them (almost) as good at their programming jobs as their more intelligent male counterparts?

            Intuitively, high IQ would be very important for the rare lone wolf type programmers (or elite teams) dealing with difficult problems such as, idk, writing game theory optimal solvers for poker, but that’s not what most programmers do. I used to work on a big software development team once and the favorite part of my job was listening to 2 sigma IQ programmers bitching about the “overly complicated” code of one of our 3 sigma IQ programmers. It was highly amusing. Yeah, I know, in an ideal world the smartest coders would produce code that would dazzle everyone with its stunning brilliance and clarity, but we live in the real world. In the real world the really smart kids often procrastinate, then hack something together at the last moment, creating a bunch of maintenance nightmares, and then move on to doing what they really want to be doing like studying their 13th language or, idk, commenting on SSC.

            So, in summary, I don’t think we have enough evidence to claim that if there were as many women in tech as there are men in tech, the women would be *much* worse programmers. First of all, programmers are not that intelligent on average, so we can’t expect that much of a male IQ advantage, and secondly, the male intelligence advantage might not translate very well into superior job performance. There’s also the somewhat separate question of whether programming currently attracts “top women” which I don’t think is true.

            Report comment

          • Chalid says:

            If you assume B, then either women in the industry should be better paid than men on average, or you there is anti-woman gender bias in pay, or as the Max/Nadja discussion shows, you have to make implausible assumptions about the cognitive requirements of the profession.

            (I wouldn’t assume B.)

            Report comment

          • ryan says:

            Think about it this way:

            If there were more white people in the NBA the overall quality of white players would go down.

            Report comment

          • Nadja says:

            Yes, Ryan, that’s assuming NBA currently attracts top talent, which is probably somewhat of a reasonable assumption. It’s reasonable because of both access and willingness. Access is there because many kids are given the opportunity to play in school, and the really tall athletic kids are going to be encouraged to try it, etc etc. Willingness is there because NBA is high status and very appealing to a great fraction of the population.

            The assumption that tech currently attracts top women is not reasonable. Access is probably there but willingness is not. As I said previously, smart women have what they perceive to be better options.

            Report comment

          • Nathan W says:

            More men study computer programming. Doesn’t have to mean women are worse by nature.

            Report comment

        • Scott Alexander says:

          That seems really speculative – the distribution for women could be the same as men, but the people who participate could be selected from a more elite portion of the distribution.

          Report comment

      • James Picone says:

        I think he means on average.

        Assume women and men have the same distribution of programming ability.

        Say that only women who are 99th-percentile programmers end up working in tech for whatever reason only 6% of commits were from women (the general view that it’s a ‘male field’, systemic discrimination, they have better options, whatever).

        Then for there to be more men in tech, they must draw from lower percentiles in programming ability – can’t be arsed doing the actual maths assuming normal curves – so the programming ability of an average male in the sample will be lower than the programming ability of the average woman in the sample, even though women and men were assumed to have the same distribution of programming ability.

        Report comment

        • Nadja says:

          Thanks for the explanation. So, I’m wondering if there’s any support for the claim that only top women go into tech. In my experience, tech is not seen as a very glamorous option, so if a woman is intelligent/skilled enough to do something else, she will likely choose to do something else. Actually, the same is true for many guys I went to school with. Being a pogrammer just wasn’t something that the most successful people aspired to. They wanted to be investment bankers, management consultants, lawyers, portfolio managers at hedge funds, entrepreneurs, politicians, etc, etc.

          Report comment

          • weareastrangemonkey says:

            Nadja, I think you nail it. They are making very strong assumptions about people’s outside options, that the outside options don’t improve at a faster rate than the option of programming with some arbitrary innate ability measure.

            Even if you grant them this assumption there is still a big problem with their analysis. They are making an appeal to the greater IQ variability hypothesis to get the differing ratios. But in order to get the 10:1 ratio we see in this data we would have to be over 3 standard deviations into the distribution. I am pretty damn certain that the lower bound for tech guys is no where near 3 SDs out from the average.

            That’s not to say that there isn’t need for concern about the selection issue. But I think you are right that it does not automatically follow that the women in tech are just the crem de la crem.

            Report comment

          • Svejk says:

            I would speculate that rather than tech, the top-percentile STEM women gravitate toward academia – particularly the biosciences, and consulting. There are a lot of talented female bioinformaticists and data scientists who do not see programming as their primary vocation, but have made important contributions to scientific programming products. These tend to be in-house efforts requiring significant domain expertise that do not appear on GitHub, or are only migrated to GitHub after reaching a certain level of maturity. Consulting is often equally remunerative to tech for top entry-level candidates, with fewer geographical constraints at the top tier, and is perceived as more welcoming to women with more attractive lifestyle benefits. Women are sought-after consultants in part because even ‘nerdy’ women are assumed to have the soft skills adequate to handle client interaction and personal presentation. The pipeline for women from a biology, chemistry, or statistics BA/PhD to McKinsey is probably significantly less ‘leaky’ than the academic or tech pipeline.
            I would further speculate that women participate in GitHub projects at a lower rate than men partially for the same reasons they tend to post blog comments at a lower rate than men, including a lower general tendency to self-advertise.

            Report comment

          • James Picone says:

            To be clear I’m not saying that’s true. I have very little personal experience with female programmers and haven’t looked into any data on it. More explaining what I saw as the thought process.

            Even if you grant them this assumption there is still a big problem with their analysis. They are making an appeal to the greater IQ variability hypothesis to get the differing ratios. But in order to get the 10:1 ratio we see in this data we would have to be over 3 standard deviations into the distribution. I am pretty damn certain that the lower bound for tech guys is no where near 3 SDs out from the average.

            I wasn’t relying on IQ variability; when I said “assume women and men have the same distribution of programming ability” I did mean it. I am assuming that whatever it is that results in less women in programming funges against less-skilled women. For example, if it’s systemic discrimination, presumably women who see programming as a calling are going to be more likely to tough it out and end up programmers, and they’re also going to be more likely to be good programmers.

            In my experience, tech is not seen as a very glamorous option, so if a woman is intelligent/skilled enough to do something else, she will likely choose to do something else

            In my experience there’s a significant subgroup of people who get into programming because they find it fascinating and fun. Those people also tend to be really good programmers. If I were to try and construct a srs biz explanation, it’d be that female programmers are almost exclusively drawn from that group, and male programmers draws from that group but also people who are in it for the career.

            Report comment

          • Nadja says:

            @wearestrangemonkey @Svejk
            Yes! Thanks for the comments.

            @James
            Oh, yes, I know you weren’t saying it was true. And I like your thoughts on why people get into programming.

            Report comment

      • Anonymous says:

        Bell curves.

        Report comment

    • Deiseach says:

      If we’re going to argue about “sexism in tech”, the fact that there is a ratio of 2.5 million to 150,000 sounds very pertinent; either men make very many more suggestions for fixes than women do, or there are a lot more men in the field than there are women.

      Which then brings us back to “why are there very many more men than women in the field?” and the hair-pulling over sexism can commence.

      I think the “insider/outsider” thing could be as easily explained by insiders knowing “Oh, this is by Sally? I know Sally, she does good work” whereas for an outsider it’s “Who the hell is Jane? Is she any good?”

      Report comment

      • onyomi says:

        I think the “insider/outsider” thing could be as easily explained by insiders knowing “Oh, this is by Sally? I know Sally, she does good work” whereas for an outsider it’s “Who the hell is Jane? Is she any good?”

        This could explain outsider women doing worse: women are less likely to be programmers in the first place, so one might be less willing to accept a woman as a competent programmer until she proves herself.

        But this doesn’t explain why the insider women do better than the insider men. My guess is that insider men feel competitive with other insider men, but have a soft spot for insider women, perhaps, in part, precisely because of their relatively small number, and also, probably, in part, because the male instinct to be protective of women doesn’t cease to function across cyberspace.

        Report comment

        • Anthony says:

          Less so for the Github context. It’s more “Sally|Eric|fqubr007 has submitted good patches; this one is likely to be good, too.”

          Open source programming isn’t pure meritocracy – if you submit a patch that’s not keeping with the general direction the project leader wants to go in, it will be rejected no matter how technically good it is. But people smart enough to program well are generally smart enough to figure out whether their patch fits those requirements which aren’t purely technical.

          Report comment

      • Thomas says:

        “Which then brings us back to “why are there very many more men than women in the field?” and the hair-pulling over sexism can commence.”

        Not saying this is your argument, but this is lazy thinking.
        Because there’s a disproportionate amount of X versus Y gender != sexism.
        Because there’s a disproportionate amount of X versus Y race != racism.

        Along the same lines, the NBA is 70% black, yet black people are 13% of the population.
        Females make up ~90% of nurses, yet 50% of the population.
        Jews comprise 0.2% of the population, yet 22% of Nobel laureates are Jewish.

        Point is: Statistical disparity != discrimination.

        Report comment

        • The Nybbler says:

          It’s not lazy thinking; the Social Justice groups have put a LOT of thought into turning a statistical disparity into proof of sexism. Step one, of course, is to place the burden of proof on those denying sexism. “We’ve got a disparity in outcome. This means sexism unless you’ve got some better idea”. Step two is to systematically eliminate all the other possible explanations on the grounds that either they indicate sexism too, or that using that explanation is itself sexist.

          So, propose that men are simply better at programming than women… you’ll be run out of town on a rail for being a sexist.

          Propose that women are less interested in programming than men… you’ll be told that this is because of sexism.

          Propose that women are less interested in programming because they prefer fields with more human interaction… you’ll be told that this preference is social conditioning and therefore sexism.

          Propose this preference is actually biologically determine… you’re a sexist.

          Propose that women don’t enter the field because they find the men in the field now weird and unattractive… this too is sexism. On the part of the men already in the field.

          Propose that it’s not the duty of the men currently in the field to change to become less weird… you’re a sexist.

          Report comment

    • Squirrel of Doom says:

      Assuming these are public Open Source projects, this is mainly the work of hobbyists and professionals contributing on the side.

      There is extremely good work being done in some of these projects, but on average it is not where the best programmers are. They work paid jobs.

      Report comment

      • orangecat says:

        Although today it’s not uncommon to get paid to write open source code; in many cases the “insiders” may be doing it as part of their day jobs. Which could explain the relatively greater proportion of women among insiders, given the reasonable assumption that women are much less likely than men to be involved in open source projects in their non-work time.

        Report comment

      • Nonnamous says:

        I’ve been in this industry for fifteen years and I suspect the opposite, i.e., the median open source contributor will dance in circles around the median SWE working a paid job, when it comes to programming ability.

        Report comment

        • Jason Miller says:

          You took the words right out of my mixed closed / open source mouth. The massive attention paid to the implementation of open source projects tends to demand a higher code quality than the outcome-driven nature of closed-source. Open-Source also allows for collaboration on architecture, which is bound to lead to better results than one person’s brainchild.

          Report comment

        • Squirrel of Doom says:

          Perhaps. It would be an interesting thing to study.

          For the major infrastructure projects you and I use daily, I’d believe it more than for the average project.

          Still, my point is that this field is mostly not where people have their professional careers, so it’s not that relevant for the question of if women are welcome in there.

          Report comment

        • noge_sako says:

          Thankfully, the CS industry has adamantly refused to try and create standardized tests of knowledge and competence* (unlike the rest of engineering), so anyone can spew whatever they think!

          *http://ncees.org/exams/pe-exam/

          Report comment

        • Marc Whipple says:

          I don’t know about dancing in circles, but IME their routine will almost certainly be noticeably better, yes. (I was employed for a long time by a company that made heavy use of OS materials. The best programmers were usually the biggest OS proponents as well.)

          Report comment

        • Orphan Wilde says:

          Working with ~50 open source libraries in my own job… yep.

          Open Source is, by and large, barely-maintained garbage. Releases are rarely backwards-compatible, releases often (but don’t always) update dependencies which also aren’t backwards-compatible, projects often include horrendous numbers of dependencies for single easily-coded functions, architecture is generally epileptic and inconsistent (too many chefs in the kitchen), basic functionalities are often left untested, any given open source project is likely to be orphaned within a year or two, and the majority of the code, when examined, was either written by an inebriated zebra mashing random keys with its hooves, or are completely and illegibly “clever”. (If a novice programmer can’t immediately identify what your code is doing, you’re a shit coder spreading shit practices, and this is doubly true for open source.) Also, quit spawning threads because I have to account for all of those and then fix the deadlocks they inevitably fall into, and also I’m one of two people I’ve ever met who knows how to multithread safely.

          That said – I use a lot of open source libraries, so I am pretty dependent on open source coders (I have maintained open source in cases I couldn’t work around, I do not consider myself an author of it). The fact that most of them do terrible work is not to detract from the fact that the work they do is often (usually?) far more valuable than most of what I do. A given piece of code I write is likely to be used by one or two companies, as opposed to dozens or hundreds.

          So they do valuable work. It’s just a pity most of them are incredibly bad at it.

          Report comment

    • Darwin says:

      How does ‘but would be much worse’ follow from the argument of a selection bias making current women programmers better? Wouldn’t it be more parsimonious just to assume that the difference disappears?

      Report comment

  2. Anon. says:

    Why do you choose to only state the name of the publication instead of the author and their editor? If you’re looking to effect any change, surely naming names would result in a bit more “oomph”.

    Report comment

    • Scott Alexander says:

      The paper has a bunch of authors, and I’m not out to shame anybody here.

      Report comment

      • Anon. says:

        I meant the articles, not the paper.

        Report comment

        • zensunni couch-potato says:

          My guess is that it’s still basically the same reason; the bad media coverage had a bunch of people responsible for it, and he’s not out to shame anybody here.

          It *might* have more oomph to name and shame the authors and editors, but that oomph might be outweighed by the elevated chance of a hate-mob if the issues spreads to the wrong corridors.

          Report comment

          • Jon says:

            It’s wrong to shame people. Unless they’ve done something really terrible. Scott — any chance we can get you to do one on this NYT article about a dude who got shot for going to the ER during a manic episode?

            Report comment

          • Deiseach says:

            Yeah, if they’re only students, it’s not fair to get them into a mauling. Let the bad “science” journalism do that- there is going to be more than enough claiming and counter-claiming stirred up by that.

            Reporting like this is why I’d advise you all to be very skeptical about anything involving religion reporting; a lot of media are dropping specialist religion-beat reporters and so you tend to get terms like “fundamentalist” thrown around without any real meaning (never mind the perennial “A woman has just been ordained a Roman Catholic priest” cheerleading pieces that pop up every so often. Short answer: no, a woman has not been ordained a Roman Catholic priest, any more than I could run an election on here, get you all to vote for me, and then declare myself President of the United States based on that).

            Report comment

          • Muga Sofer says:

            Yeah, it’s not “science” journalism per se, or that Scott has identified some cluster of Especially Bad Journalists (as Scott seems to be pointing out himself.)

            It’s that all journalism is terribly unreliable. We just only notice it when it deals with people/incidents/topics we’re actually informed about. Murray Gell-Mann amnesia, as it’s occasionally called.

            Report comment

          • Anthony says:

            Not all journalism. Just journalism about subjects which require more intelligence than typical journalists possess to understand. Sports generally doesn’t require a lot of brains to participate in or understand, so most sports journalism is pretty good. (Though figuring out *why* the GM made that trade sometimes escapes journalists, because GMs are generally smart people.)

            Report comment

          • Simon says:

            I would say sports journalism is one of the few places where there’s still some proper reporting skills required and that, rather than it being particularly simple, is why it retains some quality.

            So much else is supplied via PR and political briefing and published almost without any critical process at all. In addition the big breaking stories are often handled by non-science journalists leading to even more uncritical errors.

            Report comment

          • My snarky but possibly true hypothesis is that you can tell what people care about by where they give up irrational discrimination*. There’s less discrimination in the US in entertainment than in banking, and I believe that’s because people care more about whether they’re entertained than whether money gets handled sensibly.

            For the specific case, it’s quite plausible that people care more about sports than they do about science or religion (especially other people’s religions).

            *There’s a common (Marxist?) idea that people discriminate because they can get richer by discriminating. While this happens, I think discrimination is at least as likely to be a luxury that people engage in when they think they can afford it.

            Report comment

          • Adam says:

            Also, for people who are interested in these things, they acquire knowledge of science by studying science under actual scientists and acquire knowledge of religion from whatever form of priesthood that religion offers. Sports fans largely gain knowledge of sports through sports journalism. Sportswriters are the priests.

            Report comment

          • Tom Richards says:

            Mainstream sports journalism, with the possible, partial exception of cricket, is atrocious. Most consumers of it are interested in emotional narrative, not actual analysis, so that’s what they get. I don’t think it’s much different to anything else: there are niche websites doing good work for a hardcore audience that seeks them out, but what you read on ESPN or SkySports or the Telegraph or whatever is almost all unmitigated tripe.

            Report comment

          • Jonas says:

            My snarky but possibly true hypothesis is that you can tell what people care about by where they give up irrational discrimination.

            My highly snarky applications is that SJWs by that logic really don’t care about gender/racial/etc. equality.

            Report comment

  3. John Schilling says:

    By far the most significant result in that paper is the one where an unknown programmer revealing their gender leads to a ~10% relative drop in the acceptance rate. Once that appeared in the data, it should have been the focus of the subsequent analysis, the key takeaway from the abstract, incorporated into the paper’s name, and the only thing “science” journalists could take from it without knowing they were deceiving themselves and their audience. Everything else, being a percent or two of difference here or there, is trivia.

    Which suggests that everyone involved was so focused on the task of “proving” a gender bias they “knew” was there, as to be completely oblivious to the genuinely interesting thing they walked right into.

    Report comment

    • The Smoke says:

      It would be interesting to know if that effect has to do with gender or rather with personal identity: Maybe it doesn’t matter if the gender is perceivable but more if the user name seems to be the real name of the person (which might lead to one being perceived as more interested in status).

      Report comment

      • Anonymous says:

        Speaking from personal experience (purely anecdotally), there are two reasons programmers are on github: passion for the craft and resume building.

        Most people there are interested in both but people that are more interested in the second reason are more likely to link their profile to other social media and use their real name, they are also more likely to make crappy PRs.

        Report comment

        • Error says:

          That’s interesting. I use my legal name on GH for resume-building purposes, but I find that it makes me pay more attention to doing quality work, not less. On introspection, the implicit assumption is that anybody who actually looks at my GH code is likely going to be 1. able to recognize good code, and 2. able to dent my career if they don’t like it.

          (on the other hand, I don’t link my profile to other social media — because I have no other social media and if I did it would be HR’s anti-business. YMMV.)

          Report comment

          • HeelBearCub says:

            @Error:
            I think the correlation would work more like so. People on GitHub/open source to build resume are a) people who are more likely to need to build resume, and b) people who are less likely to be passionate about the particular open source project they are contributing to.

            There is also a special brand of a) that is trying to play a certain kind of resume game that leads to management and then to C-level jobs at small companies. Not sure whether they are large enough in number to be meaningful though.

            Report comment

      • Deiseach says:

        I would definitely be going for the personal identity angle, given that men also had a drop in acceptance rate once their gender was revealed. How else do you explain it? Somebody knows Susie and has an (accurate or not) assessment of her work, they will or won’t accept a fix suggested by Susie. They don’t know Anne, they’re as likely to go for Joe’s fix – or given the huge discrepancy between 2.5 million and 150,000, they might not even get to Anne’s suggestion in the middle of all the suggestions from Bill, Tom, Mike, Fred, etc.

        Report comment

      • Daniel Speyer says:

        I would guess that lack of genderability is a proxy for hacker acculturation.

        We tend to shorten our names to three initials (DMR, RMS) or first initial / last name (bstroustrup). I’m not sure where this tradition came from. Possibly a mix of begrudging every keystroke and hating inelegant collision handling. Things like firstname.lastname have a vague not-what-we-do feel.

        And putting a photo of myself on github just feels *wrong*. That doesn’t belong there. Like taking a dependency on X in a printer driver.

        I suspect that along with these hacker customs, people pick up hacker values (like paranoia in testing and separation of components) which lead to genuinely better code.

        Why this effect would be stronger in women I don’t know.

        Report comment

        • John Schilling says:

          firstinitial / lastname and lastname / firstinitial were the traditional defaults for user names and email addresses in the dark ages where every byte had to count, and often continue in that role to this day. I have one colleague who goes by lastnamefirstinitial (an easily-pronounceable monosyllable) in all but the most formal real-world contexts, from the email address he was assigned as an undergraduate.

          Report comment

    • Anonymous says:

      What did they do about names that may imply one gender but are actually unisex? Lynn, Tracy, and Ash are the first four that come to mind.

      Report comment

    • Michael Watts says:

      There’s a related question which has been the subject of study:

      Some black people in the US have names marked for blackness, like “LaDarius”. As far as I can tell from the discussion, they are “marked for blackness” in that they’re innovative, and it’s overwhelmingly black people do that. (related comedy bit: https://www.youtube.com/watch?v=gODZzSOelss )

      People with such names tend to have remarkably poor life outcomes, relative to the rest of the United States. The question is, are the names causing those poor outcomes, or is the problem that their parents were the kind of people who thought a name like that was a good idea?

      I believe there is some research to the effect that, after controlling for personal quality, there is not a residual effect of overt name blackness on life outcomes (this has pretty strong implications for those other studies with the methodology “we sent out matched pairs of resumes with one white name and one black name, and the black names got significantly fewer callbacks”). Similarly, it’s quite possible that people who go to the effort of making sure you can tell what sex they are differ systematically from people who don’t do that.

      Report comment

    • 4bpp says:

      Wording it like that sounds like a repetition of much of the same mistakes the media reporting committed, though – all they found is that the programmers who revealed their gender in the past had a 10% lower acceptance rate. I would guess that a significantly higher percentage of people who regularly wear jackets with elbow patches are working academics than of those people who don’t, but one (probably) wouldn’t suggest that someone put on elbow patches to increase their chances of getting that postdoc position.

      Report comment

    • Darwin says:

      I’m a bit confused by your logic – I agree that they should have talked more about the interaction effect, but that 10% *is what proves* the gender bias they are talking about.

      Report comment

      • Loquat says:

        How does the drop prove gender bias when it occurs for *both* genders?

        Report comment

        • Darwin says:

          It’s an interaction effect – there’s a drop for both genders when their gender is revealed, but the drop is about 5% for men and about 10% for women. So we can assume that about 5% of that drop is due to factors that affect both genders, and about 5% is due to factors that affect only women when their gender is revealed (the most likely of which, until someone advances a plausible alternative, would be a gender bias among reviewers).

          Report comment

          • Jonas says:

            the most likely of which, until someone advances a plausible alternative, would be a gender bias among reviewers

            Help me out here—why would that be the most likely?

            Report comment

    • Greg says:

      That would be data mining though.

      I think the way social science papers normally work is that you formulate a hypothesis and test the hypothesis.

      The way significance tests are structured assumes this. They ask, “If there was no actual effect and just random data, what are the chances that you’d see something as extreme as this?” If you get less than 5% or ideally less than 1%, then you start getting some support for your hypothesis.

      On the other hand, with millions of pieces of data, there are nearly an infinite number of hypotheses that can be tested. Simply by chance, countless hypotheses will appear statistically significant. Unless you actually hypothesized these things a priori, the tests break.

      If this effect was actually deemed interesting and worth further study, I think you’d use this data to formulate a hypothesis about insider-outsider status. Then, you’d take a completely new set of data and test that.

      Report comment

      • Luke Somers says:

        Problem: where do you get another GitHub?

        Report comment

        • Kyle says:

          You could glean similar data off older/smaller sites like Sourceforge, Google Code, BitBucket, etc. Or you could look at a single large project like Chromium or Linux.

          Problem is this study didn’t even gather their own data. They used data others (GHTorrent) had already collected and then fished around in it for any interesting conclusions.

          Report comment

        • Adam says:

          They had enough observations that they easily could have broken it into training and test sets and done all of the data mining and hypothesis formulation only on the training data.

          Report comment

  4. Matthew Russotto says:

    (Note: I ordinarily post here under a different nym, but since I’ve made this point elsewhere under my real name doing so would “out” that nym)

    One conclusion I can make from those charts, plus the overall point that women have their pull requests accepted more often than men (along with the numbers), is that a greater proportion of women are “insiders” on the projects they contribute to.

    Report comment

  5. piercedmind says:

    It seems that there are large incentives to conduct and publish studies like these that, very charitably, are honest but very unsuccessful attempts at conducting science. While I don’t want to attack any particular author, especially not the undergraduates here, I am baffled that nobody is questioning how it’s possible that eminent scientists publish extremely shoddy research without much negative backlash in academia.

    As it stands, writing an article about “SJW-ish” issues gives you great chances to be picked up by major news, which in turn increases name recognition and thus chances at promotion or grant money, so it’s not surprising that scientists do publish these studies. It’s even more hopeless to expect the press to adopt higher journalistic-scientific standards.

    The only negative incentives are long and polite rebuttals from other scientists, which do not appear to be a strong demotivating factor. Maybe it might be a good idea to actually call out this behaviour the way it is: fraudulent and dishonest. I understand that sceintists despise using that language, and I am uncomfortable suggesting using it, but I am still fairly certain that it’s fitting. There is a reason the authors did not conduct a signifance test on the actual difference, and that’s probably the same reason why they did not include the gender of the approver, and that reason is very likely not just an honest mistake.

    If incentives to publish this kind of research are so high, then maybe the counter incentives are too weak and need to be ramped up.

    Report comment

    • Theo Jones says:

      I don’t think there is much blameworthy that the authors did. The paper is a non-peer reviewed article by some undergrads. The problem here are non-scientists in the popular media that 1) amplified very weak and contingent results beyond their merits, 2) misrepresented the actual details of the study.

      Report comment

      • Anonymous says:

        “I don’t think there is much blameworthy that the authors did”

        They set out to prove a prejudice (programmers are sexist pigs) and when the data did not cooperate they engaged into p-hacking until they found something that proved their prejudice. They are either incompetent and, given how much p-hacking is at the center of attention, living under a rock or they are disonest.

        Report comment

        • Theo Jones says:

          Really, I don’t see that in my read through. I see some sloppyness and misuse of statistical tools in the paper (namely no adjustment for multiple tests ran on the same dataset). But when you consider that the authors are undergrad comp sci majors and seem to have no experimental science background that is to a degree forgivable.

          Remember that this was a draft that wasn’t even submitted for publication yet. It was released for comment on PeerJ Preprints — a service designed to allow researchers to solicit feedback on their drafts. Here is how the PeerJ people describe their service https://peerj.com/preprints/
          A PeerJ PrePrint is a draft of an article, abstract, or poster that has not yet been peer-reviewed for formal publication. …… Establish precedent. Solicit feedback. Publish updates.

          The paper had a large header on each page reading “PEERJ PREPRINT NOT PEER REVIEWED” and on the abstract webpage for it it had a text in bold font and red color reading “NOT PEER-REVIEWED. This is a rapid communication before peer review”. So, this should of been reasonably obvious to any reader.

          And the authors added the following note:

          https://peerj.com/preprints/1733/#annotation-2002
          This report has not yet been peer-reviewed, and thus the findings should be considered preliminary. Before citing this paper, please check for an updated version here: http://people.engr.ncsu.edu/ermurph3/pubs.html

          The press took that draft and treated it like a final publication. Without even bothering to note some of the issues.

          Report comment

        • HeelBearCub says:

          They set out to write a paper as an undergrad. Undergrads have only a faint notion of all of the things the don’t know, and an even fainter notion of all of the things they don’t know they don’t know.

          We should expect that any paper they write should have many flaws before going through peer review. And frankly, the odds that a paper from undergrads make it all the way throuh peer review and into a journal are very small. That paper likely was not even accepted for peer review by any journals, hence it appearing on an “open” peer review site.

          They has the right of it here.

          Report comment

        • Deiseach says:

          No, this is the kind of paper that a good supervisor will take apart (as Scott did) and tell them what was wrong with it and how to improve it.

          It was, unfortunately, the kind of headline media outlets love. It’s not about the “science” in the “science reporting”, it’s about the “get this story read and talked about for hits to sell advertising”. STEM is a trendy subject right now, with the political push everywhere to “get more kids/women/minorities into STEM fields”, “STEM work is going to be the engine of the economy and the jobs of the future”, and “sexism in X” is always going to be a winner.

          Put the two together and you have an irresistible subject for opinion writing, columnists, editorialising, etc. that stirs up enough controversy to get attention for the particular organ writing the piece.

          Report comment

          • Held in Escrow says:

            This seems like it was done in good faith and with a decent knowledge of how to conduct some basic studies. Professors love it when their students have ideas and follow them out like this. Sure, there are some major issues but that’s stuff to teach the students about so they can fix them and do better next time!

            It’s not coming from a terrible misunderstanding of basic principles like the undergrad “McDonalds could give an X raise to their employees by only raising prices by Y” paper that was bandied about a few years back. It’s a solid first attempt by a group of undergrads. IT’s just a shame that science news is governed by the worst set of incentives known to man.

            Report comment

      • The Nybbler says:

        Sorry, no. I thought perhaps they were innocent at first, but after a few readings I noticed that they have a heading “Is a woman’s pull request accepted more often because she appears to be a woman?”

        Then they go on to say “While acceptance rate results so far have been robust to differences between insiders (people who are owners or collaborators of a project) versus outsiders (everyone else), for this analysis, there is a substantial difference between the two, so we treat each separately.”

        They do this without ever presenting the numbers considering the two conditions together. This is basically a magicians trick. “Now watch carefully as we demonstrate sexism by comparing masked and unmasked participants… but that’s too easy, let’s make this a little harder, first we’ll split them into two groups…”

        Report comment

    • Samedi says:

      In my view the problem here is not the topic of the study but the credence given to single studies themselves. A single study is scientifically worthless. The findings in a single study may be invalid for any number of reasons: methodological error, unintentional mistakes, bias, fraud, etc. Only when the findings have be reproduced by different people over time does the likelihood of accuracy increase.

      A single study may be catalytic, in the sense that it inspires others to research the topic, but on its own it counts for little to nothing.

      Report comment

  6. Alex says:

    The zoomed-in scales exaggerate the magnitudes of the effects in play here.

    Report comment

    • Scott Alexander says:

      I have mixed feelings on this. Truncating the y-axis is potentially deceptive, but if the y-axis went all the way from 0 to 100, the difference between 63 and 64 would be invisible and the graph would be useless. At least this way you can see what they’re trying to show.

      Report comment

      • Said Achmiz says:

        But the problem is: the fact that the scale is zoomed-in is not clearly indicated. The axis labels. The y-axis is truncated, and this fact is easily missed, which makes the graph very deceptive.

        There is a known and well-established solution to this: axis breaks (illustrated here). (I would recommend the following: 1. Break the axis; 2. Make the labels at zero, just before the axis break, just after the axis break, and at the top of the graph bigger and bolder, to make them stand out.)

        (An alternative solution is two versions of the graph, one of which is a “zoomed-in” version of the other — but that may be infeasible given space constraints.)

        Effective data graphics aren’t rocket science. They take effort, but they aren’t some esoteric unsolved problem.

        Report comment

      • Alphaceph says:

        > the difference between 63 and 64 would be invisible and the graph would be useless.

        i.e. the study is useless. If the difference is so small that you can’t see it on a graph, that’s a clue that the effect is irrelevant.

        Report comment

        • Said Achmiz says:

          This is incorrect, because there is no reason to expect the axis range of a data graphic to have any kind of straightforward correlation with the scale of practical consequences of any aspect of the data that the data graphic is displaying.

          Report comment

    • Darwin says:

      This is a knee-jerk response that we’ve learned to use to discredit papers, but it’s not really applicable here. With over 2 million data points in the study, small percentage differences are likely to be statistically significant; and in a complex social context like gender bias in the tech industry, they’re also likely to have large effects on final outcomes. Zooming in to make the graphs clearer is reasonable in a case where the differences are ‘small’ relative to the full possible range of the scale, but very significant and important (and large with respect to the actual *used* range of the scale for the phenomenon in general).

      Think about flipping a coin; if you flip it 2 million times and it comes up heads 51% of the time, it is *definitely* not a fair coin. It wouldn’t be misleading or improper in this case to zoom in on your bar graph so readers can see that 1% difference.

      Report comment

      • Patrick says:

        “With over 2 million data points in the study, small percentage differences are likely to be statistically significant; and in a complex social context like gender bias in the tech industry, they’re also likely to have large effects on final outcomes.”

        Why.

        The second part, not the first. I understand how using an astronomically high sample size can allow you to detect statistically significant differences even if those differences are too small to noticeably affect the lived experiences of anyone in the groups.

        I don’t understand how that isn’t a refutation of the second half of the sentence.

        Report comment

        • Darwin says:

          Ok, if your question is why do I think a 4% bias against women could have a big affect on their lives, the answer has to do with long-run outcomes of complex systems; it’s related to the reason why chaotic systems are highly sensitive to initial conditions.

          If 4% more men than women get good mentorship and encouragement in college CompSci programs, and then of the ones that finish that, 4% more men than women get accepted to good Graduate programs, and of those graduates 4% more men get glowing recommendations from their advisors to put in their job applications, and of those applications 4% more from men are offered interviews than women, and in those interview they hire 4% more men than women, and so on for performance reviews and raises and promotions and invitations to speak at tech conferences and etc etc etc… that 4% compounds over a lifetime in ways that end up with huge differences in outcomes which can distort the fabric of an industry. That’s how minute differences lead to big changes in complex dynamic systems – by applying their effect repeatedly, in a cumulative manner, to *many different interactions* within the system, over time.

          Report comment

          • Jason K. says:

            In order for compounding to work like that, it has to be the same 4% getting benefits/discriminated against each time, which seems rather unlikely.

            Report comment

          • Muga Sofer says:

            That depends on whether it’s a 4% chance of experiencing any discrimination, or a 4% worse outcome – which in turn depends on whether something consists of a lot of little interactions or one big one. On job interviews, the effect would be small; but you would expect it to be larger on job *choices*, for example.

            Report comment

          • Jason K. says:

            “That depends on whether it’s a 4% chance of experiencing any discrimination, or a 4% worse outcome ”

            Most barriers (when it comes to employment) are binary, or fairly discreet. You either got something or you didn’t. You were either hired or not. Promoted or not. Accepted or not. It would be difficult to write a 4% less glowing recommendation. Treating as ‘chance of experiencing discrimination’ is probably going to be more reflective of people’s lived experiences than ‘4% worse outcome’, even though the averages are the same.

            Report comment

          • Patrick says:

            Or, all of those minute differences compound over time to yield the 4% you measure at the end. Or, all of those 4% differences affect different people in different amounts, and don’t compound at all. Or, once the difference has occurs, the populations are no longer the same and this 4% rule we’re presuming applies everywhere no longer applies at all. Or, all of those 4% differences aren’t in black and white issues that have meaningful effects on people’s lives. To use the obvious example, the difference in submission acceptance is unlikely to have a significant impact on the lives of even the subset of people who are differentially affected, AND, that difference occurs long after the chain started, not at the beginning.

            Report comment

      • Kaminiwa says:

        “As usual, it’s humanity that messes things up. When flipped by a machine, coins come up heads a solid fifty percent of the time, and tails the other fifty percent. Put the fate of the coin in grubby human hands and the odds tip slightly in favor of the side that faces up just before the coin is flipped. The side that was face up at the beginning of the flip has a fifty-one percent chance of landing face-up at the end.”

        http://io9.gizmodo.com/5826157/coin-tosses-arent-really-fifty-fifty

        So, 51% heads is actually pretty typical ^.^

        Report comment

      • Muga Sofer says:

        One could argue that a small statistical difference is more likely to be a result of the study’s design than to be “real”, and – being undetectable – we would have no way to disprove that theory.

        Report comment

    • ediguls says:

      In fact, it would be better not to use a bar graph at all for such a small amount of numbers. A table would be more informative and have a greater data-to-ink ratio, to channel Tufte.

      Report comment

  7. Nadja says:

    Bravo, this is why I love this blog.

    It boggles my mind why people can’t embrace these (preliminary) results for what they are: it’s great news there doesn’t seem to be much gender discrimination in the open source community! Shouldn’t this be helpful to the feminist cause? Show this result to young girls. Tell them that if they feared going into tech because they believed there’d be widespread gender discrimination, here’s a piece of evidence showing that their fears might be overblown. Hurray… Right?

    Report comment

    • The Smoke says:

      Totally agreed. However most feminists wouldn’t be happy to live in a world where the feminist goals are actually realized, since in such a world you can’t be fighting for feminism anymore. In particular this is good reason for them not to accept the notion that things are not actually so unfair right now.
      With feminist here I mean ‘actively part of the movement’.

      Report comment

      • phantasmoon says:

        That’s uncharitable. I’m sure most feminists would like to live in that world. Unfortunately, their worldview is based around the idea that there is a systematic discrimination against women, that good intentions and even individual actions aren’t enough to make a difference, and that people who distract from or hinder the cause of raising awareness about, and fighting against, “the patriarchy”, are hurting the cause of all women.

        If the evidence seems to point out that there doesn’t seem to be much discrimination in tech, well, the evidence is wrong because of course there’s a patriarchy and of course there’s a problem. And if you present good news about the status of women in tech, or any other area, they will attack you because you’re distracting everyone from the fact that the patriarchy is real and must be dismantled.

        Report comment

        • ThrustVectoring says:

          You’re not actually disagreeing. Feminists would like to live in a world where gender equality has “won”. They’d also be unhappy about living in that world, due to the sudden loss of raison d’etre.

          Report comment

        • Patrick says:

          Easy test- provide a feminist evidence that there is little or no sexism in an area of life she previously believed had a lot of sexism. See how she reacts.

          For example: Provide evidence that the differential effect of gender unblinding in submission acceptance for men versus for women is INCREDIBLY small, to the point where you’d need a couple hundred female contributors in a room before you’d have enough people its affected to fill out a hand of bridge. Point out that the effect size is so small that if someone tells you that their submission was negatively affected by their revealed gender, your default assumption should be that they’re wrong, because false positives are likely to easily outnumber actual incidences of that occurring. Point out that the tech world, at least in this aspect and per this survey, is less sexist than… uh, virtually every other community, ever.

          See how that goes.

          If she’s happy, you’ll have your answer.

          Report comment

          • dust bunny says:

            I don’t agree. You’re holding feminists to an unnaturally high standard of rationality, as if they weren’t fairly average people.

            A typical person, when faced with evidence that their belief system can’t score a “win” in some specific context, will not actually update their priors. They will perceive whoever is presenting the evidence as hostile to their belief system and their group. And that’s actually a very good heuristic.

            It would be nice if feminists were actually better people than literally every other major group everywhere, but sadly, that’s not the world we live in.

            Report comment

          • Mark Atwood says:

            The difference here, and why it is necessary they be crucified on this standard: their entire belief system is based on and goalled towards making everyone else in their world change their own minds, or at least pretend they have, under threat of government power and social and economic exclusion.

            Report comment

          • Luke Somers says:

            I think a better test would be if you show them a case where there WAS bias, and then over time the bias shrank to what you describe.

            That would give them the choice of YAY WE WON, which is kind of missing from what you describe there.

            Report comment

          • butts kapinsky says:

            Hi, Patrick.

            I am a feminist and am extremely interested in the evidence you’ve listed. Could you please link the relevant studies? I find it annoying both when feminists and MRAs push bad or incorrect information (Examples: feminists sometimes don’t like it when you tell them domestic violence is largely bidirectional, and MRAs sometimes don’t like it when you tell them that women ATTEMPT suicide more than men, and that this is a more relevant figure than SUCCESSFUL suicides).

            My current understanding of the research is that, in general, identical resumes will see the female name rated as less competent, offered a lower starting salary and rated as less likely to hire. Screaming patriarchy at this is a bit extreme, but it is evidence of a system which rewards women less for the same amount of work.

            Just remember to update your priors of feminists after I’ve seen the research.

            Cheers.

            Report comment

          • InferentialDistance says:

            this is a more relevant figure than SUCCESSFUL suicides

            I keep seeing this asserted, but I’m not persuaded. Society is much more sympathetic to women’s sadness and weakness, which could unconsciously encourage (poor) suicide attempts.

            Report comment

          • dust bunny says:

            @ Mark Atwood

            I agree that the problem of feminist Schlimmbesserung is real and needs to be addressed. But I don’t see how gotchas directed at individual feminists for basic failures at rationality help at all. Blaming isn’t likely to help either, since it’s going to be tribal. Feminists can entirely reasonably ignore such blaming, because supporting policy and intervention without good evidence is a ubiquitous problem, not a feminist one.

            I’m going to be predictable and offer a “more of the same” type solution. Try to shift culture so that people value good evidence more highly, and understand what constitutes good evidence better, and know why good evidence is relevant and important in policymaking. Feminists are not impervious to reason at all.

            Report comment

          • FacelessCraven says:

            @butts kapinsky – I do not think it can be defensibly claimed that surviving suicide is a worse outcome than dying from it, any more than I think it can be defensibly claimed that women are “the real victims of warfare”. I would be interested in hearing your reasoning though, if you wouldn’t mind elaborating.

            Report comment

          • Hyzenthlay says:

            women ATTEMPT suicide more than men, and that this is a more relevant figure than SUCCESSFUL suicides

            I’d say that both are relevant. And it’s kind of depressing that both sides treat it as a contest and seem to be vying for the spot of “most suicidal.”

            Report comment

          • Schlimmbesserrung: An effort at making things better which actually makes them worse.

            Report comment

      • Vox Imperatoris says:

        With feminist here I mean ‘actively part of the movement’.

        To me, that’s the problem with using “feminist” uncritically in this way.

        If I start a movement for blue-people equality, and we actually do achieve equality for blue people, well, who’s going to be left calling themselves “bluists” in the “bluist movement”? Only the people for whom equality isn’t enough, and who want actual government privileges for blue people.

        But by any objective measure, in such a world, nearly everyone actually is a “bluist”, and “bluism” has achieved its goal—except perhaps in a few minor areas here and there.

        Report comment

        • John Schilling says:

          Do we have anyone left who calls themselves an “abolitionist”?

          Do we need to? Slavery still exists in some corners of the world, but at a level where opposition is generally rolled into general human-rights activism rather than having a large movement dedicated to that singular cause.

          Report comment

          • Vox Imperatoris says:

            But is it a good idea to say “I’m opposed to abolitionism”?

            I don’t think so. Not even if people are calling for preferential racial quotas in the name of “fighting slavery in the workplace.”

            It’s deliberately ceding the moral/connotative high ground for no particular reason.

            ***

            Especially if there were a small but vocal number of people out there who were legitimate, hardcore anti-abolitionists.

            Report comment

          • Nornagest says:

            We have anti-slavery activists. I think “abolitionist” has passed out of use because slavery is legally abolished pretty much everywhere, it just continues in some places extralegally or on a customary level, where abolition (double abolition?) would do it no further good.

            Report comment

          • John Schilling says:

            But is it a good idea to say “I’m opposed to abolitionism”?

            If we had a sizable group who called themselves “abolitionists” today, I would wager that A: the ones who focused their activities on places like Mali and the Sudan would get almost zero publicity(*), and B: the ones who focused their activities in the United States and Western Europe would be doing things that sensible people should oppose.

            * Unless they hired Emma Watson as a spokesperson, of course.

            Report comment

          • Vox Imperatoris says:

            @ John Schilling:

            I am not disagreeing with anything you said in that comment. The question is whether it’s a good idea to let those people take over the term “abolitionism” and declare yourself an “anti-abolitionist”. Especially when there is the very real chance that you will be conflated (purposefully or not!) with the small segment of people who really do want to bring back slavery.

            Let me stop using metaphorical language: there really are legitimate anti-feminists out there who do believe that woman are inferior, or at the very least that men and women belong in “separate spheres”. Mostly, they are religious traditionalists, though some of them are in the alt-right or the “novo-regressive” movement. It is foolish to let yourself be grouped in with those people, if you in fact share the feminist belief in the equality of men and women but disagree on the proper means of promoting it.

            Worse, the actual anti-feminists actively want this confusion to occur because it makes them seem more reasonable. In fact, some of them are most likely posting here.

            @ Nornagest:

            That is surely true. But it’s beside the point, which is whether a group calling for racial quotas should be allowed to get away with calling themselves “the true heirs of the abolitionist movement” without any pushback.

            Report comment

          • The Smoke says:

            I agree with what you say, but let’s apply this to self-identifying as feminist. It doesn’t actually tell really anything about yourself.
            Does it mean you want i) equality before the law (that wouldn’t justify using that label, because it is the baseline assumption anyway), that you think ii) women should get special attention because of perceived prevailing inequalities, or even that iii) men should be out under special scrutiny because of the possible danger they represent?

            If I don’t have any more information I am going to put the person somewhat in the middle, dependent on the typical views of persons that identify as feminist I have encountered. For me personally this would roughly mean ii).

            I also agree that I should have just written what I mean directly.

            Report comment

          • Vox Imperatoris says:

            @ The Smoke:

            If it’s just an average person you meet at work or some social event, and you hear him or her identify as a “feminist”, the term is simply too vague to determine their precise views. Especially if it’s a woman: I think most woman these days are simply going to reflexively agree with feminism in the broad sense.

            Most people have not thought about this very much. You are going to go wrong assuming everyone who casually identifies as “feminist” is a “feminazi”.

            Now, if we’re talking about someone who runs a “feminist” blog, dedicated to “smashing the patriarchy”, then you can make some more assumptions.

            But in dinner-party conversations, you have to keep in mind that the whole thing is a conceptual morass, and most people are confused about it. One person may call himself an “anti-feminist”, while another may call herself an “ardent feminist”—and they may well share the exact same views.

            Especially when we’re talking about things like the entertainment industry. It is not at all unreasonable, in my opinion, to point out that Hollywood in many ways promotes a certain idea of gender roles—because it sells. Or in video games: only in the past decade or so has there really been a consistent move toward creating equal female characters.

            Report comment

          • transparentradiation says:

            Vox Imperatoris. You’re a very fair personage. My compliments. And you identify as an Objectivist, is that correct?

            Report comment

          • I wonder if any of the people who say “wage labour is slavery!” call themselves abolitionists.

            Report comment

        • The Smoke says:

          Whenever somebody uses the word feminist they have a certain definition in mind. Then when someone reads/hears the word they might think of another. This alone causes a lot of misunderstandings.

          There are little to no movements (-isms) where the label is best used to describe people who identify with the goals of the movement, but rather people who identify with the movement itself and who advocate serious systemic changes.

          I think my use is much more adequate and consistent than ‘someone who believes women should have equal rights’ as it is often used, the only difference is that I am upfront with what I mean.

          Report comment

          • This. If you give a bunch of people a political compass quiz, many of them end up in the socially liberal/fiscally conservative quadrant. You’d be crazy to use this fact to claim that half the population are libertarians. Probably less than half the population have any idea a libertarian movement even exists, much less personally subscribes to it.

            Report comment

          • Jonas says:

            You’d be crazy to use this fact to claim that half the population are libertarians.

            That depends on what the meanings of ‘are’ are 😉

            If people hold substantially libertarian views and vote libertarian where possible, then they are libertarians in deed. If they call themselves libertarians, they are libertarians in name. (All four combos probably exist.)

            Your point seems to be that they don’t self-identify as libertarian—that they’re not libertarian in (self-applied) name. I think it’s much more interesting whether they’re libertarian in deed, in substance.

            You could have argued that the political compass is a bad measure for this, but you didn’t. Is that because you think the name side of the name/deed split is more important (and the strength of the test is irrelevant), or is it because you think the PC actually is a fairly good measure?

            Report comment

          • Adam says:

            I think he’s saying it’s a really big quadrant and you have to be pretty far into the corner of it to really be a libertarian. Many people are in the quadrant but usually close to the center or at least close to the center on the economic axis.

            Report comment

          • Consider the statements “I’m a libertarian because…” and “the problem with libertarians is…” Hearing either of these in a normal conversation, we would expect people to be referring to (1) people who self-identify with the libertarian movement, or (2) people who vote for political candidates on the basis of their agreement with libertarian views, or (3) people whose stated views have significant overlap with views that the speaker recognizes as libertarian. We would not think they were referring to everyone whose views fall within the broadest possible definition of libertarianism.

            Similarly, when someone says “I’m a feminist because…” or “the problem with feminists is…,” they’re not referring to everyone who would agree with the statement “women are people, too!” They mean people who self-identify with the feminist movement, or whose views are strongly aligned with those of the feminist movement.

            Imagine if libertarians popped up in the comments of every anti-libertarian article to say, “libertarianism just means you think people aren’t slaves to their governments! The only reason you could possibly oppose libertarianism is if you want the universal slavery of a totalitarian state.” I think we can all agree that that would be a terrible argument.

            Report comment

      • AspiringRationalist says:

        > Most feminists wouldn’t be happy to live in a world where the feminist goals are actually realized, since in such a world you can’t be fighting for feminism anymore.

        I think you’re confusing “most” with “the loudest”.

        Report comment

    • Yakimi says:

      Analyses of bureaucracies should also apply to movements, especially to ones as thoroughly institutionalized as feminism, as entities that have to manufacture problems to justify their continued prominence. Those employed in activism and journalism can only be expected to argue in favor of a world in which they become more important and powerful.

      Report comment

      • Ptoliporthos says:

        I was wondering if it was more like Conquest’s third law and the feminist movement is actually run by the patriarchy. It certainly looked that way this past week, anyway.

        Report comment

        • DrBeat says:

          There’s a good reason us MRAs sometimes refer to feminism as “The Ladies’ Auxiliary of the Patriarchy”.

          Feminism is what you get when you immerse yourself completely in every single belief that underlies sexism before thinking about how to fight sexism.

          Report comment

          • Maware says:

            When I flirted with the dark side, the explanation was that feminism benefits high-status/elite men, by disempowering the bulk of men. Feminism sexually liberated women to chase after the high status men they prefer in their youth, while enabling them to work at traditionally male jobs thereby depressing wages and making men compete and fail more.

            The losers were the majority of men who lost the stabilizing influence of wives and who suddenly found the economy not only shedding many male jobs due to productivity gains, but the increased competition for the remaining service jobs due to women. If you look at older media, it’s surprising to find men in professions now completely dominated by women, like store clerks and clerical type positions.

            Much of MRA philosophy is quite elegant (which is why it should be distrusted, it explains things TOO well) if you read it. Reading about male disposability was a shattering experience, for one.

            Report comment

          • houseboatonstyx says:

            @ Maware
            The losers were the majority of men who lost the stabilizing influence of wives and who suddenly found [… ] increased competition for the remaining service jobs due to women.

            A vivid image! Poor fellow, he no longer has a wife cooking for him , she’s outside competing with him. Lucky fellow, if he’d like to stay home and let her bring in some money.

            Report comment

          • Maware says:

            Wives don’t marry men who don’t bring in money, although they may sleep with them. Social roles are less flexible than you think: women are dissatisfied with house husbands for the most part, and they make up a tiny part of the population. Many stay at home dads will tell you of the isolation and issues with other women they have.

            What happens is that they redouble trying to marry the smaller amounts of men who do earn money, and increasingly remain single as they age and fail. Meanwhile you have a fair amount of underproductive, isolated, and unhealthy single men.

            This is the part of the MRA analysis which is interesting, but you wade through a lot of junk to get to it.

            Report comment

        • It turns out that I was familiar with some version of Conquest’s third law, but I didn’t know it was his. Have a link.

          1. Everyone is conservative about what he knows best.
          2. Any organization not explicitly right-wing sooner or later becomes left-wing.
          3. The simplest way to explain the behavior of any bureaucratic organization is to assume that it is controlled by a cabal of its enemies.

          Report comment

          • Samedi says:

            These remind me of John Gall’s “Laws of Systems” (https://en.wikipedia.org/wiki/Systemantics). If you consider an organization a “system” then the laws explain quite a lot. For example, “Complex systems tend to oppose their own proper function.” And, “The system itself does not actually do what it says it is doing.”

            Report comment

          • Jason K. says:

            I can summarize laws 2 & 3 into one:

            The implicit goal of any organization or movement will become its own perpetuation and growth. This can also be applied to any distinct sub-group of an entity, down to an individual member.

            Report comment

        • Muga Sofer says:

          What happened this week?

          Interestingly, I seem to have absorbed a definition for “Conquest’s third law” that was … significantly less american-conservative than the original – “Any controversial movement is eventually controlled by those [in power] whom it claims to oppose.” I wonder where it came from.

          Report comment

      • Samedi says:

        In my field we call such organizations “self-licking ice cream cones”. A phrase which I find highly amusing.

        Report comment

    • James says:

      Great point.

      Some MRAs (Honey Badger Radio) have gone so far as to say ignoring such results and continuing to preach (listen and believe) gender discrimination is tantamount to abuse.

      see: self-fulfilling prophecy, no autonomy, no moral agency.

      Report comment

    • curtains says:

      You guys don’t get it. There is no conspiracy. These journalists are just farming for clicks the way they always do. Rage gets clicks. There is no hidden agenda aside from boosting pageviews. All media is now tabloid media. At best, they are individually consistent about which side they take on all the hot button issues they are addicted to covering, in order to avoid alienating their readership.

      Ryan Holiday is one of the few media insiders who is willing to speak plainly about this. It applies to social justice issues as well as anything. Luckily, we are reaching “peak content” as adblockers proliferate and venture capital money runs out. Hopefully the next iteration of journalism is better that this one, and people figure out which news sources stayed credible and which ones didn’t.

      Report comment

    • Cord Shirt says:

      “Hurray… Right?”

      Mostly, yes! 🙂

      OTOH, there is also a caveat/concern that such studies may possibly be overlooking subtler issues, and that taking such studies at face value may lead to a premature declaration of victory.

      People whose main exposure to feminism is as an excuse to attack them will understandably feel exasperated by this hesitation. But IMO it tends to come from those who have a lot of experience of sexism…and often also of prematurely declaring victory.

      I mean, what dust bunny said applies too–typical people often don’t update their priors–but, well…as dust bunny said, it often *is* a good heuristic when you have strong support for your priors…as in Eliezer’s “I defy the data.”

      Report comment

  8. vV_Vv says:

    Of course, this doesn’t rule out that one gender is genuinely doing something differently than another, so they had another neat trick: they wrote another program that automatically scored accounts on obvious gender cues: for example, somebody whose nickname was JaneSmith01, or somebody who had a photo of themselves on their profile. By comparing obviously gendered participants with non-obviously gendered participants whom the researchers had nevertheless been able to find the gender of, they should be able to tell whether there’s gender bias in request acceptances.

    So we are to believe that their automated script is able to determine the gender of participants better than the humans who decide whether to accept pull requests. I don’t buy it.

    The default hypothesis for the difference in acceptance rates between “blinded” and “obvious” gender should be that it is due to systematic measurement bias.

    Report comment

    • Scott Alexander says:

      I think the theory is that their script compares email addresses to Google Plus pages. Although a human could do that, most people don’t. In theory, right now I could Google the email address you used for your SSC registration, find your social media profile, and see if it contained gender cues. In practice, you’re just a series of Vs.

      Report comment

      • vV_Vv says:

        In theory, right now I could Google the email address you used for your SSC registration, find your social media profile, and see if it contained gender cues. In practice, you’re just a series of Vs.

        If I sent you code to publicly include in your project, possibly with the implicit promise to maintain it for some time, wouldn’t you be a bit curious about who the hell I am?

        Report comment

        • Mary says:

          The whole point is that they care about the code, not you.

          Report comment

        • ThrustVectoring says:

          Not particularly, as long as I can associate further code contributions as coming from the same source. If the patch is good, accept it. Otherwise don’t. If I’m hiring or looking for someone to speak at a conference, your code contribution has some means of contacting you. If you wanted to advertise who you are with your code, you’d be doing so already.

          Report comment

        • Eric Anholt says:

          I’m an open source developer, and do a mix of traditional projects and github. I have accepted contributions from people on github without knowing real name. However, in general we’ll have exchanged some emails outside of GitHub, so I expect a lot of the “blinded” population really wasn’t.

          I am surprised that gender-ambiguous usernames (which I think of as those people who go by MostAwesone or whatever) did better than non-gender-ambiguous usernames. I’m seriously prejudiced against contributions from people that won’t put their name on it, and I’m definitely not alone, so I wonder what’s going on here.

          Report comment

        • Anon says:

          Yeah, no, as one of the maintainers of a large project, having worked with / contributed to a number of others, no one does this. It’s basically either you recognize the handle or you don’t, and maybe glance at the other repos contributed to. Even then, it doesn’t make much difference: all that really matters is if the code would be a positive change.

          Report comment

        • As a maintainer of a few small projects, this has literally never occurred to me

          Report comment

        • patches welcome says:

          No, and the insinuation irks me slightly. If I’m “curious about who the hell you are”, I have `git log –author` and the like. Why would a maintainer ever “find your social media profile”?

          Report comment

          • voidfraction says:

            I’ve searched for people’s resumes after they contributed to a project open-sourced by my company, but that was just to see if they’d be a good candidate for an open role we had.

            Report comment

      • Anonymous says:

        “I think the theory is that their script compares email addresses to Google Plus pages. Although a human could do that, most people don’t”

        I do it all the time to people who open pull requests or issues on my github project to satisfy my voyeuristic tendencies.

        Report comment

      • Deiseach says:

        In theory, right now I could Google the email address you used for your SSC registration

        I just did that right now. It brought up a handful of results, three of which were indeed comments I had left on other sites*. The rest (as “best matches”) had to do with Connemara Pony societies.

        There was a popular fad a couple years back for an online text analyser which said that, given a sample of writing, it could identify gender. Like everyone else, I copied’n’pasted a chunk from my Dreamwidth journal into it. It said I was male.

        I am not a Connemara Pony, should anyone be in any doubt about that, neither am I male 🙂

        *Under a different name. If anyone wants to try Googling “Deiseach”, there is another Deiseach out there who’s not me 🙂

        Report comment

        • Muga Sofer says:

          To my eternal disconcertion, googling my email address (or name) will occasionally turn up sites hosting lists of cracked emails and passwords, usually in Russian.

          (Yes, I’m on them. By sheer luck, I changed my password before this was discovered.)

          Report comment

  9. I read that and my thoughts immediately jumped to Paul Graham’s essay on how to spot gender bias. The idea is that if you’re biased against women (or whatever) the ones you let through will have to be extra awesome and will subsequently out preform people who weren’t discriminated against. If one were to imagine that women were somehow less confident about submitting pull requests then the requests that were made would be fewer but higher quality and more likely to be accepted.

    Report comment

    • eponymous says:

      That depends on whether bias occurs only at an initial selection step, or is pervasive and affects subsequent performance.

      For instance, suppose fewer women enter tech because of the hostile work environment, and the hostile work environment also lowers women’s performance. Then we would expect higher-ability women to enter tech (compared to men), but they would under-perform their ability because of the hostile work environment, and so it’s not clear whether their final output would be higher or lower.

      (I’m not endorsing this hypothesis — I personally favor differential occupational preferences.)

      Report comment

    • The Nybbler says:

      Graham’s test was mostly nonsense. The definition was self-contradictory. He promises “…in some cases it’s possible to detect bias in a selection process without knowing anything about the applicant pool”. Then, one of the conditions is “the groups of applicants you’re comparing have roughly equal distribution of ability”. That’s not “knowing nothing about the applicant pool”, in fact that’s knowing quite a bit about the applicant pool. It also applies only to some sorts of bias; if you’re more strongly biased against higher-ability women than lower-ability women, the test might not work even assuming the applicant pool meets the condition.

      It also applies only to cases where you have one test and one measure. When you have many different tests, it fails due to the possibility of Simpson’s Paradox. For instance, if men are more inclined to submit to projects with maintainers with extremely high standards than women, more men’s pull requests will be rejected, even under an assumption of no bias and no ability difference between men and women in the pool of contributors.

      Report comment

      • Randall Randall says:

        In context, knowing nothing about the applicant pool means knowing nothing about the distribution of gender in the applicant pool, I would think.

        Report comment

      • Paul Graham’s intuition was right, but his math details were wrong. That’s ok, because you can fix it. Rather than looking at *median* performance of accepted groups, you can instead look at *min* performance of accepted groups.

        I.e., compare the stupidest accepted woman to the stupidest accepted man.

        https://www.chrisstucchio.com/blog/2015/paul_grahams_bias_test.html

        This works as long as the distributions are not bimodal with an empty gap right near the quality cutoff.

        Report comment

        • TrivialGravitas says:

          I think the specifics of venture capitalism prevent that from working. VC doesn’t work by trying to only choose successful clients but by choosing people with high payoff potential (and being diversified enough that the risks aren’t an issue). As a result they simply don’t care how bad the failures are, only how good the successes are.

          Report comment

  10. Github has become SJ-deranged, so it’s appropriate to use some salt. Recent coverage: http://www.businessinsider.com/github-the-full-inside-story-2016-2

    Report comment

  11. Newbie says:

    I feel like informal peer review like you’re doing could serve a function similar to snopes. Snopes calls out urban legends and blatant falsehoods and serves as a useful way to deflate viral myths. There’s another level of analysis that could be done in explaining how studies and experiments have been misreported, or where the conclusions of a paper aren’t born out by the data.

    It would basically be a media peer review site to hold news outlets responsible for their science reporting, and when they distort research/statistics to serve their own ends. A media that reports on the media, which is what a lot of these blog posts of yours have essentially been. I would love it if something like that existed and served as a trusted non partisan source for higher level fact checking.

    Report comment

  12. Anonymous says:

    As a nerdy straight white male programmer, that fact that people like me are constantly being propagandized against by the media is getting pretty wearisome. Add in the apparent surge of support for socialism among the young and it’s getting downright frightening.

    If I was an American I’d be thinking about buying a gun and at least having a backup plan in mind to escape the revolution, as paranoid as that might sound.

    Report comment

    • Anonymous says:

      As a nerdy straight white-ish male and very leftist non-programmer, it’s not the socialists you need to worry about. Both because socialists are too hopelessly divided to do anything and because many socialists (/leftypol/, for instance) are fanatically opposed to identity politics in any form and giving it its strongest critiques.

      The people you need to worry about are the regressive, illiberal social justice types. Many of whom are not socialists, many of whom call themselves socialists but seem curiously comfortable with top-down social change and corporate power.

      Report comment

      • Anthony says:

        I was reading some discussion about the “code of conduct” controversy in open-source circles, and one thought which struck me was “This is exactly the sort of corporate bullshit a lot of people in the open-source community were trying to escape.”. Funny that Social Justice requires submitting to corporation-style control.

        Report comment

      • Jon Gunnarsson says:

        Wait, why would it be curious that socialists are in favour of top-down social change? After all, that’s what pretty much every socialist government has done (or at least tried to do).

        Report comment

      • Viliam says:

        My (very simplified) model is that the “old left” used to be middle-class people trying to speak in the name of the working class; while the “new left” is upper-class people trying to speak in the name of the working class and the middle class. Of course each class projects their own thinking on those they pretend to defend.

        The old left (the socialists) wants to get rid of the “top 1%”, which would more or less bring them to the top positions of the society. The new left (the regressives) wants to divide the “bottom 99%” internally as much as possible, and eliminate all the microaggressions and free speech because those are the most frequent “dangers” that the most rich people face these days.

        Incentives. When you think about social justice, remember that their leaders (not the average members) are the rich kids, the future kings and queens of this world. Suddenly things start making sense. Both real equality and meritocracy are their enemies.

        Report comment

        • Alex Welk says:

          This. With an extra helping of This on top. I think it is almost too common to forget that there can and often is a serious disconnect between the various parts of a movement’s hierarchy (no matter how much it may say it has no hierarchy, someone is writing the academic treatises and talking to news media). Differences between the leaders and the movement as a whole could account for large parts of the movement’s appearance to outgroups.

          Report comment

    • perpippity says:

      You don’t need to be American to buy a gun. Lots of places in Europe, to name one, have plenty of guns. (If, on the other hand, you’re a Japanese citizen, you’re probably out of luck.)

      Report comment

    • The Nybbler says:

      You misunderstand the situation. If the revolution comes, it will be against these people, not by them. They are already on top. I suspect they will end up falling without an actual shooting war, however.

      Report comment

    • Samuel Skinner says:

      If you are referring to support for Sanders, the ‘socialism’ covers single payer and free college- it isn’t hard to figure out why young people would support things that would benefit them. Aside from that Sanders has ‘spend more on social security’ and… I think that is the extent of the socialism. The rest is social justice, railing against the financial industry and other populist measures like protectionism.

      Report comment

      • Thomas Jørgensen says:

        Single payer would benefit everyone except people working in medical billing and medical insurance – The US health care provision system is uniquely wasteful. It doesn’t make anyone better off that you spend twice as much on administrative overhead as Canada does.

        Thus, this isn’t so much “Socialist” as “What is currently being done is just horribly worse than known best practice, so lets quit doing it this way”.

        The free college thing would require a campaign of auditing US university administrative and prestige spending with fire and sword to be practicable. I mean, a lot of nations do this, but a lot of nations also don’t have universities that build the kind of bondogles and hands out the kind of administrative sinecures the US does…

        Report comment

        • Vox Imperatoris says:

          Single payer would benefit everyone except people working in medical billing and medical insurance – The US health care provision system is uniquely wasteful. It doesn’t make anyone better off that you spend twice as much on administrative overhead as Canada does.

          It may be more efficient to have one giant nationalized grocery chain than to have dozens of different government programs for providing groceries, all working against one another.

          But that doesn’t mean that any form of government-provided groceries is more efficient than a free market in groceries.

          Healthcare is not a public good. It just isn’t. It is rivalrous and excludable. The only thing that’s sort of a public good is fighting infectious disease.

          Report comment

          • Anonymous says:

            @Vox Imperatoris

            One argument for this is to point out obvious absurdities, like the price of an ambulance ride, and ask, “Why couldn’t I set up a competing business that offers the same thing but for closer to what it actually costs to run, completely undercutting everyone else?”. The answer to that question will point you to the factors that prevent you from doing this and thereby serve to artificially raise the price of ambulance rides (or whatever else).

            Report comment

          • Vox Imperatoris says:

            @ Anonymous:

            Exactly. Especially on your first point.

            As one economist (I think it was George Reisman) has said, take the “shortage of hospital beds”. If you had a free market in healthcare, an entrepreneur could simply buy a hotel, take on the non-critical patients from the hospitals, staff it with nurses, and then make money by charging way more than a hotel charges but way less than a hospital bed costs.

            I don’t exactly want to sing the glories of the Russian system, but when I was studying abroad in St. Petersburg, I became sick with gastroenteritis. The state-run healthcare system is free and…not good (actually, the doctors are good, but the accommodations are terrible), but I ended up spending five days in a private clinic, paid for by my American insurance. You can see it on Google Maps here.

            What was interesting to me were a few things:

            a) The price of every service was clearly posted in a little book you were given upon going in.
            b) The place had clearly been converted from a hotel.
            c) All the doctors and nurses were very good, professional, and spoke English.
            d) The accommodations and food were significantly nicer than I have seen in hospitals in the U.S.
            e) Everything cost at most half of what it would cost at hospitals in the U.S.
            f) They didn’t have every service, but they had a private ambulance constantly parked outside waiting to take people to the state hospital if necessary.
            g) It was located a block from the “Square of Proletarian Dictatorship”, where Lenin first ruled from the grounds of a boarding school for aristocratic girls, after overthrowing the Provisional Government. (The place is still the headquarters of the “Gubernator” of Leningrad Region, as can be seen on Google Maps. The left side says “First Soviet of the Proletarian Dictatorship.” The right says “Workers of All Countries, Unite!”)

            These doctors and nurses are almost certainly of the same or similar quality as doctors and nurses in the U.S., probably a large number of them would wish to practice in the U.S., but they are kept out either at the level of immigration or by not allowing their degrees to transfer. And that’s not to mention the difference in regulations that allowed them to run a place that like.

            From what I understand “private clinics for rich foreigners” is a pretty good little industry in many less-developed countries. I think they should take the principles behind those and apply them to regular hospitals for non-rich natives in developed countries.

            Report comment

          • Thomas Jørgensen says:

            The free market works well, as a general principle. As a matter of empirical fact, in health care, it doesn’t. Wait, let me modify this some. The *US* system of a health care market is an abysmal failure.

            Citizens of the united states of america pay an enormous surcharge and get no better care than is typical for a developed nation. And it’s not just that wages are higher – the differential persists when you do the accounting in “percent gdp” and when you dig down, you find that most of this extra expense boils down to far more paperwork.

            In some ways this is absurd, I mean, I am saying that having the government just take over this entire sector of the economy results in the job being done with 50% fewer bureaucrats being involved. But that appears to just be how the economy works, and as a general principle, governments and nations ought not to ignore the experiences of the rest of the world and keep doing things poorly just because doing them better runs counter to ideology.

            Report comment

          • Anonymous says:

            @Thomas

            I don’t think the claim “in the rest of the developed world, healthcare is fully nationalized, single-payer, free at the point of use, and all works much better than the US, or at least, just as well for much less money” is accurate.

            Are you perhaps from the UK or Canada? Most countries really don’t have an NHS. Most countries have a healthcare system that is private to some extent. Only a few have a nationalized, single-payer system.

            The problem is that the US’ system is in certain ways uniquely terrible, and as I argue above, this quite clearly is not down to the free market just making things enormously expensive for no reason.

            I’m all down for adjusting the theory when it doesn’t match the data, but you have to make sure your data is actually accurate first.

            Report comment

          • sweeneyrod says:

            @Anonymous

            While the UK and Canada’s healthcare systems are somewhat peculiar in having a single-payer system, the USA is a lot more peculiar (amongst first-world nations in not having universal health care of any sort.

            Report comment

          • Anonymous says:

            @sweeneyrod

            I’m not sure how that would cause the problems people cite with the US system, i.e. everything being inordinately expensive for seemingly no reason.

            Also, I’m not sure the differences with regard to who is paying for it are as great as you might think. According to this Forbes article, about 50% of healthcare spending in the US is by government.

            Report comment

          • HeelBearCub says:

            @Anonymous:
            We do have single payer for the most expensive time of life to be a healthcare consumer, when you are old and near death. So, it’s a little bit of an apple and oranges comparison to just site the 50% figure is particularly meaningful. But there is a fair amount of consen in studies that Medicare is WAY cheaper than any comparable private market costs.

            Report comment

          • Adam says:

            I have absolutely no clue what causes this, but the biggest personal problem I experience with healthcare is nobody can ever tell me what something is going to cost. Get a treatment, get quoted a price, pay it, then four months later receive a bill saying ‘oops, sorry, we actually ended up getting billed this, and your insurance actually paid less than we thought they would, and you now owe this.’ How is a market supposed to function when neither party knows what a service is going to actually cost until months, sometimes years, after it is rendered? This is probably more of an issue with the insurance providers than the service providers, who could quote an immediate cash price if you actually have the cash, but since we’re now forced to purchase insurance (and effectively were forced to before if you had a real job), doing that may as well be setting money on fire every month if you’re not going to file a claim when you require treatment.

            The other obvious issue with the insurance model is it’s not really insurance. Flood insurance works because most of the people paying premiums never need to use it, and when the really expensive disasters happen, the government pays anyway. Everyone has healthcare expenses at some point, and insurance largely works by hoping they’re old enough to be eligible for Medicare by the time those expenses get really big. It’s more of a forced savings plan than insurance, which is probably fine in principle, except for some odd reason we’re forced to contribute to multiple forced savings plans at the same time, one of which can only exist at a profit by stiffing us and the other of which can only exist at all by stiffing doctors.

            Report comment

          • jaimeastorga2000 says:

            @Adam: From Jim’s “How to do health care right”:

            The American health care system is socialism without a central plan, and capitalism without markets or prices. In America, the health care system is disturbingly expensive, and sucking up alarming and rapidly increasing amounts of taxpayer money. America has the best health care system in the world for the very rich and the very poor, but for those in between, not so good. For the non working and part time working affluent (me) it is woefully bad.

            Because there are no prices in the American health care system, there is no competition, so costs rise to absurd and astronomical heights.

            Stuff that is offered on a fixed price basis, for example dental surgery and laser eye surgery, works well, but almost all health care is offered on the basis of that they will do it, then afterwards make up a price on the basis of political power. For example my family has catastrophic coverage, which means we pay most ordinary medical charges out of our own pocket, but the insurance kicks in when we actually come down with something expensive. My wife was advised to get a colonoscopy. We shopped around, got a reasonable price at a doctor with a good reputation, negotiated with the insurance company, did all the stuff one does in an environment which actually has prices. Then after the colonoscopy was done, the hospital pulled a huge list of stupendously expensive charges out of their ass, most of which were obviously ridiculous or completely made up out of thin air, just trying it on to see what they could get away with, and all of which were charges we had definitely not agreed to, nor consented to in any way, formal or informal, written or unwritten. They just were not used to doing stuff on the basis that one has a definite price, and that the price one charges affects demand for one’s services. The concept seemed alien and incomprehensible to them. Mentally, they were socialists.

            Report comment

          • Nita says:

            @ jaimeastorga2000

            “Socialists” in the last sentence just means “badwrong”, not anything actually socialism-related. Trying to squeeze the maximum possible amount out of the buyer in every transaction (in other words, attempting to achieve “perfect price discrimination”) is clearly a capitalistic, profit-driven strategy.

            But I agree that your health care system is a good illustration of what happens when the free market assumptions don’t approximate reality even at a stretch.

            Report comment

    • noge_sako says:

      Its mostly overblown click-bait that doesn’t really professionally effect you.

      Every single guy I know who was a really good programmer in college who graduated (and also did well at school) got a good programming job.

      Report comment

  13. krstck says:

    As a female* programmer, I’d very much like it if everyone would stop analyzing me and my peers and leave us alone to do our work, goddammit. But any time I bring up my opinion on this, I’m hounded by people telling me to shut up because I’m sooooo privileged to not have experienced rampant sexism or something. But most of the anecdotes I read about how awful it is tend to be people choosing the least-charitable interpretation of some guy being socially awkward.

    Look. Women, in my experience, just aren’t interested in programming. I’m supposed to believe that this is the patriarchy or something, but in my friend-group, all of the people that have the slightest interest in the details of my job are men. And look at the sheer difference in number of PRs in this study by women versus men – 150,000 versus 2.5 million. Go browse recent projects on Github – any language – it’s all going to be men. And there’s nothing stopping any woman from forking an open source project and hacking on it, but *they just aren’t interested*. (Most men aren’t either, but the people who are interested are almost always men.) There aren’t a bunch of lonely female-owned forks sitting around, submitting PR after PR only to be rejected.

    Would I like more women to be interested in programming? Sure, I guess, I don’t really care. My being female doesn’t have the slightest impact on my ability to write code. If any woman wants to code, there’s literally nothing stopping her. In fact, there’s *tons* of resources out there explicitly begging her to learn!

    Again, I just don’t see this rampant sexism that’s keeping all these competent female programmers out of the industry. The female programmers just don’t exist in the first place, and that’s no one’s fault but their own.

    * When did we start saying “woman programmer”? That sounds awful.

    Report comment

    • perpippity says:

      Thank you! I have no problem saying if a woman can code and wants to, her work product should be judged on the same basis as anyone else’s. I have serious problems with “disparate impact as a categorical theory”. Women are capable of lots of stuff. Most just aren’t interested in the same things as men. You are– and more power to you.

      Report comment

      • The Nybbler says:

        > When did we start saying “woman programmer”? That sounds awful.

        Well, at some point, Our Betters (that is, the SJWs) decided that using “female” was some sort of geeky tic that is (wait for it) unwelcoming to women. Comparisons to the Ferengi of Deep Space 9 were brought up to show how awkward the term is.

        Report comment

        • Vox Imperatoris says:

          Terms like “woman doctor” and “woman lawyer” have been around since the Bad Old Days of outright sexism, where this was a necessary qualifier, since “doctor” or “lawyer” per se simply implied “male”.

          There was a similar debate with “woman senator” versus “senatrix”. They went with “woman senator”.

          Report comment

          • Yli says:

            I am now very saddened that ‘Senatrix’ did not become a thing

            Report comment

          • Vox Imperatoris says:

            @ Yli:

            Yes, most unfortunate. Aviatrix, too.

            I realized I forgot to include the obvious comparisons:

            “Woman doctor” versus “female doctor”: the latter did not take off until the mid-60s

            “Woman senator” versus “female senator”: same story.

            These may be confounded by “woman” appearing next to “doctor” for other reasons, so pay less attention to the relative frequency and more to the fact that virtually no one said “female doctor” or “female senator” in print before the mid-1960s.

            For additional anecdotal evidence, see the fact that Ayn Rand wrote a (ridiculous) article about why a woman shouldn’t want to be president. The title? “About a Woman President”.

            Report comment

          • brad says:

            The -trix suffix is the best part of will and trust law: testatrix, executrix, and rarely administratrix.

            Report comment

          • BBA says:

            Coining “litigatrix” is one of the finest contributions of AboveTheLaw.com to the English language, along with “benchslap.”

            Report comment

          • Deiseach says:

            Use of the term “woman doctor” from a short story collection published in 1894 (story is “The Doctors of Hoyland” by Arthur Conan Doyle):

            “Verrinder Smith, M.D.,” was printed across it in very neat, small lettering. The last man had had letters half a foot long, with a lamp like a fire-station. Dr. James Ripley noted the difference, and deduced from it that the new-comer might possibly prove a more formidable opponent. He was convinced of it that evening when he came to consult the current medical directory. By it he learned that Dr. Verrinder Smith was the holder of superb degrees, that he had studied with distinction at Edinburgh, Paris, Berlin, and Vienna, and finally that he had been awarded a gold medal and the Lee Hopkins scholarship for original research, in recognition of an exhaustive inquiry into the functions of the anterior spinal nerve roots. Dr. Ripley passed his fingers through his thin hair in bewilderment as he studied his rival’s record. What on earth would so brilliant a man mean by putting up his plate in a little Hampshire hamlet?

            But Dr. Ripley furnished himself with an explanation to the riddle. No doubt Dr. Verrinder Smith had simply come down there in order to pursue some scientific research in peace and quiet. The plate was up as an address rather than as an invitation to patients. Of course, that must be the true explanation. In that case the presence of this brilliant neighbour would be a splendid thing for his own studies. He had often longed for some kindred mind, some steel on which he might strike his flint. Chance had brought it to him, and he rejoiced exceedingly.

            [Dr Ripley goes to call on his new medical neighbour]

            …Turning round, he found himself facing a little woman, whose plain, palish face was remarkable only for a pair of shrewd, humorous eyes of a blue which had two shades too much green in it. She held a pince-nez in her left hand, and the doctor’s card in her right.

            “How do you do, Dr. Ripley?” said she.

            “How do you do, madam?” returned the visitor. “Your husband is perhaps out’?”

            “I am not married,” said she simply.

            “Oh, but I beg your pardon! I meant the doctor – Dr. Verrinder Smith.”

            “I am Dr. Verrinder Smith.”

            Dr. Ripley was so surprised that he dropped his hat and forgot to pick it up again.

            “What!” he gasped, “the Lee Hopkins prizeman! You!”

            He had never seen a woman doctor before, and his whole conservative soul rose up in revolt at the idea. He could not recall any Biblical injunction that the man should remain ever the doctor and the woman the nurse, and yet he felt as if a blasphemy had been committed.

            Report comment

          • Anthony says:

            Best in class:

            “Lady Novelist”. So much better than “authoress”.

            When I lived in the Berkeley Coops, one of the house managers gave her title as “House Manageresse”.

            Report comment

          • Maware says:

            Don’t forget asterix, obelix, dogmatix…

            Report comment

        • krstck says:

          > Well, at some point, Our Betters (that is, the SJWs) decided that using “female” was some sort of geeky tic that is (wait for it) unwelcoming to women.

          Like…I *kind* of get that, if someone is using “man” and “female” in the same sentence, as if to highlight their sex *specifically*. But “female programmer” is the correct terminology to distinguish from “male programmer”!

          Is caring about correctness some kind of male-only trait? Maybe that’s why so few women write code. I’m being facetious, but if someone’s feelings are so fragile that “insensitive” terminology is going to cause them to log out of their terminal, shut their text editor, and never write a line of code again….. well, you didn’t care enough about it to be doing it in the first place.

          Don’t get me wrong – I don’t think anyone should be a dick to women on purpose or whatever. I’m just tired of people talking about it! I want people that talk about gender instead of code to feel bad. I don’t want people to feel like they can’t reject my PR because I might cry about Code of Conduct. I don’t want people to wonder if I only have my job because I’m a diversity hire. I’m tired of this idea that my gender means I need my hand-held and I need to be tip-toed around. The only thing that should matter is: do my tests pass?

          Report comment

          • Jason K. says:

            It isn’t going to go away because the core motivations driving the behaviors are intrinsic to the genders and how they interact. It is worse now because these social signalling games used to be largely restrained to one’s circle of acquaintances. With social media and a content starved 24hr news cycle, a story that plays to the narrative we want to see (in these cases, we want to see it because it allows us to play and reaffirm our roles) can spread cross-country with nary any scrutiny and bring intense pressure onto the ‘villain’ of the moment.

            Report comment

        • Svejk says:

          I though it was ‘female-as-a-noun’ that was objectionable, not ‘female-as-an-adjective’? The former does seem to have an inbuilt Ferengi sneer.

          Report comment

          • Vox Imperatoris says:

            You’re right.

            This whole little subdiscussion is misguided, as I tried to indicate above.

            If anything, “woman x” sounds traditionalist and dismissive of women to me, but that has been the standard usage. As in “woman doctor”. Is it ungrammatical? Not really. No more than “research paper”. English has no strict separation between nouns and adjectives.

            Report comment

          • rockroy mountdefort says:

            feminists object to both, because their objections to “female” are built on the eminently rational basis of “inbuilt Ferengi sneer”

            Report comment

    • jsmith says:

      * When did we start saying “woman programmer”? That sounds awful.

      I don’t know, but it’s poor grammar. Woman is a noun, not an adjective.

      Report comment

      • Liskantope says:

        I’m not sure that it’s grammatically incorrect to stack two nouns like that. Compare to “lion king” or “child prodigy”.

        Report comment

      • Viliam says:

        I am not a native English speaker, but I always felt like in English every noun can be an adjective, or more generally, anything can be anything. You take a noun and ask “is there a specialized adjective form of this word?” — if the answer is yes, then you use the specialized form; if the answer is no, then go ahead and use the noun.

        (Even in the previous paragraph, I used “adjective” twice, once as a noun, once as an adjective. Actually, I was tempted to end it by “…go ahead and adjective the noun” so I would also verb it the third time, but I didn’t feel completely comfortable with that. Perhaps “adjectivize” would be the proper form here?)

        Report comment

    • Anonymous says:

      When did we start saying “woman programmer”? That sounds awful.

      I suspect it’s because “female” continues to have strong connotations of certain common biological features. Whereas “women” have penises, don’t have penises, and everything in between (at least so I have been educated, over and over).

      Report comment

    • Good Burning Plastic says:

      “Woman programmer” sounds like it means someone who programs women to me.

      Report comment

    • Winter Shaker says:

      * When did we start saying “woman programmer”? That sounds awful.

      Hey, at least it wasn’t programmeress 🙂

      Report comment

    • Muga Sofer says:

      I must say, I’ve been surprised by how much more conservative – or at least antifeminist – the women I’ve met in programming courses at college have been. Much more than either the men or women I know from outside colledge.

      I don’t have any hard numbers – or a decent sample size – but anecdotally, women in computing fields seem to be noticeably more conservative than men (who seem to skew conservative/libertarian themselves.) I have absolutely no idea why this would be, but it’s interesting.

      Report comment

      • Liskantope says:

        Because women who are more immersed in left-wing feminism are more likely to receive dire warnings about how hostile the male-dominated environment of programming is?

        One obvious problem with that hypothesis is that it would imply that women in other male-dominated arenas, including math departments, would also be more conservative and/or less feminist, which isn’t consistent with my experience. Still, people in academia seem to be more liberal in general.

        Report comment

      • NN says:

        There seems to be a definite pattern here, since it shows up internationally as well. The countries in the world with the greatest percentage of female science degree holders are, in order: Iran, Oman, Saudi Arabia, the United Arab Emirates, Romania, Algeria, Bulgaria… You get the idea. Or at least, that was the case back in 2011 (I’ve heard that the Iranian government has since instituted strict gender quotas in their universities, so their numbers might be different now).

        I would speculate that it might have something to do with people from conservative backgrounds valuing potentially lucrative jobs more than “following your dreams,” but I don’t know enough to say for sure.

        Report comment

        • Sandy Beach says:

          Look up the gender equality paradox. Basically, the more equal a society is, the more it will follow gender norms.

          Report comment

        • Francisco Boni says:

          To the extent that the sex roles of society-at-large affect sex differences in aggression, one would expect smaller sex differences in cultures where boys and girls are socialized more similarly, and larger sex differences in patriarchal in which boys and girls are socialized dissimilarly. Empirically, studies find the EXACT OPPOSITE. Sex differences are largest in progressive sex-role cultures (e.g., Scandinavia) and smallest in more patriarchal cultures (Schmitt, 2014).

          It is also possible that gender roles and socialization practices may, themselves, be evolved mechanisms (see Pirlott & Schmitt, 2014). For example, adaptations in parents may lead to differential socialization of boys and girls; these adaptations may be sensitive to local contexts—such as societies with high levels of warfare socializing boys more than girls to especially tolerate pain.

          To the extent that the sex roles of society-at-large affect sex differences in aggression, one would expect smaller sex differences in cultures where boys and girls are socialized more similarly, and larger sex differences in patriarchal in which boys and girls are socialized dissimilarly. Empirically, studies find the EXACT OPPOSITE. Sex differences are largest in progressive sex-role cultures (e.g., Scandinavia) and smallest in more patriarchal cultures (Schmitt, 2014).

          There are many problems with the “patriarchy” or “it’s just because society doesn’t let girls be as aggressive as boys” explanation of sex differences in aggression. These sex differences demonstrably stem from genetic-developmental experiences (including in utero masculinization), are consistent across species, and are universal across cultures. And to the extent they do vary culturally, sex differences are largest in progressive sex-role cultures (e.g., Scandinavia) and smallest in more patriarchal cultures (Schmitt, 2014). If patriarchy and sex roles are supposed to cause large sex differences in aggression, these social forces aren’t doing a very good job.

          “Mate preferences actually get LARGER (not smaller) in nations with more gendered economic equality. Zentner and Mitura (2012) found sex differences in preferences for physical attractiveness increase from a small effect size (d = 0.24) in lowest gender equity nations to a moderate effect size (d = 0.51)”

          “These large sex differences are especially noteworthy as they emerge from a highly egalitarian nation (Norway) with high paternal investment expectancy, and because they contradict social role theories that predict a diminution of psychological sex differences as gender economic equality increases.”

          http://www.sciencedirect.com/science/article/pii/S0191886915003748

          Report comment

    • Error says:

      As a female* programmer, I’d very much like it if everyone would stop analyzing me and my peers and leave us alone to do our work, goddammit

      Do you want to date?

      That’s not entirely rhetorical, but I’d like to know if, in your professional life, you get asked that far more often than you’d like. I’ve always had the secret suspicion that the feminist complaints about insufficiently-feminized tech have less to do with gender politics than sexual politics; an environment filled with too many men they don’t want to date, but so demographically tilted that there aren’t enough No’s in the world. That could easily feel like a hostile environment even if it isn’t.

      Last year I attended a convention panel on the subject. Most of the panelists were women (as a rule I pay much more attention to complaints about the field from women who are actually working in the field). One of them had an interesting anecdote: She was discussing the subject with male coworkers who were also complaining about not enough women in the field. She objected and pointed out a number of them in the office. They sort of grudgingly admitted that those weren’t the kind of women they were talking about. i.e. not young and attractive. Her conclusion: “For some of these people its not about improving the lot of women, it’s about improving their dating pool.”

      And my thought was “Well…duh? I happen to like it when my dating pool is made up of people with shared interests, and that’s more than enough reason to support getting more women into tech. I expect most other people do too.” Beyond that I don’t care, because I don’t really see artificial barriers; it’s not like the manuals and compilers and build tools care about the shape of your genitals. Except that’s not what women (or maybe just feminists?) want to hear — because they don’t want to date you, and they don’t want you thinking about dating them. It’s not enough to have shared goals; one must have shared motives to not Be Evil.

      Of course, I’ve just done exactly the analyzing you’re complaining about, so maybe I should shut up. I’m not actually sure what my point is here, so I guess I’ll close with that.

      Report comment

      • Nita says:

        they don’t want you thinking about dating them

        Seriously? Is that what you got from the story you just passed on?

        Male nerds resent being considered “not really men” (see a couple of bans in the previous thread), female nerds resent being considered “not really women”. I don’t see much of a difference at all.

        Imagine the gender-flipped scenario:

        Nurse Mary: Hey girls, isn’t it terrible that there aren’t any men working with us?
        Nurse George: Uh, that’s not quite true? For instance, I’m a man.
        Nurse Mary: Haha, you know what I mean, George! When I said “men”, I meant hot men.

        It seems like Mary was too busy thinking with her pussy to avoid blatantly insulting a colleague — kind of a bad thing, IMO.

        an environment filled with too many men they don’t want to date

        For me, as someone who still happens to be in the “really a woman” category, the only dating-related problem with guys in tech is that I had to choose just one (he was worth the sacrifice, though) 🙂

        Report comment

        • Error says:

          Seriously? Is that what you got from the story you just passed on?

          In retrospect, posting about a six month old impression of an anecdote that probably didn’t go exactly the way I remember it being described was probably a bad idea. Oh well, can’t edit it now.

          I do remember her complaint was about the motivation rather than about being personally insulted by it, which leads me to suspect the original incident wasn’t quite as blatent as you suggest (or perhaps that the man in question counted her in the ‘hot’ category).

          the only dating-related problem with guys in tech is that I had to choose just one

          I know I’ve internalized local weirdness when I have to read this twice to figure out what you’re talking about. O_o

          Report comment

        • Anonymous says:

          > Male nerds resent being considered “not really men”

          Well, male nerds also sometimes resent being considered men (the whole nerd scrupulosity thing), and female nerds sometimes resent being considered women (e.g. theunitofcaring).

          Nerd scrupulosity was a big part of my decision to transition, and I think it’s fairly common for trans women to have [had] the nerd scrupulosity thing.

          Anyways, I think “they don’t want you thinking about dating them” (at least as a ‘feminist’ thing) is a consequence of circlejerking on places like tumblr.

          Report comment

          • Error says:

            I’m not getting this. How does not wanting to be considered a man follow from nerd scrupulosity?

            Report comment

          • Cord Shirt says:

            I…sort of feel like in an ideal world nerd scrupulosity wouldn’t be a part of anyone’s decision to transition. Like…I guess I…alieve that nerd scrupulosity isn’t or “shouldn’t be” related to gender identification. Do you think I’m wrong?

            Report comment

      • Aap says:

        That could easily feel like a hostile environment even if it isn’t.

        That is a logical consequence of being a small minority, but I’ve actually never seen feminists make that argument. They always seem to claim that the men in these fields are more misogynist than men in more equal fields.

        Report comment

        • Nita says:

          They always seem to claim that the men in these fields are more misogynist than men in more equal fields.

          More like “the culture of these fields is about as misogynist as we expect a culture of a male-dominated field to be”.

          I don’t really agree with that either (I think it’s really heterogeneous, and much better than one would expect from the gender imbalance alone), but I’ve never seen the “nerds are the worst misogynists” complaint that’s frequently referenced around here.

          Report comment

          • Cord Shirt says:

            Nerds are more open about it (or any other “offensive” belief) and/or worse at hiding same. That’s all.

            Earlier feminists generally thought this was a good thing, because if someone’s willing to admit to something, you can argue with them about it. If OTOH the people with a given belief choose to avoid conflict by always hiding it…well then, you can never discuss it with them; you never get the chance to convince them they might be wrong. That’s one reason earlier feminism was pro-nerd. (Another is that Second Wave feminism actually is IMO a movement created by and for female nerds.)

            But as feminism has become more widely accepted, it’s been adopted by many less perceptive, less thoughtful people…who see nerds’ higher rate of “openly expressed sexism/misogyny” and mistake this for a higher rate of *actual* sexism/misogyny.

            Report comment

          • Nita says:

            @ The Nybbler

            Thanks, I took a look at the articles. Here are my impressions:

            (1) The Daily Beast – the author identifies as a nerd, argues that popular media promotes unhealthy attitudes, argues that reactions to Elliot Roger’s killing spree demonstrate these attitudes, says that “we [nerds] all need to grow up”

            (2) The Mary Sue – the author identifies as a nerd, criticizes a movie, argues that “we mistakenly assume that the guy wearing a Green Lantern T-shirt who can recite Monty Python and the Holy Grail in its entirety is harmless”

            (3) Charleston City Paper – the author identifies as a geek, ties his personal experience in the subculture to Elliot Rodger via the fact that he also was “a lonely, bright-ish virgin who retreated into the world of video games”

            (4) New Statesman – quotes a Facebook rant by a comic book artist, also mentions “gamers, comics and sci-fi fans, and even [..] board- and tabletop-gaming enthusiasts”

            (5) Jezebel – based on a reddit discussion, quotes redditors, also mentions gamers

            Sorry to be so contrarian, but literally none of these articles talk about men in tech fields, STEM students, or anything along those lines.

            Plus, 3 out of 5 are in-group criticism, written to caution fellow “nerds” against self-righteousness, and in all cases, the general idea seems to be “sexism in nerd subcultures does exist”, not “nerds are much worse than average”.

            Report comment

        • The Nybbler says:

          I’ve seen the argument made. It even has a name, which I don’t remember. However, the SJ types insist that this means it _really is_ a hostile environment. The logical conclusion is that in an environment where the gender ratio is lopsided in favor of men, the men should never ask the women on a date because that’s harassment. Fortunately not a problem for me, as I am monogamous with a woman in a different profession, but the children of some of my colleagues might have some things to say about such a policy.

          Report comment

      • krstck says:

        Well, I’m married, but even before that I never really received any interest from men. Going along with what @Nita said, female nerds are often considered “not really women” (by all parties involved!) and so our presence and concerns are dismissed. I have gotten a lot of “ok yeah, you’re a girl, but…” but what they really want to say is “you’re not one of THOSE girls”.

        Here’s the thing. Girls that grew up programming are generally not going to be the “hot chicks” that dudes typically want to date. I don’t feel like I have to disclaim “not all girls” to this crowd, but everyone KNOWS what I’m talking about. My husband thinks I have Aspergers, and though I disagree with him on that, I do feel a disconnect in my being able to communicate and relate to other women (and men). Being a “nerd” might be “cool” now, with cutie-girls loudly proclaiming their interest in Comic Books and Star Wars and going to conventions, but that absolutely wasn’t the case in the late 90’s-early 00’s when I was growing up.

        Personally, a lot of the complaining coming from the “women in tech” movement reads like a lot of #humblebrag to me. It’s just so hard to get work done when everyone wants to fuck you all the time, you know? I don’t have that problem, I’ve never had that problem, and my entire adult life I’ve been pretty much surrounded by men. But, you know, my feelings about being a “woman in tech” are irrelevant because I’m not a sexually-attractive woman in tech, which is *really* what everyone is talking about.

        Report comment

        • Error says:

          female nerds are often considered “not really women”

          Somewhat tangential, but: I do get the impression that a lot of the “women are excluded from tech/tech is misogynist” complaints don’t actually come from nerd girls, attractive or otherwise.

          Anyway, thanks for answering. It’s hard to get a handle on what things actually look like on the ground through all the noise.

          Report comment

        • lisa says:

          I think the sort of experience you describe is one that often gets lost in the discussions of sexism and harassment in tech and elsewhere. That can give off the sense that the concerns of feminism, “Women In Tech”, or of female nerds are not about you. I understand feeling that it’s always the shiny golden people who get to tell their story, and like your story is not listened to. It sucks to be treated like an outsider.

          I can relate somewhat to the experience you describe. I’ve been charitably described as “blunt” in how I talk to others. I was terribly nerdy growing up, a skinny, too-pale, glasses-wearing, braces-having, flat-chested geek. I don’t think joining the school server maintainance / AV club won me any points in social status. I am one of the girls you mention who grew up programming.

          Expressing interest in mainstream geek culture like Star Wars is quite unrelated to expressing one’s personal experiences with sexism in tech. Whether or not a woman’s interest in such things comes from an authentic place, I certainly agree that as geek culture has become more mainstream, it has become easier for women (as well as men) to show interest in these things without negative social repercussions. Today’s young women don’t have the same experience as we might have had in the early oughts.

          As a fellow Woman In Tech, however, I have had some bad experiences with unwanted male attention at my job. Talking about being sexually harassed in the workplace isn’t a humblebrag like “oops, teehee, all the boys want me!” It actually really sucks. Being touched without your consent or having one of your seniors offer to tutor you in JS in exchange for sex isn’t flattering or exciting, it is scary and gross – especially when you also have to worry about maintaining a good professional working relationship with your harasser and any other men who might be present. While there may be exceptions, I very, very much doubt that many of the women who discuss their bad experiences as Women In Tech are doing so to boost their ego by bragging about how all the men want them. They are sharing how bad it feels to be treated like your greatest way of contributing to the group is as a sex object. Again, it sucks to be treated like an outsider. The implication that we secretly want it is a pretty dangerous one.

          Report comment

          • Error says:

            Expressing interest in mainstream geek culture like Star Wars is quite unrelated to expressing one’s personal experiences with sexism in tech.

            It may be worth distinguishing here between geek culture (scifi, comics, games, anime) and nerd culture (physics, technology). The former has been mostly mainstreamed, the latter certainly has not; and while the overlap is heavy they’re not identical.

            Report comment

          • Jiro says:

            I think there’s a difference between an interest in a subject, and devoting a certain amount of effort to the subject. Watching Star Wars is mainstream. Writing Star Wars fanfic or making your own Stormtrooper costume isn’t.

            Report comment

          • My bet is that women have very different stories about what it’s like in tech because different companies (and probably divisions and such as well) have different cultures.

            Women aren’t having the same experiences.

            Report comment

      • Adam says:

        I think this discussion, certainly here but on the most of the Internet too, gets dominated by an extremely narrow view of ‘tech’ that basically just means websites and cloud services providers. My wife is a hardware engineer who has worked in defense her entire career, a field with far fewer women than most software and especially web development. But it’s also a field dominated by older men, so no, her complaints have never been about being hit on too much or being viewed as a sex object. It’s more to do with rarely being taken seriously, being talked over all the time, suggesting something and having it ignored only for the exact same suggestion to be enthusiastically adopted when it comes from a man, the general feeling of exclusion that comes from nearly always being the only woman in the room, etc.

        In her particular case, I think there are compounding factors. One, she’s younger than everyone she works with. Two, she’s extremely hard of hearing and very soft spoken and shy because of it and often difficult to hear when she does speak. Still, it pays to remember that not everyone in the world is under 30, and evaluating every person you ever meet as a possible hook-up partner kind of goes away when you get older, at least for most people.

        Report comment

        • Error says:

          It might be an accidental artifact of the term STEM. I know I think of tech as referencing computers and engineering as referencing the sort of thing your wife does, by your description, and to me those feel like very different professions and cultures.

          Report comment

          • Adam says:

            Oh believe me, she’d be perfectly happy with that. She’s an electrical engineer by trade, which sometimes involves coding, but for the most part, her work in the past involved integrated circuit design, not software, and at this point she’s graduated to technical lead on a field-testing team. She can’t stand the job-title creep of how many people these days call themselves ‘engineer’ and would be happy to distinguish between the engineering of physical systems and software systems.

            It’s just weird to me that all the hand-wringing seems to come from web companies, but when you look at numbers from the US BLS, web dev is like 30% female. Mechanical and electrical engineering are hovering at 8-9%. Theoretically, they should have it much worse, but the substance of the actual web company complaints seems to have less the character of being annoyed at a workplace dominated by men so much as being annoyed at a workplace dominated by people in their 20s.

            Report comment

        • Aapje says:

          I think there are compounding factors. One, she’s younger than everyone she works with. Two, she’s extremely hard of hearing and very soft spoken and shy because of it and often difficult to hear when she does speak.

          Don’t you think that these could be the primary reasons and her gender is at most a compounding factor?

          The problem with anecdotal proof is that it’s often very subjective and prone to be forced into a narrative. For example, I had a very similar experience in school as your wife, due to me being very low on the social ladder. If I was a woman, I might have concluded that this was because of my gender. But being a man, I obviously can’t use that narrative.

          Report comment

          • Adam says:

            To be perfectly blunt, I think she just suffers from crippling self-doubt and anxiety. They keep promoting her and giving her raises and she’s the only true engineer working with a bunch of retired pilots. The evidence to me suggests they tremendously value her for that, but she doesn’t see it because she’s a paranoid wreck. For instance, she’s convinced she’s going to lose her security clearance in spite of no criminal record, no credit problems, she passed the first two re-investigations, etc.

            But she’s one person. I didn’t mean to prop her up as an example of industry issues, more to point out that problems with being in a room full of raging hormones aren’t as much of a problem in industries where the average employee isn’t 24, which is fairly unique to web companies.

            Report comment

      • Liskantope says:

        There are two issues to worry about when contemplating skewed gender ratios in the workplace, although they are often conflated. One issue is the possibility that there might be some kind of prejudice behind the gender skew, or really any other mechanism at play other than innate interest or ability which is keeping the minority gender out. The other issue is the effect in and of itself of having less diversity in the workplace. When a man says, “I wish there were more women in the office”, in most contexts he’s probably referring to the second issue and not expressing his opinion on the first one at all.

        An overly simplistic way to look at dating dynamics when a work environment is heavily male dominated is that it’s bad for the men because they have a harder time finding someone there to date, and it’s bad for women because they have to fend off a lot more unwanted advances. This is magnified by the dating dynamic in society-at-large, where single men tend to have a harder time getting dates and single women tend to have a harder time staying single. I remember seeing a metaphor for this on Reddit where men are dying of thirst and women are drowning. Anyway, I think it’s fair to say that men and women are symmetrically disadvantaged in this situation (though when you take each particular disadvantage to its potential extreme, this symmetry becomes dubious). It also seems fair to say that in identifying the downsides of having a skewed gender ratio in the workplace, this is a major factor at least for single employees (assuming most single men are open to dating their coworkers), although it’s far from the only downside to lack of diversity.

        Report comment

    • Cord Shirt says:

      Are you familiar with Megan McArdle’s old piece on same-sex marriage? (Disclaimer: I’ve supported same-sex marriage since I read Andrew Sullivan’s “Here Comes the Groom” in The New Republic in 1989. But McArdle gave a good argument for caution.)

      She mentioned the economics concept of “the marginal case”:

      Now, economists hear this sort of argument all the time. “That’s ridiculous! I would never start working fewer hours because my taxes went up!” This ignores the fact that you may not be the marginal case. The marginal case may be some consultant who just can’t justify sacrificing valuable leisure for a new project when he’s only making 60 cents on the dollar. The result will nonetheless be the same: less economic activity….

      Marriage…looked, to those extremely smart and well-meaning welfare reformers, practically unshakeable; the idea that it could be undone by something as simple as enabling women to have children without husbands, seemed ludicrous. Its cultural underpinnings were far too firm. Why would a woman choose such a hard road? It seemed self-evident that the only unwed mothers claiming benefits would be the ones pushed there by terrible circumstance.

      This argument is compelling and logical. I would never become an unwed welfare mother, even if benefits were a great deal higher than they are now. It seems crazy to even suggest that one would bear a child out of wedlock for $567 a month. Indeed, to this day, I find the reformist side much more persuasive than the conservative side, except for one thing, which is that the conservatives turned out to be right. In fact, they turned out to be even more right than they suspected; they were predicting upticks in illegitimacy that were much more modest than what actually occurred–they expected marriage rates to suffer, not collapse.

      How did people go so badly wrong? Well, to start with, they fell into the basic fallacy that economists are so well acquainted with: they thought about themselves instead of the marginal case.

      You’re apparently not the marginal case.

      I am. I was a high-IQ kid with lots of interests, one of which was programming. One of the reasons I dropped that interest was because I didn’t find the culture congenial.

      I almost did. (Which means this was a more painful experience than if the culture had been “just an overall bad fit.”) It was actually quite a good fit in most ways. It was a better fit than “the general surrounding culture” in many ways. But.

      But there was a subtle sexism / male-oriented-ness. This is often tolerable (though still off-putting) as an adult, but as an adolescent, I decided it was more than I wanted to put up with…and that’s part of the problem: Even if I picked up the interest again now, I skipped developing it in adolescence–exactly the time when it’s most important to develop one’s talents–and I did so because of the male-oriented culture. I’m not the only one.

      (I agree that part of the problem is “uncharitable interpretations of men being awkward”–but that’s partly because most people tend to typical-mind and underestimate inferential distance, especially in adolescence. Another part of the problem is “uncharitable interpretations of teenage girls being awkward”…for the same reason.)

      SJWs want the problem to be simple and easy to solve, so they rant at male nerds as if they could “just” easily change their behavior. In fact this is a *subtle* problem that’s *difficult* to solve. Not least because the very way in which the culture is “male-oriented”/”sexist” is itself very subtle and difficult to quickly explain. Deborah Tannen in the ’90s did a good job. Her explanations are books. Because a book is what it takes. (Example: Talking from 9 to 5: Women and Men at Work, 1994.)

      Anyway…both adolescents in general, and high-IQ individuals in general, tend to be sensitive; put the two together and you get *extreme* sensitivity. And sensitive high-IQ adolescents, who really would’ve enjoyed the field and been good at it, *are* getting put off for reasons unrelated to their competence.

      One possible reaction to that situation is, “Well, that often happens, to many people considering many fields. So oh well.” Another possible reaction is to want to change it. Yet another is to agree that it’s bad while still being unsure anything we could do *would* change it.

      I agree that it does often happen. I dislike it ever happening, because I prefer for people to enter the fields they would enjoy and be good at. I also reserve the right to focus my attention in this matter especially on the field(s) that *I too would enjoy* and that people *like younger!me* are getting put off of.

      But as a non-SJW, I agree with you in that I don’t know if there’s a solution.

      The traditional feminist solution was, “Just convince men-in-general to better empathize with women-in-general.” That solution assumed that intention to empathize was all that was necessary for understanding (because it also assumed no gender differences existed), and hence that this task would be easy.

      Neither of which has turned out to be the case, it seems. So if we update that solution, it would now be, “Accomplish the difficult task of teaching men-in-general to better understand women-in-general[, so that if they want to, they know how to change their culture in ways that will actually help].” (Or for female-dominated professions, “…of teaching women-in-general to better understand men-in-general.”)

      …when it comes to the issue of *adolescents* being put off from entering a field, that might be somewhat doable. After all, it is the job of the adults in a field to try to understand and support adolescents considering entering it. (Isn’t it?)

      Perhaps we could also help “teen-girls-in-general” better understand “teen-boys-in-general” and vice versa…though that is much more difficult, as members of each group tend to (mis)interpret the others’ preferences through the lens of their own and then be horrified, and again, adolescents tend to be especially sensitive and prone to typical-minding. (See: Redpill. And the old rec.arts.sf.* thread Mark Atwood mentioned. Even adults are horrified; adolescents are worse.)

      …whatever. I care about the issue, is my point. :/

      Report comment

      • Nadja says:

        Fascinating. I agree with the point you made in your previous comment about not declaring victory too early. I was just about to respond with: well, but then what do we do about it? What would be the productive way of approaching the slight sexism that might exist? Isn’t the current feminist approach doing more harm than good? (As suggested by many others here, including Garret M. Petersen, Kendall, and the Nybbler.) But after reading your current comment, I see that’s a question you’re raising yourself: “But as a non-SJW, I agree with you in that I don’t know if there is a solution.” I was also going to emphasize that I very much care about young girls making good career choices for themselves. And then you said exactly that.

        I’ve listened to talks by many women who were successful in their technology careers. They all (I’m not exaggerating) had one thing in common. When asked the obligatory question about how it was for them to be a woman in technology, they didn’t have much to say. They were often apologetic about it. They knew they were “supposed to” care and have an opinion. But they just never paid enough attention to how their gender affected their career. What if this sort of approach makes success easier? Maybe if sexism is so subtle that one can’t quite put a finger on it, then the best thing is for an individual to fight back the way you’d fight back any other sort of bullying/mistreatment. Stand up for yourself. Keep doing good work. Keep learning. People are always going to pick on others. I don’t think “nerdy” guys have it any better than women.

        I really like your point about marginal cases. I’m very curious to learn more about your specific marginal case. Where did you grow up? What was it about the tech culture that turned you off? What did you end up doing instead?

        My story is very different from yours, but there are some parallels, so I’d be curious to compare notes.

        Report comment

        • Cord Shirt says:

          But they just never paid enough attention to how their gender affected their career. What if this sort of approach makes success easier?

          Yes, Barbara Kerr came to a similar conclusion in Smart Girls, Gifted Women. (This was a multiple case study that started out focusing on the women who’d been in her Sputnik-era gifted program, and then expanded to, IIRC, National Merit Finalists and so on.)

          But…people differ in personality and “social perception,” such that some people will find this much easier than others.

          Also, “It’s all in your head / you’re just oversensitive” is a favorite bullying technique.

          *Also*, a group of children selected for high IQ is likely to have a higher percentage of members who are “good targets” for that kind of bullying. This shows up in the research on such children–when bullied they often object more to the idea that “any human could treat another human that way,” than to the pain of physical attacks or insults directed at them personally.

          IOW, the issue of “standing up to bullies” is much more complex for some high-IQ kids than it might appear on the surface. Or than it *is* for most typical kids. As described here:

          Ten-year-old Greg Barnes was acknowledged by school personnel as highly gifted. His scholastic achievement test scores placed him in the 99.9th percentile, as did his score on the Stanford-Binet Intelligence Scale. On this particular day, when he returned home from school, Mrs. Barnes knew immediately from his despondent expression that the day had been less than ideal.

          “Something wrong?” she probed gently.

          “Yeah,” he said in a tone of thorough disgust, “I got into trouble. We’ll have to see the principal tomorrow.”

          “What for?”

          “I got into a fight with Joe and beat him up.”

          Mrs. Barnes was shocked. Greg was not an aggressive child. He had never reported such an incident before. In fact, he was an extraordinarily sensitive boy who genuinely cared about other people.

          “What happened?” she inquired further.

          Greg explained that he and Joe had exchanged insults during music class. Both boys, Greg insisted, were at fault. Later, Joe had cornered Greg by the lockers, taunting him, threatening to beat him up, and egging Greg on to fight. Greg responded by punching Joe, who punched Greg back. When the teacher came onto the scene, Joe was crying while Greg continued to rain punches upon him.

          “Well, it sounds like you stood up for yourself…” began Mrs. Barnes. She was surprised at Greg’s immediate and heated denial.

          “No, Mom – it wasn’t that simple.”

          “But wasn’t he threatening to beat you up?”

          “No, Mom! You don’t understand!”

          Greg was getting visibly more upset as Mrs. Barnes attempted to convey that she was not being judgmental. Unable to comprehend why her efforts to convey caring and understanding were being met with mounting frustration, Mrs. Barnes decided to defuse the issue.

          “OK. Why don’t you write down what happened and explain how you feel about it. Obviously you were there and you know why it happened better than I do.”

          Greg willingly took a seat at the typewriter and laboriously typed out his story and explanation. An hour and a half later, he handed the pages to his mother: “It all began in third grade…” started the first paragraph. Greg went on to describe in careful detail how he and Joe had met and embarked upon a rocky friendship. At certain times, Joe seemed to want to be friends. At other times, Joe refused to allow Greg to participate in ongoing playground activities. Greg admitted to sometimes levelling “insults” at Joe in retaliation for these playground rejections.

          Greg listed incidents from 3rd and 4th grades as well as the 5th grade incident that precipitated the immediate problem. For each incident, he detailed each child’s behaviours with painful accuracy in an effort to render an objective view of what had happened. Greg’s outburst was, according to him, not only a response to the day’s happenings, but a reaction to the entire pattern of incidents composing their relationship over the past two years. The argument of the day was simply “the straw that broke the camel’s back”.

          The next day, Joe, too, wrote out his version of the fight. He wrote simply, “Greg hit me and then I hit him back and he kept hitting me.”

          Greg and Joe had participated in the same fight. Yet, Greg was fighting over a broader and more complex issue than was Joe.

          This is the type of thing that it’s common to write off as “oversensitivity,” but labeling it that doesn’t lead to any reliable methods of addressing it.

          Morelock points out that such children have “an emotionally infused need to understand the truth of their situation and to communicate that truth to others…. They live life by analysing it step by step while emotionally responding to that analysis.” That…really seems like the same thing that’s driving their high intellectual achievement and desire to share their discoveries with others. If so, it’s something we should allow them to keep.

          Morelock’s suggestion:

          “Greg must be helped to see, of course, that others do not always perceive experiences with such detailed clarity or infused with such emotional intensity and significance especially other children his age. Nevertheless, this should not be conveyed as a defect on his part, such as “oversensitivity” or a tendency to “carry a grudge.” Rather, he needs first to have his view of reality validated, and second, to be helped to find ways of building bridges with other children and not faulting them because of their differing level of awareness…. At the same time, children like Joe need to be made aware of the role they play in instigating unnecessary ill feelings.”

          …strikes me as more likely to actually work. But that technique isn’t something I’ve personally experienced.

          So, back to your point…my friends (some of whom have Asperger’s) and I have kicked around the idea that Asperger’s might sometimes be more of a help than a hindrance for a high-IQ person. Might keep you from even noticing some of the subtle social stuff. As Meredith Patterson described. (But probably only for a specific, mild, degree of Asperger’s–and Asperger’s comes with other problems.)

          More in another comment.

          Report comment

        • Cord Shirt says:

          I’m very curious to learn more about your specific marginal case. Where did you grow up? […] What did you end up doing instead?

          First of all, what about you? Where did you grow up, and what’s your story?

          Anyway, to answer your questions…I keep trying to summarize quickly, and it keeps…turning out long. Sorry about that.

          I grew up in Yankeedom, USA and attended non religious, “progressive-ish” private schools. (Where, as I said on the other thread, I never learned any academic competitions existed. Maybe competition was too “square” for my schools.) I took the Actual Real MBTI as part of a study I was in :waves to Universal Set: and came out as INTJ (but I think INTP fits better).

          I liked playing around with computers from the time my parents first got one when I was a preteen. Wrote a bunch of stuff in ROM BASIC (it’s what I had)–so I couldn’t save anything, always had to print it out and then type it in again when I came back later. Learned about hardware from reading PC Magazine–it’s what I had access to (I didn’t know BBSes existed, and we didn’t have a modem then anyway). Started homebuilding my own desktops as soon as I could.

          After I got online as a teen (in the gopher era), I…basically just the usual: spent all night downloading Linux onto dozens of floppy disks, spent years trying to make Wine work (for games), had one computer with an uptime measured in years…etc. I wrote a fair number of little fixes and tools for myself, but for myself they stayed. It never really occurred to me to even *try to figure out* how to share them with others, let alone to actually do it. At that stage, I think the reason for this was mostly my school experiences–I had learned very thoroughly that I was a CHILD, and children are NOTHING, and WHO DO YOU THINK YOU ARE TO EVEN TRY TO TALK WITH AN ADULT–etc.

          (Another aspect, though, was…”perfectionism” is the wrong word. I’d call it, “That thing where smart girls are held to a higher standard of i-dotting and t-crossing than smart boys are.” It’s like the book I once read for novelists, about character archetypes, which listed two archetypes of a smart man–“anal scientist” and “absent-minded professor”–but only one archetype of a smart woman: “anal librarian.” An absent-minded *woman* is just an “airhead”; a *smart* girl is expected to also be tidy/polished. If there *is* an elementary-school “leak in the pipeline” as Scott has hypothesized…then this might be one of the contributors. Because IMO the standard of “polish” to which smart little girls are often held is high enough to create habits that discourage sharing. If you have two equally-lazy youths, who both just kludge together an ugly fix…the guy is more likely to go ahead and share it; the girl is more likely to…not even have it occur to her that she could, because “*of course* anything you actually *share* needs to be more polished than that.” KWIM? Or is it different where you’re from?)

          I went to boarding prep school for 9th grade, which experience taught everyone involved that I really belonged in college, so the next year that’s where I went. There and online, I encountered geek/hacker and physics cultures…with aspects I ultimately decided I’d rather not put up with (more on that below).

          This didn’t convince me to give up my interests immediately; it only meant I continued them in isolation instead of trying to interact with most of the others who seemed to share them. Flaw in this plan: When I did meet someone I wanted to befriend and even date…well, we all know what “the group’s” opinion tends to be of someone who “was brought there by her boyfriend.” It often doesn’t occur to people that maybe one of the reasons she’s dating their friend is *because he shares her interests*. Yes, some of these “girlfriends” may be annoyingly “closeted” geeks, coming into the group after having been reluctant to give up their higher status; but others will be more like I was, coming in from…complete isolation.

          It’s around this point (age 17 or so) that I started focusing more on other interests. (So come to think of it, I guess I didn’t give up quite as young as I half-remembered above!) I left college one semester short of a degree and started a couple small businesses. (Years later I went back to pick up those last few credits so as to have A Degree Any Degree in case I ever need it.) My businesses are in fields I don’t have much natural interest in, but…one of them I could enter easily (I don’t envy the men in that one. Do try to support them though), the other I considered necessary/important despite not initially enjoying it. I’ve found things to enjoy about these fields; I’ve even, rather oddly for an INTwhatever, learned to enjoy the fact that the items I create/develop make my customers happy. These days the only vaguely coding-related thing I do is the occasional Sims mod. (Which I don’t share. I’m too lazy to make them “minimally polished enough” to “deserve sharing.”)

          You could say I fit Barbara Kerr’s claim that high-IQ women often don’t “achieve on paper” because they’re “too well-adjusted”: I’ve adjusted to the world in which I happen to be, learned to enjoy things my personality didn’t initially pull me toward, etc. (Which is kinda sad considering that when I first read her work, I was sure I never would.)

          What was it about the tech culture that turned you off?

          I already mentioned a minor roadblock I encountered on the hobby end of things. And, again, Deborah Tannen’s Talking from 9 to 5 described in detail a lot of the issues I saw.

          To describe the general issue concisely, I’m going to have to oversimplify to an extreme, offensive degree–so please keep in mind the following *is* an *extreme* oversimplification:

          The culture had been created by and for (mostly straight) men, including the aspects of maleness that women often (a) don’t know about, (b) don’t understand, and (c) especially in adolescence tend to be horrified to discover. The culture just had that stuff all hanging out there–that was off-putting aspect #1.

          Off-putting aspect #2 was that since the culture was created by and for men, its traditions and usages treated the subject–“you,” the member of the culture–as a man with (straight-)male-typical sexual proclivities and assumptions and perceptions. IOW, from the teen girl POV, this “horrifying” attitude isn’t just something you discover exists, it’s also something you too are being assumed to hold.

          And so–off-putting aspect #3–though women did appear in the culture, it was as (sorry to use the old term but) objects rather than subjects. “Foreign creatures” rather than “you.” (nb: To summarize it as “foreign creatures” is an extreme oversimplification, but it’s the only way to keep the sentence short enough not to distract from the larger point.)

          All three of these aspects show up in, for example, the vignette in Reviving Ophelia where the physics teacher begins an explanation with (paraphrase since I’m too lazy to look it up), “So, imagine you’re speeding in this cool car with this hot girl and…”

          “Cheesecake lying around all over the place” was another manifestation, of course.

          I think it’s maybe similar to the complaints men often voice about romance novels. I think many teenaged boys might have to love an interest much more than “the equivalent” teenaged girl would in order for the boy to be willing to continue with the interest if its culture was “like living in a romance novel.”

          (And, yes, this would be a problem with *any* mostly-male culture, *not* just tech. Tech is not “worse” IMO.)

          BTW, a problem with pointing out this type of issue to members of another group that doesn’t have it: There’s a human tendency to go, “I don’t have this issue” -> “Having this issue seems really weird” -> “Having this issue means you’re weak” -> “OtherGroup is weak.” No, this is one specific issue and that’s all it is. There’s not just one single axis of sensitivity, and wanting consideration on a single issue doesn’t make someone *generally* “weak” or “oversensitive.” If I could have *one* way in which members of a SomeGroup-dominated culture could be asked to pull up their socks and try to overcome their natural human flaws, it would be this one: I’d have them overcome the natural tendency to overgeneralize “We’ve been asked to be sensitive on this *one* issue” to “OtherGroup is a bunch of weaklings and we need to treat them with kid gloves about *everything*.”

          …uh, [/soapbox]

          There were other issues…but it’s late, and I’m tired. Sorry to repeat myself again but Deborah Tannen did a good job describing them. I’ll just say that in a way, more direct issues, “deliberate ‘attacks'” or etc., were easier for me to deal with than this subtle “male-oriented-ness.” (Also why I like nerds: Nerds tend to be direct–so if there’s a problem, you can actually address it. Except on this one issue, unfortunately.)

          Reading Meredith Patterson’s other essay touching on this topic makes me both nostalgic and sad. Yes, I recognize that culture; yes, in many ways it fits my instincts, my natural personality, better than any other…except. Except for the things above that I decided I’d rather not deal with. Having left that culture behind (somewhat. I mean, here we are on the periphery of it), having found other things to enjoy and to support myself with, other people to exchange ideas with and fit in with…has changed me, and in some ways the change has been away from my natural personality. Well, that’s life, I guess.

          BTW. Today my first seeds of the year came up. Future stock of starts: growing. 🙂 …and now I’m remembering a certain “Lone Gunmen” flashback episode. Kid!me had similar ideas to kid!Langly; today, I grow things. I suppose many people here are still with Langly 100%; I’m…not so sure. There’s good and bad in both.

          Sorry again for the length.

          What’s your story?

          Report comment

          • Nadja says:

            @ Cord Shirt

            I have just ordered “Smart Girls, Gifted Women”. Darn, no Kindle version! I’m not very good at finishing paper books these days. But, I also got “Talking from 9 to 5”, which – thankfully – is available in audiobook!

            I really like your discussion of bullying. I realized that what I wrote came out wrong. Whereas I think it’s reasonable to expect adults to learn how to deal with jerks (not in all environments, but definitely in most white-collar office environments), I would never, ever, ever tell a child to “just stand up for themselves.” I’m very grateful for researchers who are trying to figure out how to help bullied kids. Unfortunately, for many, learning these techniques is a necessity. At the same time, I strongly disagree with people who say “kids *need to* learn how to deal with bullies in school, because, well, one has to deal with bullies later in life, too.” An adult has more freedom, and all sorts of resources and strengths a child doesn’t have. There’s just no comparison.

            See, I’m glad I asked about your background. Your story shows me that I really just have no idea. First of all, everyone really is different. Secondly, there does seem to be something weird/wrong about the culture in the US in terms of how it affects adolescent girls. And, finally, wow, see, initially I was going to tell you: “hey, it’s not too late to become a programmer; in fact, I can help you get started, if you’re interested.” But now I think that this wouldn’t be something you’d be interested in. Given that you are incredibly gifted, what you may have missed out on because of the male-centeredness problem is not just the opportunity to be some programmer somewhere. Instead, you may have actually missed out on becoming one of those tech people who build companies, who create amazing products, and who shape the world. I now understand so much better where you’re coming from.

            But, I am really happy to learn that you’ve been able to build your own businesses anyway. It seems like you did very well for yourself. I don’t know enough to see why you’d think of yourself as not having achieved enough on paper, but – hey – is it ever too late to achieve more? =)

            Some other stuff in your comments: the point about girls and the sharing of polished products rang very true to me. I had to get over this sort of attitude myself. It’s probably still holding me back a bit. I’d never thought of it as a “girl” thing, but I can see that it very well may be.

            “Reviving Ophelia” – another book I haven’t read. =) But I think I get the point. I looked through it on Google books just now. It mentions, among other things, that girls often suffer from a lack of female role models. Come to think of it, most of my childhood role models were men, but I was fortunate in that it never bothered me. All the books I loved as a child had male heroes. My best friends were guys. I was on a math/science track in high school, so most of my classmates were guys, too. And it never felt strange. I never felt different. It was all just so natural. But then… Then, I came to study at an American university, and things stopped being easy. I talked to guys, the way I was used to talking to guys, and it wasn’t working any more. Now they all thought I was hitting on them! Some actually told me they could never be close friends with a girl! (Fortunately, I was able to find a really cool, reaaaally smart girl to be best friends with.) So, anyway, that shift: was it because we were all older or because there *is* something about American culture that makes women into “foreign creatures” in the eyes of men?

            And while we’re on American culture, quite unrelatedly, this whole jocks versus nerds thing… Really? In my high school, if you wanted to be cool, you wore black sweaters, worked on math puzzles on paper napkins, and carried a copy of “The Drunken Boat” under your arm. Ahh, I miss high school. We were all such posers. (If you’ve seen “To Rome with Love”, think the Ellen Page character.)

            Anyway, it’s getting late. I’m going to be looking out for your comments around here on SSC. I’m looking forward to chatting more. That is, of course, unless the reign of terror gets us first. =)

            Report comment

          • NN says:

            Since we’re trading anecdotes, it’s probably worth bringing up that my mother is a computer programmer. It wasn’t her first choice; she wanted to be a veterinarian but switched majors when, in her words, she realized that she would have to spend years studying creepy crawly animals before getting to the cute ones. So at least one “marginal case” woman was pushed into the tech industry by external factors. But then the tech industry was different in all kinds of ways back in the early 1980s when she started working in the field. Wikipedia says that in in 1984 37.1% of Computer Science degrees from American colleges were awarded to women, which is probably why my mother has never once suggested that there was anything unusual about her ending up in the tech industry.

            I think it’s maybe similar to the complaints men often voice about romance novels. I think many teenaged boys might have to love an interest much more than “the equivalent” teenaged girl would in order for the boy to be willing to continue with the interest if its culture was “like living in a romance novel.”

            Having spent an awful lot of time around fan fiction communities (particularly Livejournal fan fiction communities back when people still used Livejournal), I can’t help but notice that practically all of the “male dominated” aspects of tech culture that you mention are present in a gender-flipped form to a much greater degree in the female dominated aspects of fan fiction culture. Yet, as far as I can tell, what we generally see is not males being driven away from being interested in fan fiction entirely. Instead we generally see them forming their own fan fiction communities (usually significantly smaller and less productive than the female dominated ones, but they exist) that cater more to, for lack of a better term, male-typical tastes.

            We see this most famously with the Bronies, but HPMOR also sort of counts. For whatever reason Eliezer wasn’t driven away from Harry Potter fan fiction by the gigantic amount of Harry Potter slash fiction and artwork floating around the internet, nor by the fact that most fan fiction communities assume (not without some justification) that everyone involved is female, to the point that one LJ fan fiction community I came across back in the day was formatted to display each post’s comment count as “X fangirls” instead of “X comments.”

            Obviously, fan fiction has a far lower barrier to entry than programming/hacking, so this may not be the best comparison, but I still think it’s worth looking into.

            The infamously poor quality of many works of fan fiction would also seem to be a data point against your theory that “don’t share your work until it’s really, really polished” is a typical female trait.

            Report comment

      • NN says:

        You have a point about marginal cases, but the question then becomes why didn’t a similar effect keep women out of medicine, biology, psychology, much of the humanities, and other male dominated fields which rapidly achieved gender parity or even female dominance over the past few decades? I have a really hard time imagining that medicine has a more “female friendly” (or friendly-to-anyone-at-all) climate than computer programming.

        I don’t doubt that various forms of bias and sexism exist, but I have yet to see any convincing evidence that they are major drivers of gender disparities (at least when it comes to driving adolescents and adults out of the field; I’m still open to ideas about early childhood socialization).

        Especially given that the gender gap in computer science has actually greatly increased over the last few decades:

        In the United States, the proportion of women represented in undergraduate computer science education and the white-collar information technology workforce peaked in the mid-1980s, and has declined ever since. In 1984, 37.1% of Computer Science degrees were awarded to women; the percentage dropped to 29.9% in 1989-1990, and 26.7% in 1997-1998. Figures from the Computing Research Association Taulbee Survey indicate that fewer than 12% of Computer Science bachelor’s degrees were awarded to women at U.S. PhD-granting institutions in 2010-11.

        https://en.wikipedia.org/wiki/Women_in_computing

        If sexism is to blame for the programming gender gap, then someone has to explain why programming culture became far more sexist over a period of time when the rest of society was becoming less sexist.

        Report comment

        • Cord Shirt says:

          I didn’t say it’s “a major driver of gender disparities.” I said it’s a thing that happens that I’d like to reduce the frequency of.

          I suspect that a sincere but mistaken belief that this problem was *the* “major driver of gender [and racial] disparities” is what inspired policies like affirmative action which have not worked as well as people expected them to, and which have also had unexpected negative effects.

          It’s tempting to respond to those negative effects by denying that this situation even happens at all. I’m just saying it does.

          To put it another way: I did not go around demanding that hackers change their culture for my sake. Instead, I quietly left. These days, OTOH, we have a situation where some women (at least some of whom may actually be interested in STEM) loudly demand that “nerdy men” change for their sake. And we have some nerdy men resenting that and saying or wanting to say, “If you can’t stand the heat, get out of the kitchen.” My point is just that if you say “If you can’t stand the heat, get out of the kitchen”…some will, in fact, get out of the kitchen. And that’s not always because of simple lack of talent and/or interest; sometimes, it’s because of “nothing more than” culture clash.

          As to why it’s played out differently in different fields: I’d say it’s because each field had its own culture and its own personality type that was most attracted to it. Some fields had a culture that more women found congenial, and some fields tended to attract personality types that were common enough among women to form a critical mass who then influenced the culture. Other fields did not.

          BTW…don’t assume that having made this mental connection means I somehow agree with this author on everything but…the 1915 novel Herland by Charlotte Perkins Gilman also expressed the opinion that different male professional cultures were more or less…welcoming…to women. Her three male characters are a sociologist (the narrator), an engineer (the “worst” according to her), and a man who was “born to be a poet or a botanist” but his family convinced him to be a doctor and who loved (biological) science (the “best” according to her). Interesting how she seems to have accurately predicted that smart women would find literature, biology and medicine more welcoming, and engineering less so.

          Report comment

      • houseboatonstyx says:

        @ Cord Shirt
        Not least because the very way in which the culture is “male-oriented”/”sexist” is itself very subtle and difficult to quickly explain. Deborah Tannen in the ’90s did a good job. Her explanations are books. Because a book is what it takes. (Example: Talking from 9 to 5: Women and Men at Work, 1994.)

        Suzette Hayden Elgin’s books were around that same time, and also very good, with some methods individuals can use to deal with this (or to use ‘verbal self-defense’ in general).

        Report comment

  14. perpippity says:

    “Every time I say I’m nervous about the institutionalized social justice movement, people tell me that I’m crazy, that I’m just sexist and privileged, and that feminism is merely the belief that women are people so any discomfort with it is totally beyond the pale.”

    Feminism is so NOT “merely the belief” yada yada yada. In terms of Louis C.K.’s shtick about boys and girls (below), feminism is the project to make people believe that there aren’t any bodies buried at all, and that the only harm out there is the nasty hurricane of harm caused by boys. Feminism is a mean-girls clique, a deranged lynch mob out after your emotions, society gone rancid. It’s very Soviet in terms of the nasty mental and emotional shit that goes on.

    https://www.youtube.com/watch?v=RTrCBcrFMCI

    Report comment

    • Dahlen says:

      Feminism is a mean-girls clique, a deranged lynch mob out after your emotions, society gone rancid. It’s very Soviet in terms of the nasty mental and emotional shit that goes on.

      Can we please not have a two minutes hate on every one of these articles?…

      Piggybacking on a discussion in the latest open thread:

      It would be interesting to have more serious feminists/social justice people and/or Noam Chomsky socialist-anarchist types; I’m not sure what one could do to attract them.

      Well. I wonder what.

      Don’t get me wrong, what’s happening with science journalism downstream from the actual study is unmistakeably shitty and in no way novel, and I’ll grant that it’s because of feminism that this is the new direction in which the headlines are sensationalised, but there’s a strong case to be made that we haven’t hit the pinnacle of gender harmony a decade or several ago and that there are still changes that could roughly be considered “feminist” to be made that would help rather than hinder society (even if they’re not likely to come from feminists of the SJW persuasion). In case you find that statement highly objectionable, on top of all that, this community is, however reluctantly, adjacent to a blogosphere that literally collects all the most patriarchal aspects of all historical societies ever into one big idea system of misogyny on steroids, and what they consider feminism is anything onwards from the 19th century. In the light of that fact, maybe it wouldn’t be wise to contribute to conditioning in the reader an aversive reaction to the word ‘feminism’ and anything it could stand for?

      There’s a baby in that bathwater, that’s all I’m saying.

      Report comment

      • The Nybbler says:

        > In case you find that statement highly objectionable, on top of all that, this community is, however reluctantly, adjacent to a blogosphere that literally collects all the most patriarchal aspects of all historical societies ever into one big idea system of misogyny on steroids

        Really, guilt by association? I’m not sure what “adjacent” means in this context, but I think most measures of “adjacent” which would put SSC adjacent to them would also put it adjacent to the SJW blogosphere. Neither “adjacency” imposes any duty on SSC’s commentariat, or on Scott for that matter.

        As for feminism, I’m afraid the baby may have drowned. I’d like to see the name reclaimed by non-SJWs, but it doesn’t seem likely.

        Report comment

      • Zorgon says:

        You had my fervent agreement right up until this bit:

        >this community is, however reluctantly, adjacent to a blogosphere that literally collects all the most patriarchal aspects of all historical societies ever into one big idea system of misogyny on steroids

        Is this the standard now? “Adjacent”? BRILLIANT! I can get on with pointing out how feministing and Alas and all that lot are adjacent to Stormfront.

        Report comment

      • The Smoke says:

        I have my aversion to feminism to the most part from reading feminist blogs / media. There is of course the really bad stuff, which is still pretty much socially accepted, and then there are a lot of more moderate feminists.
        What I find interesting about the latter is that most of them are thoughtful and intelligent persons EXCEPT that when talking about gender they regard ‘feminist theory’ as facts and based on that sometimes even feel justified to express really sexist views themselves (E.g. Cathy O’Neil, Piper Harron, I don’t know how to hyperlink, else I would be more specific).

        Pointing out specific problems without drawing overblown conclusions, as done here, actually helps everyone getting a better understanding. In particular the civil tone in the comments here helps (me) getting a more nuanced grasp on what exactly my problems with feminism are, and so being less angry at feminists in general.

        Report comment

      • Dahlen says:

        @ The Nybbler and Zorgon:

        Whoa. Chill. What I meant by “adjacent” was that Jim, Heartiste et al are a few (< 5) links away from the SSC blogroll and the pool from which commenters are drawn includes a handful who also comment over there, which means said idea system is not some distant nemesis but something that occasionally leaves traces here.

        Report comment

    • Muga Sofer says:

      I’m pretty sure this violates the comment policy.

      It’s definitely unnecessarily rude, regardless. You’re well aware that there are self-identified feminists here and that your description is incredibly unfair to them.

      Report comment

      • perpippity says:

        I’m well aware that there are self-identified feminists here, but I don’t care whether their feelings are hurt. What they need to do is not care what I think.

        Report comment

    • Feminism is literally millions of people, many of whom have no idea what others are doing and saying under the banner of feminism.

      I challenge the idea that we can sum up feminism any more easily than we can sum up Christianity.

      Report comment

      • John Schilling says:

        Is it even possible in this media-saturated age to “have no idea” that lots of feminists are conducting and promoting dubious studies to prove the pervasiveness of sexist discrimination by privileged men who need to be put in their place?

        Meh. If Christianity were a million Crusaders slaughtering the Infidel at the Pope’s command, and a billion people who showed up at church for an hour every Sunday to proclaim a “religion of peace” and obliviously ask, “What’s a Crusade?”, I’d have no trouble summing up Christianity as a religion of murderers and idiots. I can hope that we might find a more charitable description of feminism, but it won’t come from the supposed ignorance of the majority of feminists.

        Report comment

        • Nita says:

          Is it even possible in this media-saturated age to “have no idea” that lots of feminists are conducting and promoting dubious studies

          Yes, absolutely. Most people can’t tell a dubious study from a good study, or, for that matter, from vaguely scientific-sounding crackpottery. That doesn’t make them “idiots”, just uneducated and confused.

          And the ongoing comparisons between posting social criticism online and killing people (“lynch mob”, “slaughtering the Infidel”) are unhelpful, no matter how misguided the criticism might be.

          Report comment

        • Anonymous says:

          Well, Christianity has the advantage of having an unchanging source document, and an official hierarchy of priests interpreting the rules, which helps to keep everyone not living in a cave on the same page. Feminism has neither, it is just a collection of roughly adjacent ideological points that nobody has any control over.

          Report comment

    • houseboatonstyx says:

      “Every time I say I’m nervous about the institutionalized social justice movement, people tell me that I’m crazy, that I’m just sexist and privileged, and that feminism is merely the belief that women are people so any discomfort with it is totally beyond the pale.”

      This 1970s Feminist wishes that people would not identify ‘feminism’ with any kind of SJW stuff. (Particularly SJW-style retreat to a trivial motte like “the belief that women are people”.)

      Here’s my definition of ‘feminist’ (in a useful zone between trivial motte and the far edge of bailey). “Important practical goals of us 1970s feminists aren’t yet done, and are high priority.”

      Report comment

      • John Schilling says:

        Could you elaborate on the goals that were important and practical in the 1970s and have not been substantially achieved in the 2010s?

        Report comment

        • Adam says:

          Paid family leave seems like the most obvious. Also the general sense (that may or may not be accurate, but is definitely there) that reproductive rights gains are fragile and could disappear any minute if they stop fighting.

          Plus, there are at least a few feminists who legitimately care about international issues, i.e. genital mutilation (even seen quite a few who are strongly anti-male circumcision), sex trafficking, honor killings, the few countries out there where women can’t drive, go to college, whatever else is still left.

          Report comment

          • John Schilling says:

            On the second point, fair enough, I should have specified “in the industrialized west”.

            As for reproductive rights, I think I am both dispassionate and reasonably well-informed about this and I see the odds of women in the United States being more than modestly inconvenienced in their pursuit of many forms of birth control up to and including first-trimester abortion as being negligible. And mostly subsumed into the remote general possibility of a civil war or other complete sociopolitical upset, which would basically negate any present activism.

            The illusory threat/promise of a return to the 1950s is deployed by both sides to motivate their base to fight ever harder for, well, whatever the issue du jour is even if it isn’t abortion or birth control. If the goal is reproductive rights, that one has in fact been accomplished even if you fear it might be unaccomplished in the future. If the goal is eliminating the fear, that cannot realistically be accomplished by so long as there are feminists stoking the fear.

            Report comment

          • Adam says:

            I agree that the chances of it actually happening are basically nil, but it clearly does motivate them and I get the impression the goal at this point is eliminate all politicians trying to remove reproductive rights, and those politicians certainly do exist so that isn’t a finished goal. I’d even say that seems to be the single issue most driving the recent spat of ‘stop the infighting, Bernie and Hillary supporters’ memes and articles I keep seeing, the conviction that this election is special above all in recent memory because of anticipated supreme court appointees who will overturn Roe v. Wade if a Republican is president.

            Which is really weird, by the way. I always thought when I was younger that single-issue abortion voters were universally Republicans.

            Report comment

          • houseboatonstyx says:

            @ Adam
            I’d even say that seems to be the single issue most driving the recent spat of ‘stop the infighting, Bernie and Hillary supporters’ memes and articles I keep seeing

            Clintonista here. I’m glad to hear pushback against infighting is starting. Apologies for ad hominem, but I model Sanders supporters (like Nader supporters in 2000) as being motivated by head in the clouds stuff, and by negative personal emotions toward celebrities etc. Abortion/RR cuts through that level: it’s real to Democrats who are otherwise not so pragmatic.

            Report comment

          • John Schilling says:

            I get the impression the goal at this point is eliminate all politicians trying to remove reproductive rights

            Which, as a goal, is right up there with removing all terrorists or all Nazis. We’d still be fighting WWII in the streets of Germany if we went with that last one. These are again poor metrics if your real terminal goal is to eliminate the threat, and a recipe for fanaticism if you make them your terminal goal.

            And I’ve been hearing literally since 1973 that one more conservative Supreme Court justice, one Republican president in office at the wrong time, means an end to legal abortion and is thus the One True Cause for every woman everywhere. When do we get to call bullshit on that one?

            Report comment

          • houseboatonstyx says:

            @ John Schilling
            And I’ve been hearing literally since 1973 that one more conservative Supreme Court justice, one Republican president in office at the wrong time, means an end to legal abortion

            The Supreme Court likes its own precedents, and politicians on both sides like the 3-letter label Roe that fits well on placards and in headlines. But the fact is that access to abortion is being chipped away, seriously and steadily, by restrictions that are subject to review by several levels of courts … and a SC Justice serves for decades.

            How attacking PP would fare in a 2017 SC we don’t know, but a law can be thrown out on problems in its fine print and President and Congress can change before it gets re-worked, and attacking PP* is something that Republicans are trying now (which is another reason to beware of a Republican President).

            (*and the actual sources of PP’s abortion funding)

            Report comment

          • houseboatonstyx says:

            @ John Schilling
            I see the odds of women in the United States being more than modestly inconvenienced in their pursuit of many forms of birth control up to and including first-trimester abortion as being negligible.

            'Modestly inconvenienced' and 'neglible' are your opinions. On the 'testimonials' thread, Nita and I posted numbers, ie facts. Here are links to some more:

            http://www.cyclebeads.com/blog/815/why-arent-all-women-using-contraception?-

            http://www.guttmacher.org/pubs/FB-Unintended-Pregnancy-US.html

            http://www.guttmacher.org/pubs/win/contraceptive-needs-2013.pdf

            Report comment

          • Mark Atwood says:

            ‘Modestly inconvenienced’ and ‘neglible’ are your opinions.

            I agree with John Schilling about the fact that having to pay $8 for a well defined product/service is only a negligible modest inconvenience. However, as you correctly point out, backed by a long list of citations, that it does still seem to be beyond the planning ability and self-control abilities of entirely too many young fertile women. (Why that may be, I could not possibly guess.)

            However, I do support fully gratis BC treatments, for the simple fact that the kind of people for whom such a negligible modest inconvenience is so overwhelmingly burdensome are the LAST people who should ever breed.

            I would go a step further, and say that as a society, we should flat give a few hundred dollars in cash to any fertile person, age 30 or below, who asks for a gratis permanent sterilization.

            Report comment

          • houseboatonstyx says:

            @ Mark Atwood
            I do support fully gratis BC treatments, for the simple fact that the kind of people for whom such a negligible modest inconvenience is so overwhelmingly burdensome are the LAST people who should ever breed.

            I would go a step further, and say that as a society, we should flat give a few hundred dollars in cash to any fertile person, age 30 or below, who asks for a gratis permanent sterilization.

            These are actions so worth agreeing on, that y’all’s opinions and motives seem hardly worth arguing about. 😉

            This might be elegant reasoning against the argument that the ‘kind of woman who needs help’, doesn’t deserve it.

            (But for the record:
            1. Temporay sterilization would be less scandalous than permanent, and
            2. ‘Modest/trivial’ is different for different budgets. To take a flat cost that all budgets must pay, and pronounce it ‘modest’ because it would be modest on the budgets you’re familiar with — almost deserves the p-word. Even if and when all US states stop requiring physicians’ prescriptions including pelvic exam etc for BC pills (so far not quite three). Not to mention, for abortion, the inconvenience of travel to a far clinic, getting ultra-sounds, counseling, etc etc, often needing several trips.)

            Report comment

          • John Schilling says:

            To take a flat cost that all budgets must pay, and pronounce it ‘modest’ because it would be modest on the budgets you’re familiar with — almost deserves the p-word

            $9/month is modest on pretty much any budget that doesn’t involve starving or freezing to death and thus rendering the whole issue moot. I don’t think it is particularly privileged to say that if the price for having a full and safe sex life is a modest fraction of the price for having any life at all, this is a problem that I’m not going to worry about any time soon.

            But I don’t think there would really be enough controversy to matter over a plan to provide a fairly broad range of free birth control options to any adult woman who wants them. One thing that is controversial enough (in the US) to almost certainly scuttle any such proposal are government-funded third-trimester abortions. Another would likely be having public schools or similar government institutions providing contraception directly to minor children.

            If you want free birth control for adult women, you’ll need to decouple it from late-term abortion, probably from birth control for schoolgirls, and maybe from abortion entirely for a generation or so. Not abandoning those fights, just separating them. Since this is exactly the opposite of Planned Parenthood’s strategy, and PP is the Progressive vanguard on all related fronts, I’m guessing universal free contraception is going to have a rough time in the United States for a while longer.

            Report comment

        • houseboatonstyx says:

          @ John Schilling

          Disambiguation:
          a) practical/possible to accomplish (given X date/resources)
          b) of practical use if accomplished

          Small example, professional dress code. In the 70s we wanted to wear b) practical-useful-comfortable stuff instead of high heels and fancy hairstyles and make-up; we got that allowed in a lot of jobs, but not yet a) everywhere; 1970s goal not yet accomplished.

          Bigger 1970s b)outcomes, not yet a) (and perhaps impossible ever) accomplished: 51/49% lawmakers, CEOs, etc.

          Referencing Ernestine for tautology, and this clip shows where journalism was then too.

          Report comment

          • John Schilling says:

            Is “51/49% lawmakers, CEOs, etc”, a terminal goal, or a metric for a goal?

            It is a poor metric for even the broadest conception of equality of rights or opportunity. Men and women are not interchangeable, cannot be culturally programmed to be interchangeable, and the residual differences will drive >>1% spreads in any profession or avocation.

            It is a poor metric even for ensuring that men’s and women’s interests are represented at the highest levels of power, because the mechanism that gets you 51/49 instead of what the natural ratio is unlikely to select for a sincere desire to fairly represent gender-linked interests.

            If it is a terminal goal, I genuinely have difficulty understanding why people would value it, and that doesn’t happen often. As a metric, it will lead to increasing fanaticism in pursuit of increasingly marginal gains.

            Report comment

          • Aap says:

            It’s especially problematic when an imbalance in favor of women is apparently not a problem (see gender of college graduates). ‘At least 50% is not equality’, it is dominance.

            Report comment

      • Sandy Beach says:

        So what you’re saying is they’re not true Scotsmen? Sorry, Scotspeople.

        Report comment

      • Viliam says:

        “Important practical goals of us 1970s feminists aren’t yet done, and are high priority.”

        In my opinion, most of those goals aren’t achieved in the developing countries… and the problem is that in the current political climate it is problematic to mention that.

        In other words, overthrowing the power of SJWs seems like a logical first step towards achieving most of the feminist goals. Then we can start freely criticizing genital mutilation or “honor” killings without being immediately accused of racism. (Also mansplaining, if the person who criticized that happens to be male.)

        Report comment

        • Nita says:

          What makes you think it’s problematic to mention?

          (Unless you mean saying “Shut up and be grateful you’re not in Saudi Arabia!” a la Dawkins — that might indeed be perceived as rather unkind.)

          Report comment

          • Aap says:

            The feminist response to Cologne clearly showed that SJWs think that it’s problematic to say that certain cultures have more sexual violence than others.

            Not surprising, since most are strongly Marxist, so they divide society in groups who are either oppressed or oppressors. Middle Easteners are automatically in the oppressed group, so they can’t then be oppressors, unless you ignore that ‘vector’ and regard them simply as men.

            Report comment

          • Nita says:

            It seems that both Muslim and Western cultures feature a significant undercurrent of “slutty women don’t deserve to be treated with basic respect” — only the threshold of what counts as “slutty” is higher in the West (due to liberalism and feminism), and lower in Muslim countries (due to an “arms race” in signaling modesty).

            Therefore, feminists see anti-liberal, anti-feminist groups as people who want to make Western culture more like Muslim culture — which, if true, would make their cultural critiques a transparent rationalization of ethnic tribalism.

            Also, it’s not like muggers and thieves in the West are respectful of women or representative of Western culture, either.

            Report comment

          • Aapje says:

            It seems that both Muslim and Western cultures feature a significant undercurrent of “slutty women don’t deserve to be treated with basic respect”

            The difference is that in one of these cultures, it is a mainstream belief and in the other it isn’t. Of course, many feminists have the false belief that it is mainstream in the West (which they call rape culture), despite all the obvious evidence that it isn’t.

            feminists see anti-liberal, anti-feminist groups

            Conflating these two is exactly why large parts of feminism is guilty of what it purports to fight: crass stereotyping and bigotry.

            Plenty of feminism is anti-liberal and plenty of anti-feminism is liberal.

            Also, it’s not like muggers and thieves in the West are respectful of women or representative of Western culture, either.

            Those criminals target men more than women, so I don’t see why this would prove that ‘we’ treat women badly.

            This kind of hugely biased cherry picking that you engage in (ignoring all male victims and only focusing on women) is endemic in feminism. Of course, you can prove anything by just ignoring all data that disagrees with your beliefs.

            Report comment

          • Nita says:

            The difference is that in one of these cultures, it is a mainstream belief and in the other it isn’t.

            Do you really think that the majority of Muslims believe that men should grope women on the streets? Why?

            I don’t see why this would prove that ‘we’ treat women badly

            Well, obviously — because that’s not what I’m saying. I’m saying that you’re comparing average, law-abiding Western citizens to mobs of Muslim muggers.

            Report comment

          • John Schilling says:

            Do you really think that the majority of Muslims believe that men should grope women on the streets? Why?

            You left out an important word in translating Aapje’s claim there. The word is “slutty”.

            I think a more charitable reading of Muslim culture would that a majority believes that men can’t help but grope slutty women and so shouldn’t be punished, shamed, or thought less of for succumbing to that irresistable temptation but that slutty women should be punished, shamed, and thought less of for offering it. Even this may not be a majority view across a billion or so Muslims, but the insistence on highly restrictive codes of female dress and conduct suggest it is fairly common.

            Report comment

          • Nita says:

            @ John Schilling

            a majority believes that men can’t help but grope slutty women and so shouldn’t be punished, shamed, or thought less of for succumbing to that irresistable temptation but that slutty women should be punished, shamed, and thought less of for offering it

            This is the same misconception that some feminists have about Christian proponents of modesty, and I’m surprised to see you repeating it.

            No, they believe that not only groping, but even staring is wrong, even if “provoked” by the wrong behavior of others.

            Report comment

          • Aapje says:

            Do you really think that the majority of Muslims believe that men should grope women on the streets? Why?

            I never said ‘majority,’ I said mainstream. I refuse to defend straw men.

            Pretty much all Muslim cultures are heavily patriarchal (real patriarchy, not feminist patriarchy). In a patriarchy, there is a a strong separation of the sexes which increases sexual frustration, a strong belief that men lack the ability to restrain themselves (so women need to protect themselves or be protected by men who are relatives) and a strong sense that people should be punished when deviating from the norm too much. Note that these beliefs are just as common among men as women.

            The result is that there is both much more harassment in general and a much greater sense that women deserve it when they act/dress wrong. Finally, the justicial system will often refuse to go after perpetrators of severe sex crimes.

            Although there are strong cultural differences in the Muslim world, of course, Saudi-Arabia is a lot worse than Turkey.

            No, they believe that not only groping, but even staring is wrong, even if “provoked” by the wrong behavior of others.

            Ehhhhh, no. A common belief is that it is wrong for other men to do that to their relatives, but that women who ‘provoke’ deserve to be harassed. See chapter V:

            http://egypt.unfpa.org/Images/Publication/2010_03/6eeeb05a-3040-42d2-9e1c-2bd2e1ac8cac.pdf

            Report comment

          • Nita says:

            @ Aapje

            Thanks for the source.

            1. I don’t know an objective way to tell whether something is “mainstream” or not. Egyptians in your source report that harassment has become more common in the recent troubled years. Redditors were really entertained by that video of a female singer being molested by the concert crowd — and reddit is pretty mainstream, IMO.

            2. You do realize that the study you linked was conducted by a feminist organization, right? (And funded by the UNFPA and the EU — which are presumably representative of “the current political climate”.) If modern feminists don’t allow anyone to criticize non-Western cultures, how is that even possible?

            3. “Sexual harassment is wrong, so I condemn it wherever I see it” is a consistent stance.

            “Feminists’ whining about sexual harassment has destroyed our wonderful culture, where men were manly and women dressed modestly” + “Muslims demand that women dress modestly, therefore they support sexual harassment, which makes them inferior to us” is not.

            I know this peculiar pair of statements probably doesn’t describe your views, but I have seen this combination, and it’s what causes the angry feminist reactions.

            Similarly, “feminists inflate the definition of rape!”+”there were hundreds of rapes in Cologne!”.

            4. Here in Eastern Europe, a blessed land still largely free from the dreaded “PC police”, I have also noticed a correlation between anti-Muslim views and racist views, as expressed in online comments — so the idea that some of the cultural criticism may be racially motivated does not seem implausible to me.

            Report comment

          • John Schilling says:

            No, they believe that not only groping, but even staring is wrong, even if “provoked” by the wrong behavior of others.

            The two are not contradictory. “It’s wrong, but we can’t blame them for it” is a common thing, factoring into e.g. most defenses of rioting ever, and a disturbing number of date-rapes in western cultures. Sometimes even infanticide, and there we don’t have the ability to blame someone else instead.

            I don’t see any good way of distinguishing the two hypotheses in this case,but I haven’t examined the matter closely. What evidence are you using?

            Report comment

          • Aapje says:

            @Nita

            1. Unfortunately, there doesn’t seem to be a single survey with consistent methodology for different countries. It seems like such an obvious study to do that it’s quite amazing that no one did it. In the absence of such, I agree that it is a subjective analysis of various bits of evidence.

            But regardless of whether you want to call it mainstream or not, I think it is very difficult to argue that Muslim countries don’t have a higher rate of sexual violence. Especially for feminists, since it is clear that Muslim countries are more patriarchal than the West, so by their own reasoning (patriarchy causes sexual violence), it logically follows that Muslim countries would be more sexually violent.

            BTW. please don’t conclude that all Reddit users have the same beliefs of a few Reddit users in 1 subreddit. It should be obvious that there are many sampling biases and thus your conclusion about mainstream beliefs is erroneous.

            2. I was talking about mainstream Western feminists. I have a gazillion times more respect for non-Western feminists. There is an obvious difference between claiming to be suffering from patriarchy when you are actually in a patriarchy vs claiming to suffering from something you aren’t. It’s the difference between a homeless person claiming to be dead broke and Hillary Clinton making that claim.

            3. That wasn’t a common reaction at all though. The common complaint was: ‘we are taking in a lot of people from a culture where many people have beliefs that result in a very high rate of sexual violence and this is being covered up.’ When people weakmen their opponents, it says more about the people who use that tactic (that their ideology can’t deal with certain facts) than about their opponents.

            4. Eastern Europe is quite different from the West, IMO. I was talking about Western feminists. Anyway, my point of disagreement is not whether racist beliefs exist, which they do. My issue is that large parts of feminism allows some very crude generalizations (all men are culturally conditioned to use sexual violence), but react angrily when confronted with a claim that certain cultures have more sexual violence than others. Note that they are hugely hypocritical about it, since I’ve never seen a feminist who didn’t think that ‘bro’s’ are more likely to engage in sexual violence. So it’s not like they don’t believe that some subcultures are more sexually violent. However, they are bigoted against white men* and if non-white men are worse in any way, it would undermine their hierarchy of oppression (where white men are the worst oppressors and everyone else supposedly oppresses less badly).

            * It’s amazing how often the term ‘white men’ is used in feminist writing, in ways that would be seen as racist if you’d replace it with ‘Jews’ or ‘blacks.’

            Report comment

          • Nita says:

            @ John Schilling

            a common thing, factoring into e.g. most defenses of rioting ever, and a disturbing number of date-rapes

            You’re absolutely right. And after seeing those date-rape defenses, feminists tried labelling their source “rape culture” — and everyone else went “how dare you!?! our culture is 100% against rape! now I’ll never listen to feminists again!”. So, if it’s offensive/unhelpful to call these unfortunate attitudes “rape culture”, then it’s similarly unhelpful to say that Muslim culture promotes mass molestation.

            Of course, we’re not glad that a large number of the people displaced by the war happen to have these sexist cultural beliefs. But at least they’re not the majority and don’t have the law on their side — and feminism did succeed in changing Western culture, despite starting from a much weaker position.

            Currently, the main barrier to changing beliefs in non-Western cultures is their suspicion that the West is an immoral society with alien values. And the things our local Muslim-haters tend to say and do are really not helping.

            Report comment

          • Aapje says:

            @Nita

            And after seeing those date-rape defenses, feminists tried labelling their source “rape culture” — and everyone else went “how dare you!?! our culture is 100% against rape!

            That is a straw man. The actual argument is that our culture is not a rape culture because:

            – The overwhelming majority of people think that rape is one of the worst crimes that a person can experience
            – Very few people think that people deserve to be raped as a result of something they do (‘victim blaming’ is almost never blaming victims for the crime, but rather blaming them for taking risk seen as irresponsible, which is an important distinction that many anti-rape activist seem unable to make)
            – The police/courts actually do their best to prosecute, within generally accepted judicial standards, which unfortunately have their limits, especially for crimes that are often he said/she said
            – There are very few rape jokes and those that exist, trade on the taboo, rather than on the legitimacy of rape

            There is one specific exception, which is prison rape of men, where the attitude of society is very different. My experience that feminists always talk about rape culture as something that effects women, while ignoring only real rape culture that exists in the West, is one of the reasons why I think that there is an extreme pro-woman/anti-man attitude in feminism.

            So, if it’s offensive/unhelpful to call these unfortunate attitudes “rape culture”, then it’s similarly unhelpful to say that Muslim culture promotes mass molestation.

            It’s not offensive to call out people with bad beliefs, it’s offensive/unhelpful to claim that the West has a ‘rape culture’ in general.

            The main reason is that (like a lot of feminism), it really seems to be an attack on white men. Why else would these feminists refuse to use the word for Muslim culture(s), but use it liberally for a culture with much less sexual violence? You fail to address those double standards, Nita, but those are why people get so upset with feminism. It’s one thing to be wrong, but it’s particularly irritating when people cannot even be consistent in their wrongness.

            Currently, the main barrier to changing beliefs in non-Western cultures is their suspicion that the West is an immoral society with alien values. And the things our local Muslim-haters tend to say and do are really not helping.

            They’ve started giving sexual behavior education to immigrants thanks to the ‘Muslim-haters’ who demanded that the government actually addresses these crimes. Isn’t it strange that one this issue, it is the feminists who try to downplay the sexual violence, just because the perpetrators are a ‘victim group?’ Isn’t that rape culture by their own standards?

            Report comment

          • John Schilling says:

            I’m not trying to be helpful, I am trying to be truthful. Once we have truthfully established the zone of possible agreement, we can either try to help reach an agreement or pursue some alternate strategy. We aren’t at that point yet.

            Report comment

          • Nita says:

            @ Aapje

            The police/courts actually do their best to prosecute

            Feminists have worked extremely hard to make that happen. In the past, it used to be “oh, are you married to him? that’s not rape”, “oh, are you a prostitute? that’s theft, haha”, “oh, were you drinking? tsk tsk, what did you expect?” — and everyone still believed that their culture was totally against rape, it’s a terrible crime, etc. etc.

            And yes, someone should do that kind of hard work for the rights of male prisoners, too. Making everyone hate feminists for being biased won’t solve this problem.

            @ John Schilling

            I don’t mean helpful for me or for our discussion here. I mean helpful for changing the beliefs we consider harmful.

            Personally, I’m against using the phrase “rape culture”, precisely because it’s so counterproductive. If you want to argue with someone who believes that Western countries do have “rape culture” but Muslim countries don’t, you’ll have to go find that person.

            Report comment

          • John Schilling says:

            I mean helpful for changing the beliefs we consider harmful

            How do you propose to do that without truthfully acknowledging what those beliefs are?

            Report comment

          • Aapje says:

            @Nita

            I am discussing feminism today, you are derailing the conversation into a discussion about historic feminism.

            Making everyone hate feminists for being biased won’t solve this problem.

            It’s not about hate or us vs them. It’s about countering a harmful narrative that creates more victims.

            It’s not very useful to engage in emotional tribalism.

            Report comment

          • Nita says:

            @ John Schilling

            How do you propose to do that without truthfully acknowledging what those beliefs are?

            Why, by sneaking competing beliefs into their minds, and letting cognitive dissonance do its job, of course. Certainly, I’ve never persuaded anyone by telling them what their beliefs are.

            Then there’s the inconvenient little fact that the previous time Europeans collectively focused on the bad qualities of specific minorities, we ended up killing members of these minorities in rather large numbers. So, many people are wary of heading in a similar direction again.

            Report comment

          • Aapje says:

            @Nita

            It is really rather offensive to argue that:

            A. Honestly assessing cultural differences and asking for non-violent solutions, such as proper education and migration limits, is on par with genocide. This kind of over the top response is exactly why the P.C. left has such a bad name. Furthermore, the ironic truth is that this unwillingness to honestly debate the issue and be truthful out of fear of extremism, is actually driving people away from moderate politics towards extremists.

            B. Europeans collectively murdered the Jews. My grandparents on one side hid people from the Nazi’s and my other grandfather was a forced laborer. Neither chose to commit genocide or had the power to stop it. When you are holding Europeans collectively responsible, you are insulting my family and the many other Europeans who suffered from Nazi occupation.

            Now, you are free to act as the worst stereotype of people who have beliefs similar to yours, but that may have the opposite effect of the one you seek to achieve.

            Report comment

          • Nita says:

            @ Aapje

            A. If you re-read my comment, you will notice that I did not say that. I said that the things some people are currently saying (“their culture is bad! get rid of them!”) remind other people of the beginning of that chain of historical events, and that has an impact on their actions.

            B. As far as I know, my ancestors didn’t join any death squads, either. But was there antisemitism in my country? Yes. Did it make recruiting for death squads easier? Most likely, yes. It doesn’t make everyone in Europe guilty of anything, but it is a clue to what kind of thing we should probably try to avoid.

            Yes, it would be fantastic if we could all come together and discuss the issues caused by the refugee crisis calmly, without accusing anyone of being either an orc or a Nazi. Then we might actually have a chance of getting through this mess without too many casualties. Unfortunately, polarization and extremism on both sides has become a huge barrier to that.

            I don’t know what beliefs you’re attributing to me, in particular. I’ve tried to explain some of the psychology and reasoning of so-called “SJWs”, to the best of my ability, in the hopes that it might help you form a more accurate idea of their opinions and talk to them more effectively.

            And I note that my original question to Viliam — what made him think that FGM and honor killings are impossible to discuss — still hasn’t been answered 🙁 (I’ve actually seen those subjects discussed in modern feminist spaces, so I hoped to learn how Viliam got such a contrasting impression.)

            Report comment

          • Aapje says:

            It just legitimizes overly negative narratives about certain ethnic groups when they get countered with overly positive narratives. When these feminists argue that people from these cultures are no more sexually violent than natives, the facts becomes a weapon for the other side. In the German context, where the the government tried to cover up the truth, more and more people will simply stop believing their government and lose faith in the political process (with violence as the result).

            To answer your question about FGM and honor killings (and more): in my country we had a unified front by mainstream political parties against anyone who dared say anything negative about immigrants and an effort to prevent inconvenient facts about immigrants from being researched/published. Beliefs shared by a large minority in society were not represented in politics (except 1 racist party which was silenced and for which most in that minority wouldn’t vote for either). Because negative aspects of migrant cultures could not be discussed in the political arena, it was obviously impossible to get them addressed with policies, which just made problems worse and worse (like FGM and honor killings, although these were actually lesser issues).

            Ultimately, this resulted in a political revolt by an outsider, who was murdered during the election campaign by a SJW. This destabilized the political arena permanently, greatly increased polarization and distrust of the government by that minority, caused a lot of violence against minorities, etc, etc. I put most of the blame for this unfortunate chain of events on the P.C. multiculturalists, who tried to put out a fire with gasoline, making everything a 100 times worse than if they had simply chosen the truth over ‘they can’t handle the truth.’

            Very similar polarization happened in France, Denmark, Belgium and now the UK, Sweden and Germany. Meanwhile, the do-gooders wallow in their moral superiority, while they unwittingly cause(d) many of the problems they smugly blame on the other side.

            Mark my words, with the huge influx of migrants in Germany, the inevitable friction between different cultures will become so large and the costs of integrating people so massive, that you will get a huge blow-back once citizens realize that the consequences of open borders are not just a pleasant feeling about ‘doing the right thing,’ but more crime, terrorism, cuts in government services, etc. Don’t be fooled by how the opposition to open borders is currently not that strong. It was similar in my country until it suddenly flipped and culturally, we are probably more like Germany than any other country (although we don’t have the same feelings of guilt, which is why Germany is lagging behind the rest of Europe on this issue, but this won’t last forever).

            Report comment

      • perpippity says:

        I guess you really feel for mainstream Republicans, then, who wish their party weren’t associated with the Religious Right, Ann Coulter or Trump. The thing about both groups (feminists and political parties) is that about all a person needs to do to belong is to self-identify as such. (Yeah, you have to register for the party, but that’s a rubber stamp. The party itself can’t prevent it, as Trump’s example demonstrates.)

        Report comment

    • Cord Shirt says:

      Just wanted to repeat here that IMO *second-wave* feminism was originally a movement by and for *nerdy* women. Not “mean girls.”

      Oh, did the popular crowd co-opt it? Well gee, it’s not like *that’s* ever happened before. :facepalm:

      Report comment

  15. BBA says:

    Maybe everyone’s looking at it the wrong way, and it’s not that we need more women in programming (and other STEM fields), it’s that we need fewer men.

    Report comment

    • ThirteenthLetter says:

      Even better conclusion: we need fewer journalists.

      Report comment

      • Ptoliporthos says:

        We need more journalists and fewer hacks that copy-paste press releases. Most science journalism looks like this. Why does anyone believe journalists do any better covering economics or politics? About the only reporting I ever find consistently reliable covers current weather and traffic conditions, and whether the visiting team or the home team lost.

        Report comment

        • Muga Sofer says:

          This wasn’t a press-release. Someone tracked this down so they could turn it into a story.

          Report comment

          • Ptoliporthos says:

            I have a hard time believing that 2-3 dozen science writers were all perusing the preprints of the same impact-factor 2 journal and all decided to write the same story about the same article.

            Most universities will prepare a press release to accompany a publication that includes authors, no matter how tangentially involved, from their institution, if they think the study might attract significant press interest. I imagine the most likely scenario is that somebody in the PR office at Cal Poly or NC State has a nose for clickbait, and sent out a press release to accompany this preprint.

            You could probably reconstruct the text of the press release from the articles using the same techniques scholars use to infer the text of lost manuscripts from surviving copies, quotations, and summaries.

            Report comment

          • HeelBearCub says:

            It’s not a preprint is it? It’s a submission for open peer review. That means it hasn’t even been accepted by a journal for peer review.

            My prior is that somehow some journalist found out about it and wrote the first story. All of the other stories just piggyback off that one.

            Report comment

          • Muga Sofer says:

            Yeah, I assumed this was a case of everyone else copying the initial fake/distorted story.

            This happens all the time; in fact, in my experience it’s more common than the “press release” thing.

            Report comment

      • anon says:

        fewer journlolists

        Report comment

    • patches welcome says:

      The careers definitely need to stop being shilled to people who are just desperate for a secure paycheck. Fortunately for open source, it doesn’t care what your day job is.

      Report comment

  16. Alex Mennen says:

    > For example, suppose that the reason revealing gender decreases acceptance rates is because corporate contributors tend to use their (gendered) real names and non-corporate contributors tend to use handles like 133T_HAXX0R. And suppose that the best people of all genders go to work at corporations, but a bigger percent of women go there than men. Then being non-gendered would be a higher sign of quality in a man than in a woman.

    In that case, being gendered would be a sign of quality, not the other way around.

    Report comment

    • Theo Jones says:

      Agreed. If I were running an opensource project, I would take people who use their real names as more credible. And I would be very skeptical of someone who used some stupid username like 31137t_Hax0r. And its not just me. This is what ESR has to say on the matter.
      http://www.catb.org/esr/faqs/hacker-howto.html
      The problem with screen names or handles deserves some amplification. Concealing your identity behind a handle is a juvenile and silly behavior characteristic of crackers, warez d00dz, and other lower life forms. Hackers don’t do this; they’re proud of what they do and want it associated with their real names. So if you have a handle, drop it.

      Report comment

      • Aegeus says:

        This doesn’t seem supported by the data, because using their real name (which reveals their gender) *decreased* acceptance rates. Apparently the people who run github projects think the opposite of the way ESR thinks.

        Personally, I don’t give a crap whether you use a handle or not. Certainly, l33thax0r-type names show a lack of sophistication, but in this day and age, separating your real name from your online identity just seems like common sense. And it’s easy to come up with a handle that’s recognizable without sounding like a tryhard or script kiddie.

        For instance, you might use a normal-sounding pseudonym, like “Scott Alexander.” 😛

        Report comment

        • Theo Jones says:

          I realize that the evidence seems to contradict the favoritism for real names idea, but my point was that its a result I wouldn’t have expected.

          Report comment

        • Good Burning Plastic says:

          “Scott Alexander” would have been counted as a gendered name in such a study.

          Report comment

          • Muga Sofer says:

            I thought they checked if your screen-name matched your real name to determine whether it was “gendered”? (Alongside other indicators, like listing it in your bio.)

            Actually, I hope they didn’t, because I thought of Ozy as male for the longest time based on their username.

            Report comment

          • Cadie says:

            My full legal name would have been counted as a gendered name too – and they’d have put it in the wrong category unless they actually went to the trouble to dig up my Facebook account to verify. “Oops. [Male-Sounding Name] is a woman, looks like one, says she’s female, and is listed as [Other Person]’s sister. Better move her data to the other category.”

            Of course, I doubt there are enough women with masculine names to significantly affect the data… unless a big enough minority of women are choosing masculine names as pseudonyms on purpose. If “Erica Lastname” doesn’t think she’ll be taken as seriously or just wants to avoid the issue, there’s nothing to stop her from using “Eric Lastname” as her handle and email name, and checking her handle and Google account isn’t going to reveal anything. Plus there are plenty of nicknames that are unisex but usually assumed male unless otherwise specified (Alex, Chris, Sam…) that some women would use online without specifying.

            Report comment

        • God Damn John Jay says:

          And it’s easy to come up with a handle that’s recognizable without sounding like a tryhard or script kiddie.

          Yup.

          Report comment

      • suntzuanime says:

        ESR is incorrect about a lot of stuff.

        Report comment

        • TrivialGravitas says:

          In this case I think he’s just old. It’s hardly the first time I’ve heard it, but I’ve never heard it from a programmer younger than me, and I’m into my 30s now. the typical 20 something on github has different fashion around names.

          Report comment

      • The Nybbler says:

        Well, 1337 5p34k has been stupid since its inception, even if a few good things like “pr0n” have come out of it. But there’s good reasons to use a handle; I use mine to keep the “real world” separate. It wouldn’t be hard to for a determined person connect my real name with “The Nybbler”, but it isn’t going to be a simple Google search. There may be good reasons (even aside from gender issues) for a person not to use their real name on an open source project

        1) Their government might be hostile to the project, even if there aren’t actual laws against it. Think encryption, or anything touching on the wide world of copyright.

        2) Their real-world employer might be hostile to the project or the whole idea of their employees “moonlighting” on open source.

        3) They are concerned a dispute on an open source project might spill over to the real world, or vice versa

        4) They’ve become known under a handle and they see no reason to change it.

        But what do I know? My handle is a reference to a tool used in cracking copy protection schemes, and also not one I use on open source projects.

        Report comment

      • Shieldfoss says:

        He says it behind a handle. Just because you and I know that “ESR” are his initials doesn’t magically turn it into a name instead of a handle.

        Additionally: That was written before the current social milieu, before people got the notion into their heads that if you disagree with somebody, it is acceptable to call their boss to try and get them fired, or a SWAT team to get them killed. I used to post many places under my real name, but I don’t any more.

        Report comment

    • HeelBearCub says:

      Upthread the point is made that one reason to have your real name as your user name is to build resume.

      People trying to build resume are less likely to submit great code.

      Report comment

      • Adam says:

        Data point of one here, but I use my real name on GitHub (mostly because it’s my name and I use it pretty much everywhere), but virtually everything I commit is to private repos I have commit access to. I’ve only ever submitted three pull requests I can remember to open source projects (honestly not even sure if they were accepted or not, was just forks I added features to for my own work and decided to offer it back to them because other people found it useful and said I should). If I were to frequently submit pull requests to open source projects, I can’t imagine devoting the same level of effort that I devote to my actual paid work. That isn’t even just on interest, but partly out of duty. Someone paying me deserves more effort.

        Report comment

    • Tommy says:

      >In that case, being gendered would be a sign of quality, not the other way around.

      You missed the second half of Scott’s claim:
      >Then being non-gendered would be a higher sign of quality in a man than in a woman.

      In the specific situation he describes (gendered names = corporate; high-quality go corporate; women more inclined to go corporate than men), non-gendered men will be of higher quality than non-gendered women. I believe this is what he means, despite the awkward phrasing.

      Report comment

    • I have no idea to what extent if any this applies to the community being studied, but for what little it’s worth I gather that at least some open-source communities are pretty much openly hostile to anyone working for “the man”.

      Report comment

  17. “Declare how many different subgroup groupings you tried, and do appropriate Bonferroni corrections.”

    Argh, no! We have good methods now for dealing with this problem: include predictor variables for the subgroups and do a multilevel Bayesian regression (http://www.stat.columbia.edu/~gelman/research/published/multiple2f.pdf). And we have readily-available, open-source software to do it, such as the rstanarm package for R.

    If any researcher reading this blog would like to analyze their data in this way but needs some help, drop me a line; I’ll walk you through it. (Offer limited to the first few people who ask.)

    Report comment

  18. Richard says:

    One question I had reading the paper that I’m hoping some of the statistically-minded readers here can help answer: Looking at Figure 5, I was curious how many people in each category were insiders and how many were outsiders, so we could get a sense of what the average numbers look like without the insider/outsider split. Considering men see a medium dropoff upon being gendered in both insider and outsider categories while women see no dropoff in the insider category and a large one in the outsider category, it looks like the two could average out to about equal values, but it depends on the relative sizes of those groups. We can get close by looking at the earlier results section (p.7), when they say they have n=3064667 pull requests (which can be verified by adding up the total of the values in the table on that page), and that if we exclude insiders we have n=2473190. This tells us that a bit over 80% of the pull requests come from outsiders, although we don’t know if the insider:outsider ratio is the same for women and men. What confuses me, though, is that if you look at the graphs on p.15 again, the confidence interval seems to be roughly the same width for outsiders as insiders, and if anything, it’s wider for the outsiders. Am I misunderstanding how those error bars work to think that they’d by necessity be narrower if there was a 4x larger sample size for outsiders than insiders, or did they make a mistake in that graph?

    Report comment

    • Douglas Knight says:

      Yes, that’s weird. It’s not logically impossible, though, because not every profile in the previous set goes into this graph. Also, the gendered profiles have very small error bars. I believe that is not because profiles are overwhelmingly gendered, but because the set was defined by an automated process, while the neutral profiles had to pass manual inspection, and thus are a small sample. But for the error bars you mention to make sense, there would have to be a correlation between outsiders and being excluded from the data.

      (Also, error bars do not depend only on the sample size, but also on the point estimate. That the outsiders have a rejection rate closer to 50% raises their error bars. The contribution from the 1:4 sample ratio is 2:1, while the contribution from moving from 90% to 70% is 2:3. Combining these gives 4:3, which is not very far from 1. We observe a ratio near 1, but on the other side.)

      Report comment

  19. merzbot says:

    “Science journalism is a garbage fire. Seriously, it’s complete trash,” has been the rationalist refrain since the dawn of time. I wouldn’t be so quick to point the finger at “institutionalized social justice,” whatever that means. Besides, the mainstream media explodes with articles about how ridiculous student SJ activists are whenever that flares up. They sensationalize anything for views. Their only ideology is ad money.

    Report comment

  20. Mark says:

    Clear proof that tech is run by hostile androgynes who are sexist against both men and women (but slightly more the latter).

    Report comment

  21. Henry Sartain says:

    > For example, suppose that the reason revealing gender decreases acceptance rates is because corporate contributors tend to use their (gendered) real names and non-corporate contributors tend to use handles like 133T_HAXX0R. And suppose that the best people of all genders go to work at corporations, but a bigger percent of women go there than men. Then being non-gendered would be a higher sign of quality in a man than in a woman.

    Wait, isn’t this backward at the end? Being non-gendered is a sign of *low* quality in this story, being one of those non-corporate 133t haxx0rs. And it’d be a weaker sign of low quality for men than women, or conversely being gendered would be a higher sign of quality. Or am I missing something?

    > In one subgroup, unblinding gender gives women a bigger advantage; in another subgroup, unblinding gender gives men a bigger advantage.

    Similarly this sounds backward — these data show unblinding gender giving everyone a disadvantage, not an advantage.

    Report comment

    • Murphy says:

      Think in terms of the projects rather than just the people.

      If company repos are tougher on code on average and people submitting to company repos are more likely to be gendered.

      If a larger fraction of identified women are submitting to corporate repos than hobbyist repos then you could see both higher quality from women and a higher pull request rejection rate.

      Report comment

    • Deiseach says:

      Wait, isn’t this backward at the end? Being non-gendered is a sign of *low* quality in this story, being one of those non-corporate 133t haxx0rs. And it’d be a weaker sign of low quality for men than women, or conversely being gendered would be a higher sign of quality.

      If gendered names means someone working for a corporation, and the very best go to work for corporations, but more women work for corporations than men, then a name that is identifiable as a man’s name means he must be one of the very best.

      But not all the very best men go to work for corporations. Some (many, lots?) are non-corporate contributors and use neutral names or pseudonyms. Whereas it is only the less capable women who use neutral names/pseudonyms (all the really good women are working for corporations and use their real names).

      So if you can identify gender by name, then it is more likely to be a woman and you can assume she is one of the best. If the contributor’s name is gender-neutral and it turns out to be a man, then you can assume he is better than, or at the very least equal to, a similarly pseudonymous woman.

      Report comment

  22. Kyle says:

    As a daily user of GitHub it’s pretty clear that this study was performed by people unfamiliar with how the site (or version control in general) is used. I think looking at sexism in acceptance rates is a perfectly fine endeavour (if the stats are interpreted properly) but this study went off the deep end by claiming that women are somehow more competent which reveals their bias and perhaps compromises their work as a whole.

    1) I am surprised that there are only 20x more PRs by men. I would say it’s more like 100-200x more men on GitHub then women.

    2) But the women that are there are usually always employed by some small to medium tech company. So this study is possibly comparing mostly professional women with mostly hobbyist men.

    3) PRs from companies are less likely to be the work of a single person. It’s often the case that someone has bundled up fixes the whole team has contributed to and then made them public.

    4) In my experience a lot of people on GH with feminine sounding names are actually men from Eastern Europe (Sasha, Nicola, etc) or India (Sandy, Sunny, etc).

    5) A lot of Americans who do have photos as their avatars often use the douchiest looking hipster picture. This makes me a lot more prejudiced against them than say the Europeans who have a regular pic of themselves in work clothes.

    6) True “insiders” would usually not be making pull requests as they would have commit access to the repo. So when you look at PRs you are by definition looking at inter rather than intra-group communication. You can’t look at PRs and say women are more “competent” while totally ignoring everyone else who has gained commit access which is a much higher achievement than having a PR merged.

    7) The number of lines altered by a PR has little to do with the complexity of it. It could be that you only altered 10 lines of code but had to add 1500 lines of regression tests.

    8) Generally speaking smaller PRs are better and large PRs are likely to get rejected and you will be asked to split it into multiple small PRs. If women are getting larger PRs merged then it’s possibly because men are being less critical of them.

    9) For more complex patches PRs are often used as a code review. They’re then rejected and another one is opened with the advice taken. That could be rejected as well, and repeat… Thus you might find that the best coders who take on the biggest features have the lowest acceptance rates.

    10) Finally, a closed PR does not mean the code was bad. The author could have closed it because they selected the wrong branch or the reviewer could have closed it but taken inspiration and implemented the fix themselves in a slightly different way due to their better understanding of their own project. Similarly more complex PRs are more likely to sit as open for weeks or months because nobody has time to review them but this study counts them as rejected.

    Report comment

    • Murphy says:

      Actually 2 sounds quite testable, check if repos are company-owned, see if women are more likely to be submitting to those and see if company owned ones are tougher on submissions in general.

      Report comment

      • Kyle says:

        I think you would probably end up biased towards San Francisco startups. The big players (Amazon, Google, Microsoft, etc) who pay programmers six figures keep most of their repos private and/or on their own platforms.

        One thing I’m sure you will find with company owned repos is that they won’t merge requests without tests. I bet this is why PRs from women are 50% longer.

        Alternatively you could look at organisation accounts on GitHub which are special groups that can own repos and then add all their members as committers. To be a member of a non-commercial organisation is pretty much a stamp of approval that you’ve been a long time high quality contributor and it would be a much better signal of quality than whether or not random PRs were merged.

        Also nobody has mentioned that all this study looked at is public GitHub repos and GH is heavily biased towards web developers. So any conclusions shouldn’t be applied to programming or tech as a whole.

        Report comment

        • Kyle says:

          In regards to tests: sometimes it can be easier to get a PR merged to a company’s repo.

          Big projects tend to have tests, style guides, and so on. If you follow the rules then they will often merge your request with little resistance.

          However smaller projects (especially one man projects) often have no tests, only personal style, and suffer from “not invented here”. Getting PRs accepted to them is harder and they will often tell you to fork instead.

          Report comment

    • Jeff Kaufman says:

      RE 6: I work on an open source project for work, and every commit goes in via pull request, even for “insiders” with full commit access, because that’s how we do code review.

      Report comment

    • Also re 6, I maintain an open-source project, and our project workflow is also pull-requests-only, including maintainers’ commits. We adopted this process several months in to the project, though.

      Report comment

      • Kyle says:

        (replying to Jeff as well)

        That’s a common pattern for projects larger than a group of friends but what sort of projects make up the majority of GH and what sort of project are the majority of PRs going to? If we’re going to assume that every project is run professionally then we might as well only look at the top 100 projects in detail rather than scraping data about millions.

        For example when I just now looked at famous web projects (Django, Node, RoR) some things stood out that were probably missed by the authors:

        1) People with commit access are often merging their own requests once the review is done (especially Django).

        2) Many people are closing their own requests after noticing a mistake (especially RoR).

        3) Many merged PRs are showing up as closed presumably because they are being merged locally rather than through GH (seems to be the majority of PRs in Node and about 50% in Django).

        Yet this study assumed that people are merging or rejecting other peoples PRs and that a closed PR is a rejected PR. If the above three points are widespread then they alone could scuttle this study.

        Report comment

    • Ricardo Cruz says:

      I have posted some of my projects in github, but I don’t know what a PR is… Is it related to the bug tracking system? (I never used one, my projects are too small for that :))

      Report comment

      • brad says:

        PR = Pull request. Anyone can ask you to include a patch in your repository via a pull request. For small out of the way projects you aren’t likely to get very many.

        Report comment

  23. Daniel Speyer says:

    If women submitting hidden have a higher success rate than men submitting hidden, then that means they must be submitting higher quality work. That probably means that their standard for how good code has to be before submitting is higher (since basically everyone creates code of many qualities). That could mean that they’re accustomed to being held to a higher standard, or that they’ve *heard* they will be so often that they’ve internalized it, or that they’re more likely to be intrinsically motivated and therefore take greater pride in their work…. I guess there are too many possible explanations to conclude much. Still, there’s probably something interesting going on.

    Report comment

  24. Tom Bull says:

    “Every time I say I’m nervous about the institutionalized social justice movement, people tell me that I’m crazy, that I’m just sexist and privileged, and that feminism is merely the belief that women are people so any discomfort with it is totally beyond the pale. I would nevertheless like to re-emphasize my concerns at this point.”

    Really, what is the worst that could happen if there is an institutionalised social justice movement like you say? Where people deliberately or subconsciously misinterpret results to skew their implications towards benefiting a marginalised group? What’s the absolute worst thing? That people become more conscious of disadvantaged groups and implement positive action in their favour? That groups who have had hundreds or thousands of years of institutionalised advantage are now marginally disadvantaged?

    Damn! That sounds like a hellish dystopia if I ever heard of one. Sign me up for meninism, we’ve got to fight this!

    Report comment

    • Murphy says:

      When you start throwing reality in the bin in favor of political positions you lose touch with reality.

      When politics starts to set what is “acceptable” for science to conclude then it’s a short step from there to book burnings and academic purges of those who come to the “wrong” conclusions.

      I’m not talking in the abstract here, it’s what actually happens and has happened many times in history.

      Report comment

      • Tom Bull says:

        Hey Murphy, I’m probably being thick, but your comment seems somewhat unsubstantiated. Maybe if you could cite an example of one of the many times this has happened in history – and the dire consequences that ensued… that would be really useful in convincing me. Especially if you could show how some similar dire consequences could result from our current situation…

        Report comment

        • Montfort says:

          Here are several classic examples.

          Report comment

          • Tom Bull says:

            So, to be clear: the comparison is between soviet scientific suppression in favour of a brutal entrenched regime on one hand and modern bias-led misreporting of science where the bias is somewhat towards historically marginalised groups?
            Presumably then, the suggestion is that the potential dire consequences of our current path is that as the black lesbian elite consolidate their wealth in Moscow, the white male proletariat will be left to starve in the fields?

            Report comment

          • Montfort says:

            Lysenkoism is hardly a “brutal entrenched regime”. It even (allegedly) increased crop yields a little bit. It also stalled a lot of biological research that the Soviets might have otherwise done.

            The dire consequence here, then, is that we might get stuck believing false things instead of true things. True things have the benefit of working, and so enable things like (at its more dramatic side) effective antibiotics and safe, cheap transportation.

            The communist government was (one) instrument of the enforcement of lysenkoism, not a result of it, so it makes no sense to imagine one springing up here just because we refuse to ask for rigor in social sciences. Perhaps you have pattern-matched the mere oblique mention of the soviet union to other, bad arguments?

            Report comment

          • Tom Bull says:

            It wasn’t entirely pattern-matching: the examples you originally posted included others besides Lysenkoism. But what I expressed was supposed to use a bit of humour and satirise the situation to make a point (I hoped that was mostly clear… but expressing oneself on the internet is difficult).

            To be as clear as I can, what I’m saying is that whilst it’s deeply important to strive for accuracy everywhere it’s important to keep the levels of potential harm done by certain biases in perspective.

            Report comment

          • Samuel Skinner says:

            The problem with using ‘level of harm caused by missteps’ is that is also something discovered by scientific findings. If those people are also systematically biased, things tilt more and more. It isn’t the Hodomor (the peasants have little grain, but I’ll report high output and things are obviously fine because everyone else is reporting high output), but it does form a clear trend. If you think that past scientific racism was caused by such a pattern (which unlike the USSR involved no coercion by the state), that gives a pretty good idea of how crazy things can get.

            Report comment

          • Tom Bull says:

            So, the conclusion is that the importance of every inaccuracy is paramount? They all REALLY matter. And we should put as much effort into righting the inaccuracies caused by bias in favour of equality as we should in (for example) correcting the chasm of inaccuracy that separates worldwide drug policy from reality?

            Report comment

          • Samuel Skinner says:

            Yes. To do otherwise is to admit that you aren’t actually interested in getting true results, but results to support ‘your side’ at which point credibility dies and you are back to no one believing anything you claim.

            Report comment

        • Earthly Knight says:

          Maybe if you could cite an example of one of the many times this [i.e. book burnings and academic purges] has happened in history – and the dire consequences that ensued…

          I don’t know about book burnings, but academics are being purged right here and now for dissenting from social justice orthodoxy. Here are some victims:

          Tim Hunt, University College London
          John McAdams, Marquette
          Teresa Buchanan, LSU
          Subramanian Swamy, Harvard
          Thomas Thibeault, East Georgia College
          Patricia Adler, UC Boulder
          David Barnett, UC Boulder

          The Foundation for Individual Rights in Education (FIRE) keeps a fairly exhaustive archive of all major violations of academic freedom that occur in the US. I encourage you to look through it, you will find no shortage of professors fired and students disciplined for reasons of political correctness.

          Report comment

          • Randy M says:

            It sounds like most of those are white or men. So what if they are now slightly disadvantaged? We probably didn’t need any of the advances they were developing anyway.

            Report comment

      • At first I thought you were being satirical, but by the time I got to the end of your comment, I’m adequately sure you weren’t.

        One possibility is that you don’t get your disadvantaged groups quite right, and end up being unfair to people as a result.

        Report comment

        • Tom Bull says:

          I agree that that is not just a possibility – it’s probably already happening. But what I’m asking is: really is it that disastrous if white men have a marginal disadvantage for a while, with all the privilege we already have?

          Report comment

          • Tom Bull says:

            Also, just to be clear… I’m totally in favour of pursuing accuracy in these endeavours – and I’m impressed by and in general agreement with the analysis of the paper by the post’s author. I just don’t draw the same conclusion: namely that a somewhat widespread bias towards pursuing social justice has got out of hand and the world is being made worse because of it.

            Report comment

          • Samuel Skinner says:

            If politics has taught us anything, it is never ‘just a little bit’. Movements take as much as they can before they hit entrenched opposition. Setting the entrenchments on ‘this is what can be justified by the evidence’ is a good rule of thumb and helps prevent things from spiraling out of control. After all if your policy is ‘screwing over white males is okay’, you find that you start creating white male interest groups which is bad if your goal is to reduce gender and racial identification.

            Report comment

          • Tom Bull says:

            Yep, I agree mostly (apart from the word ‘never’, which should probably be rewritten ‘sometimes, on the worst occasions, not’). However I’m trying to say that portraying it as ‘us v. them’ and being ‘entrenched’ is not productive. By working together towards more equality, by guiding the debate rather than digging in at one extreme, by letting some of the less harmful, less severe biases slide… we can achieve great, productive, useful things together.

            Report comment

          • anon says:

            I believe I’m working towards more equality by not letting biases slide. Why don’t you work together with me instead of me working together with you? We could achieve great things together.

            Report comment

          • Tom Bull says:

            OK anon, I’m with you! What is the next stop in our global crusade against any bias? Whilst we’re making sure that the press don’t make any more missteps that could disadvantage white men, maybe we should rigorously fact-check some of those nasty articles that suggested there was a hint of impropriety that led to the 2008 financial crash?

            Report comment

          • anon says:

            As long as it is sloppy journlolism, I’m always up for a laugh!

            (same anon as above, I keep changing email and I don’t know what I used for that one)

            Report comment

          • Jason K. says:

            “But what I’m asking is: really is it that disastrous if white men have a marginal disadvantage for a while, with all the privilege we already have?”

            You cannot rob Peter to pay Paul and accurately call that “justice”.

            Report comment

          • “really is it that disastrous if white men have a marginal disadvantage for a while, with all the privilege we already have?”

            I think it’s pretty bad if white men with significant disadvantages (most likely poverty) are skunked and then blamed for resenting it.

            I wish the work which has gone into manipulating outcomes for and against various groups had gone instead into figuring out how people can perceive competence and reliability more accurately. Or, if they’re hopelessly bad at perceiving competence and reliability, into promoting random choices in the ranges where there’s no good reason to make distinctions.

            Report comment

          • John Schilling says:

            But what I’m asking is: really is it that disastrous if white men have a marginal disadvantage for a while, with all the privilege we already have?

            Any such disadvantage will not be applied uniformly over all white men. It will almost certainly be applied preferentially to the white men who in fact lack privilege on account of A: easier and B: everybody still wants the actually-privileged white men to like them.

            Doing collective punishment right is really hard and almost nobody bothers. Doing collective punishment wrong is always bad and can be disastrously bad.

            Report comment

          • Mark Atwood says:

            It will almost certainly be applied preferentially to the white men who in fact lack privilege

            And historically, such men have demonstrated a MO of eating shit for a generation or so, and then suddenly trying to burn down your civilization.

            I can’t even say they are wrong for it.

            One of the things I do with my career is mentor selected individuals of that class into having the soft and STEM skills necessary to get an open source developer job, despite the fact that all the current “fix the diversity problem” work infesting my field is transforming the online and facetime forums for the field into breathtakingly “unsafe spaces” for impoverished working class white men .

            If something doesn’t change soon, there is going to be a social space fracture, as these men form their own growing venues and forums. And again, I really can’t say they will be wrong in doing so.

            Report comment

          • onyomi says:

            Yeah, I see it at work in my own field, where being a white man is definitely a disadvantage, but most of the people with power are old white men (because being a white man was an advantage when starting a career thirty or forty years ago, but is a disadvantage now; hence the people with seniority who are making decisions are old white men, but it is harder than average for a young white man to start a new career).

            The move to address the historical privilege of white men has taken away no power from the old white men who grew up with that privilege, but it has hurt young white men who did not. And they wonder why young white resent all this “privilege” talk.

            Report comment

          • Vox Imperatoris says:

            @ Mark Atwood:

            Of course they would be wrong in doing so. At least if you admit that groups like the Nation of Islam were wrong in their reactions to racial discrimination.

            Report comment

          • Jaskologist says:

            Really, is it disastrous if we oppress a group of poor people just because they have the same skin color as some other people we don’t like?

            Report comment

          • patriarchal landmine says:

            “really is it that disastrous if white men have a marginal disadvantage for a while, with all the privilege we already have?”

            we’re done here.

            feel free to give all your money away to the nearest “victim lacking in privilege.” I’ll just be over here, a white male, hoping to find another way to help men and boys abused by their mothers who never got the help they needed.

            Report comment

          • Mark Atwood says:

            I’ll just be over here, a white male, hoping to find another way to help men and boys abused by their mothers who never got the help they needed.

            If any of them express a desire to get into FLOSS sw dev as a career, send some of them my way.

            Report comment

        • Murphy says:

          Once it becomes acceptable to decide what results science is allowed to get based on political orthodoxy a lot more things get affected than just disadvantaged groups.

          Unrelated theories start getting judged based on whether they feel like they promote or pattern match to unacceptable ideologies.

          Also, as other have pointed out, crusades against reality rarely tend to hurt those actually in power and almost always get turned against the weakest or most unpopular.

          Report comment

    • Decius says:

      The idea that “We know that ‘scientific’ ‘research’ will show implications in our benefit” is normally not associated with the marginalized group.

      Report comment

      • Tom Bull says:

        I think you’re conflating science misreporting (which we all agree is bad) and institutionalised social justice movement (which the post author suggests is bad and I’m suggesting might not be the end of the world).

        Report comment

        • Anonymous says:

          As has been demonstrated with the likes of Ferguson, social justice identity politics isn’t just about marginal situations, but can easily get out of hand and turn to violence.

          Also purity spirals leading to people getting fired for opinions that were very recently normal, etc.

          Report comment

    • Anonymous says:

      If you refuse to acknowledge the actual cause of a problem, then you’ll keep looking and looking for someone to blame, while getting more and more upset that the problem isn’t getting better, thus getting more and more willing to use extreme measures (because what you tried so far clearly isn’t good enough). This is the kind of dynamic that leads to witch hunts (though for literal witch hunts it was caused by accidental rather than willful ignorance).

      Also, you’re assuming what you’re trying to prove – part of the argument is over whether certain groups are even meaningfully marginalized in the first place. (“Marginalized” being a term which tends to imply that the group is being treated unfairly).

      ————————————————————————

      In cases like race and IQ, some people try to say, “we can’t acknowledge the actual cause because it might increase prejudice”, but that only works as long as the people in power have the true knowledge in the back of their minds. The problem is when you have the next generation of kids being raised with the propaganda version of reality, they’ll tend to become true believers and start wrecking everything trying to fix the problem based on their faulty understanding. I guess the alternative is lies for the plebs, with only important people being told the truth, but that obviously has all sorts of problems, especially in this modern age of pleb empowerment via the internet.

      Report comment

    • Deiseach says:

      “Two wrongs don’t make a right”.

      ‘All we want is the right to love one another just like everyone else enjoys’ somehow became ‘if you refuse to take our wedding photos/bake our wedding cake, we’ll take you to court, sue you for as much damages as we can, and activist groups will whip up a firestorm of hatred towards you aimed at driving you out of business’.

      See the pizza parlour employee who said for a hypothetical “We wouldn’t cater a gay wedding” and what happened to them – including that foolish woman so carried away by self-righteousness she put up a stupid “So who’s coming with me to burn the place down?” tweet.

      Report comment

      • ‘All we want is the right to love one another just like everyone else enjoys’ includes both the legal right and the social support that heterosexual marriage gets.

        Trying to force a sudden change in social support may not be a brilliant idea, but I don’t have huge amounts of faith in my ability to predict what people will do.

        Sidetrack: How do you pronounce your name?

        Report comment

        • Deiseach says:

          If someone is claiming at one and the same time that it is no business of government or society at large what they do in their private lives and that government and society owe them social support to help them promote what they do in their private lives, I wish they’d make up their minds which it is they want.

          But my point was that “What harm could bending over backwards to promote a marginalised group possibly do, even in the worst-case scenario?” was that the answer was not self-apparently “Why, no harm at all, there is never a downside!”

          Pronunciation something along the lines of “day shock” 🙂

          Report comment

          • Vox Imperatoris says:

            If someone is claiming at one and the same time that it is no business of government or society at large what they do in their private lives and that government and society owe them social support to help them promote what they do in their private lives, I wish they’d make up their minds which it is they want.

            I am opposed to anti-discrimination laws (of any sort), but the charitable way to interpret the case for laws against LGBT discrimination is as follows. Here are all the “protected classes” in the United States, on the basis of which one is not legally allowed to discriminate:

            Race – Civil Rights Act of 1964
            Color – Civil Rights Act of 1964
            Religion – Civil Rights Act of 1964
            National origin – Civil Rights Act of 1964
            Age (40 and over) – Age Discrimination in Employment Act of 1967
            Sex – Equal Pay Act of 1963 and Civil Rights Act of 1964
            Pregnancy – Pregnancy Discrimination Act
            Citizenship – Immigration Reform and Control Act
            Familial status – Civil Rights Act of 1968 Title VIII: Housing cannot discriminate for having children, with an exception for senior housing
            Disability status – Rehabilitation Act of 1973 and Americans with Disabilities Act of 1990
            Veteran status – Vietnam Era Veterans’ Readjustment Assistance Act of 1974 and Uniformed Services Employment and Reemployment Rights Act
            Genetic information – Genetic Information Nondiscrimination Act

            Now, what is the actual reason for the fact that “sexual orientation” is not on that list, while all the rest are? Answer: prejudice against LGBT groups. The reason is not: “none of these laws are a good idea” (which is what I believe). If the general public believed that, these laws would be abolished.

            So as long as they are not grouped as a “protected class”, LGBT people correctly perceive that discrimination is being allowed toward them of a sort which is not allowed toward virtually any other similar group.

            Report comment

          • Anonymous says:

            @Vox Imperatoris

            Genetic Information Nondiscrimination Act

            Sounds like a biodeterminist lawsuit waiting to happen.

            Report comment

          • Anonymous says:

            @Deiseach

            Will you mind if I keep vocalizing it as “daisy”? 😉

            Report comment

      • Theo Jones says:

        I think one of the biggest o_o O_O woah things I’ve seen in mainstream politics is the idea that anti-discrimination laws should apply to artists (ie. photographers) deciding what works to produce. Its an extremely dangerous idea on free speech grounds. And it seems to be spreading in mainstream politics. Unlike some of others here I’m not per se opposed to anti-discrimination law. But that application of it is very dangerous.

        Report comment

        • FacelessCraven says:

          …artist here. I don’t understand what you’re talking about. Care to elaborate?

          Report comment

        • Muga Sofer says:

          I must admit, I’ve never heard of that (with the exception of casting.) It’s certainly within the overton window of internet (and, to a slightly lesser extent, offline) SJ that people should be *punished* for creating such art, but these punishments seem to be entirely extrajudicial.

          But then, I don’t really follow the art world. Has this seriously been attempted? Do mainstream individuals speak out in favour of it?

          Report comment

          • Muga Sofer says:

            Ah, of course, wedding photographers. And cake decorators, now that I think of it.

            Yeah, that’s extremely worrying.

            EDIT: Still, at least it’s based on accepting commissions, rather than a general policing of subject-matter; I think there’s a much stronger case to be made for policing which commissions are accepted by analogy to “black people can’t eat/shop here” stuff than for more general restrictions.

            Report comment

          • FacelessCraven says:

            @Muga Sofer – “Ah, of course, wedding photographers. And cake decorators, now that I think of it.”

            How about musicians?

            Just read the article, and a good chunk of the comment section. extremely worrying seems about right.

            The comment section, in particular, was enough to convince me that Clark has the right of it. We are not talking about accommodation and compromise any more; we are actively working as a society to make plurality and diversity impossible.

            Report comment

    • Dr Dealgood says:

      As for why this sort of thing bothers people, coming from someone in STEM but outside of tech:

      I just finished interviewing for PhD programs in molecular bio (my handle is a Mad Max reference, I’m not actually a doctor of any kind), and one thing I noticed is that the interview pool is invariably >75% composed of women. And when you look at the grad students, the postdocs, the lab techs, or even the younger (<50) PI's you see it's about 50-60% women. I'm certainly not complaining about this, honestly I strongly prefer this ratio to the sausage-fest that was engineering school, but one other bit did stick out at me.

      The push for more women in biology didn't seem to have lessened any for all of that. I sat through a speech, by a woman in the administration to our predominantly female audience, about how there are too few women in our field and more must be done to be inclusive / diverse / etc. so as to remedy that. Even when women are the clear majority they are still considered underrepresented and are still given extra resources and consideration as such.

      My worry is that this kind of study, and the reporting it attracted, is part of why this happens. We can never say "OK, women are (more than) represented and there is no longer any detectable bias: time to go home" because there is no longer any connection between the data and reality. Women will continue to be marginalized at 70% 80% and 90% of the positions in a field, and the whole time it will be the fault of guys like me who are just trying to do some science.

      Report comment

      • The Smoke says:

        I actually have a hard time believing that. Don’t they usually show graphs or numbers when giving a talk about such things? Are they avoiding that or is it that they refer e.g. to lack of women in top management positions?

        Report comment

        • Zorgon says:

          The graphs will almost certainly consist of “gender balance in STEM” rather than “gender balance in Biology”.

          I’ve seen this myself. I’ve seen the soft sciences do it, too. I’ve even seen it in tech, since so many management positions are held by women. I’ve sat in a conference room with 70% women and watched a speaker talk about gender bias in HR positions. Again, the numbers for the whole software sector come up, not the HR numbers.

          The problem is not the stats, the problem is the narrative. Nothing may contradict the Sacred Narrative, and all things that might do so must be cast into the Pit lest they encourage and embolden the evil neckbeard MRA misogynerds.

          Report comment

          • Deiseach says:

            Zorgon, there is a real – I won’t say problem, but a really existing thing – that the further up the ladder you go, the more senior the management position, the more power/authority/status, that it begins to skew male.

            I have seen it in my workplace. What you say about “a conference room with 70% women”? Sure, I could match that with the HR department at the organisation – mostly women – until you hit the Director of Services, who is a guy. Five Directors of Departments, one Acting Director, one C.E.O. – all top level positions, all guys.

            Same in the other departments: ground level up to about Grade Five – overwhelmingly female. Levels above that – noticeably male.

            Why that is, how that is, I don’t know. But it’s a fact. Even with efforts and policies to promote equality and diversity, if you’re talking about the boss of your boss, it’ll be a guy – and his boss will be a guy.

            Report comment

          • Zorgon says:

            As someone says below, it’s mostly a combination of Experience Fetishism with the Permanent Executive Class. Nearly all larger companies feature a specific subset of society that move in and out of executive positions and nearly all of those people are old men. The women in that group tend to be significantly younger, though.

            End result? Young men are locked out of those positions, just as mentioned. Not least since it’s now pretty much an open secret that advancement to middle management and moving to administrative positions (which often make a move to management much easier) in most STEM companies is much easier for women than men.

            Report comment

          • Yrro says:

            Really, the problem is the time delay. It takes 40-50 years of actual meritocracy and equality for women to rise to a representative number of senior positions. We could have passed that point twenty years ago and the metrics won’t show it yet.

            Report comment

        • Earthly Knight says:

          According to this article, women now make up 52% of PhDs in biology but occupy only 18% of full professorships. This explains the disconnect pretty straightforwardly– the administrator lecturing Dr Dealgood is concerned about underrepresentation of women among biology faculty, particularly senior faculty, but that underrepresentation is not reflected in Dr Dealgood’s cohort of students.

          Report comment

          • Dr Dealgood says:

            Yeah I sort of hinted at that by mentioning that there was a parity / majority in PIs under 50, rather than all PIs. The new faculty and current students are where I see women dominating.

            The issue with that is that senior faculty are, well, senior. Mostly senior citizens as well for that matter. Even if my incoming class was 100% women there wouldn’t be any change in the makeup of the senior faculty for at least a few decades. It’s a solution which cannot possibly solve the problem.

            Report comment

          • CandyceRogers says:

            Can’t some of that be explained by simple matters of age? You all are dancing around this point but no one seems to have come right out and said it. If in fact there are more women represented in today’s graduating classes than there were 30-40 years ago (when today’s senior executive group was in college), then doesn’t it hold that in 30-40 years we’ll see that effect become more noticeable in the upper ranks? Women are increasing, it’s just going to take a while for them to gain the experience and influence and status that today’s senior executive folks have now. IOW *it is already happening* and feminists can stop wringing their hands so much.

            Report comment

    • The Nybbler says:

      The absolute WORST thing? I don’t know if this is the absolute worst thing, but suppose the institutionalized Social Justice movement chooses as one of its “marginalized groups” a group distinguished by its ruthlessness, propensity for violence, and hatred for the culture in which the Social Justice movement is embedded.

      This so-called marginalized group takes advantage of this to work their way into positions of power through everything from rigged elections and tests to intimidation and murder of their opponents. All of this is either applauded, excused, denied, or ignored by the Social Justice movement because it is done by members “marginalized group”.

      Then once the “marginalized group” has enough power in their own right, they eliminate all their rivals (most particularly including the Social Justice leaders) and institute a genocidal and tyrannical regime.

      That’s among the worst things that could happen.

      (Any resemblance to real events occurring in certain European countries is probably intended)

      Report comment

      • Muga Sofer says:

        >This so-called marginalized group takes advantage of this to work their way into positions of power through everything from rigged elections and tests to intimidation and murder of their opponents.

        ?

        Report comment

        • Vox Imperatoris says:

          The Musselmen, the Mohammedans, creating Eurabia as we speak.

          Report comment

          • Muga Sofer says:

            I live in Europe. I got the reference to The Muslim Immigrants easily enough, I just can’t for the life of me get what the quoted paragraph is referencing (unless it’s just a hyperbolic prediction for the future.)

            Report comment

          • The Nybbler says:

            Yes, it’s a hyperbolic extrapolation. Not a prediction, I don’t think Europe is dumb enough to let it go all the way. The question was “what is the worst that could happen”, so extrapolation was called for.

            The real-life basis of the extrapolation is the apparent willingness of the countries involved to downplay certain crimes committed by refugees, because it would be politically incorrect to acknowledge them.

            Report comment

          • Zorgon says:

            “I just can’t for the life of me get what the quoted paragraph is referencing”

            I took it as a reference to numerous accusations of electoral fraud in majority-Muslim inner city areas in the UK. (Personally I tend to assume those specific accusations are isolated demands for rigour, since electoral fraud is ubiquitous and almost expected here. See: MI5 securing the 1992 general election for Thatcher.)

            Report comment

    • James Picone says:

      Marginalisation and discounting of male victims of rape and abuse, particularly when the perpetrator is female? IIRC we have a semi-occasional commenter around here who was shunned by his fairly-feminist friendship group after telling them he was raped by a woman. It’s not like male victims get taken seriously in the current culture, of course, but some strains of the social justice movement are very hostile to the idea.

      Hope the right people end up calling the shots; if the radfems are the ones with power, you wouldn’t want to be trans.

      Report comment

    • vV_Vv says:

      Where people deliberately or subconsciously misinterpret results to skew their implications towards benefiting a marginalised group?

      Like male nerds?

      Report comment

    • Kendall says:

      One concern I’ve seen raised by some women in STEM and the skeptic movement is that scaremongering about sexism within a field could actually put off women from entering it.

      Tell women that misogyny is rife in a particular career, that unequal treatment is likely to limit their success, and some might choose to work elsewhere, i.e. where there’s already a better gender balance.

      Celebrating improvements in equality, and the achievements of women who do enter and excel in male dominated fields, seems more productive to me.

      Report comment

    • Zorgon says:

      There are two problems with your idea here.

      1) There’s an implicit assumption that “white males” are a homogeneous group with evenly distributed privilege. That is certainly a common social justice concept, but also one that bears no relationship with reality.

      (Yesterday evening as I hobbled to my girlfriend’s flat I noticed a young White Straight Cis Male (TM) I’ve seen around town a few times sleeping in the overhang entrance to my town’s rather grandiose library. I don’t think he’s been homeless long, he has numerous bags and a big, high-quality sleeping bag, which is good as it’s February. Even so, I defy anyone on the face of the planet to tell me that guy is “privileged” by any meaningful definition whatsoever.)

      2) What exactly constitutes an acceptable amount of disadvantage to inflict on “privileged” groups purely for being members of that group?

      Are we just talking about having more difficulty in access to justice, education and abuse services? Actually, oops, all of those already exist for “white men”. OK then… direct negative bias in employment? Forced pay cuts in order to pay them the same as a woman in the same job class (regardless of experience or job description)?

      How about restricting their access to healthcare so women and minorities can have better treatment? (After all, women live longer so they need more healthcare over the course of their lives and besides, everyone knows men invent illnesses all the time.)

      OK then, how about making it so that only men are subject to the prison system and that women are instead put into hospitals or rehabilitation clinics? (After all, it’s not like women can be properly violent in the way a man can.)

      All sounding too reasonable? How about chemically castrating all men after they reach adulthood and provide sperm samples in order to “prevent rape”? Chromosome testing all babies and aborting all but a tiny percentage of male babies, thus ensuring future generations will not have to deal with the Male Scourge?

      It’s not like we want to put men in death camps, you understand. Just in controlled environments where they can live out their lives in relative luxury, kept from the possibility of hurting women and minorities.

      ***

      OK, I know the latter entries are coming from the psycho end of the radfem movement (although not as obscure as I’d prefer… far too many wannabe edgy teenage Tumblr idiots spout this stuff). I was just having fun. But my point was perhaps straightforward. If you inflict systematic disadvantage using regulatory means on a group, where exactly does the line of “acceptable” lie? Is a White Male Tax OK? How about quotas on lottery winners (since “white males” already won the genetic lottery)? How about we just have mandatory beatings every day for anyone white and male?

      Where does the line lie? What specifically about the demands of the more “moderate” SJWs makes them more reasonable than it would be to do any of these things to any other group of people, ESPECIALLY on the grounds of unselected traits like race or gender? Why is it suddenly OK if it’s the right race and gender?

      The answer, in case you’re wondering, is that it isn’t OK, it was never OK, it will never be OK, and no amount of shitty rationalisation is going to make it any better.

      Report comment

      • Sastan says:

        “Privilege” is just the current magic word being used to excuse racism and bigotry.

        Once upon a time it was that the jews were thought to control all the world. I mean, everyone knew that! Now it’s white men, who secretly control everything! Even the US, with its lilly white president…….

        We all know that “Those People” secretly control X. And therefore our hatred and persecution of Those People isn’t rampant bullying, but insurgent tactics against a superior foe! Any reaction by normal people is just the result of Evil Propaganda and Dirty Tricks by X. Any one of Those People who seems reasonable is merely trying to trick us with their (commie/man/jew)splaining!

        Report comment

      • Loyle says:

        “Even so, I defy anyone on the face of the planet to tell me that guy is “privileged” by any meaningful definition whatsoever.”

        I don’t know how meaningful it’d be, but were you comparing him to the bulk of society, or to others in his similar position? I can certainly imagine someone trying to argue that if he were black he’d be shooed away, have his stuff stolen or vandalized, and he’d be arrested or shot. Or maybe that he was less at risk to be in his situation in the first place?

        All I know is that the use of the word is based on “statistics~” and an individual anecdote isn’t going to change anyone’s mind.

        Report comment

        • Muga Sofer says:

          Yes, there’s something of a – I’m so sorry – Motte-and-Bailey going on there.

          Report comment

          • Zorgon says:

            There is certainly a motte-and-bailey going on, but it’s not me that’s using it. I’m more the one pointing at the big castle on the hill and saying “That! See! THAT!”

            Report comment

          • Muga Sofer says:

            Yes, that’s what I meant – the motte is “he’s better off in one specific way than someone else in the same position”, and the bailey is “he’s Privileged and everything in his life has gone perfectly”

            Report comment

          • Muga Sofer, I don’t think you’ve got the bailey right.

            It’s actually “he’s privileged, so anything that goes wrong in his life is too trivial to be worth concern when so many other people are suffering more.”

            Report comment

        • Cauê says:

          I can certainly imagine someone trying to argue that if he were black he’d be shooed away, have his stuff stolen or vandalized, and he’d be arrested or shot. Or maybe that he was less at risk to be in his situation in the first place?

          These are two very different hypotheses (for instance, the second one does not work in any way as a rebuttal to Sastan). Which one is the one that’s based on statistics?

          Report comment

          • Loyle says:

            The ideas of “Privilege” and “Systematic Oppression” are based on statistics, whether or not those statistics are honest when the subject is referenced.

            And those were just examples of possible arguments a person might make in response to the challenge. I don’t hear people talk of homelessness enough to know what arguments they would actually make. Just wondering whether “Others would have it worse in the same situation” would be something Zorgon considered people would suggest.

            Report comment

          • Loquat says:

            @Loyle –

            Having spent time around social-justicey parts of the internet, I can guarantee you one of the main counterarguments would be some variation of “but that homeless white guy would have it so much worse if he were homeless and black/gay/female/etc”.

            Report comment

          • Anonymous says:

            @Loquat

            Which is a totally absurd argument, as it can be turned around with zero effort: “That black/gay/female person would have it so much worse if they were homeless”. There are multiple disadvantages that exist, so… the one you didn’t mention must be worse? Huh? What?

            Report comment

          • Loquat says:

            @Anonymous

            Well, the theory is that being white gives you a lot of privilege no matter how many other problems you may have. People of any color, gender, etc, can be homeless, or addicted, or scraping by on minimum wage, or whatever, but if you’re a straight white male at least you don’t ALSO have to deal with racism/sexism/homophobia.

            As for the question of what good is *accomplished* by using Privilege Theory to prove that an impoverished white person isn’t quite at rock bottom of the social ladder… I really couldn’t tell you.

            Report comment

        • Zorgon says:

          >> All I know is that the use of the word is based on “statistics~”

          That’s the thing, though – it really is not. The definition of the word is (very loosely) based on statistics, but the use of it is not.

          Privilege as defined by social justice-y academics is a somewhat vague concept that seems to start out as “lack of structural oppression” and then in usage somehow manages to become a set of coherent advantages provided to one specific group of people… regardless of whether or not those advantages are meaningful, relevant or even existent to the members of that group.

          The problem with this approach is twofold. Firstly, it’s blatantly original sin, since you don’t normally get to choose whether to be assigned male at birth and you don’t get to choose your skin colour. Screw that for a game of soldiers.

          Secondly, it outright erases every single person in that group for whom those “privileges” do not exist. Probabilities are not actually a part of everyday life. “A black person is more likely to be homeless than a white person” doesn’t change the fact that the guy I mentioned upthread is clearly embarking on a period of homelessness at all. Indeed, by invoking that very thing you’ve both erased his clear (and real) suffering AND effectively blamed him for having it because clearly he’s just not enough of a Victim Class to deserve pity.

          There is no context in which that guy is “privileged” by any meaningful definition. “Statistically” is not a real-world context. It does not affect his life. It doesn’t get him off the library steps and it won’t stop the police moving him on.

          And our dear friends in SJ? They actively encourage the idea that resources that might be spent to prevent that specific, real-world person’s suffering should instead be spent on things that benefit their chosen Victim Classes.

          The reason an individual anecdote isn’t going to change anyone’s mind is not because it does not affect the statistics, meaningless as they are to real-world cases. It’s because an anecdote about a Cis White Male in clear difficulty does not fit the Victim Narrative. An anecdote about a woman, or a black person, or any other Victim Class would suddenly become a moral imperative to SJ people. The sole reason this one does not is because SJ people are incapable of even the slightest hint of the “intersectionality” they so fervently profess. They’re hypocrites, all of them, political hipsters playing at saving the world.

          Report comment

          • Loyle says:

            I was with you til that last paragraph there. Like, yeah regardless of how the population finds itself spread out, we should treat individuals with regards to their circumstances.

            I’m not sure how you’ve experienced things, but I’ve noticed that people in general have a really hard divorcing their understanding of the world from established memes, unless they have a specific desire for the memes to not be true. And it’s not just when talking about how white men have it good/bad compared to victim classes that people are willing to make excuses how an individual story shouldn’t change how they think of the narrative.

            This isn’t new. It isn’t right, but it isn’t new.

            Report comment

          • Anonymous says:

            Agreed. More generally, there are two related mistakes made by Social Justice folks. One is to notice only a tiny handful of the factors that can advantage a person but ignore all the rest. So race counts but intelligence doesn’t. Gender counts but attractiveness doesn’t. Sexuality counts but social ability doesn’t. And so on. The disturbing thing is that I think in many ways they miss the factors that have the most impact. Personally, as a straight white male with a reasonably high IQ, I would rather become a black female lesbian and keep the IQ, than become low-IQ and remain straight, white and male.

            The second mistake is more ridiculous – treating averages as absolutes. If some disadvantage is suffered by 40% of people in Group A and 80% of people in Group B, that does not equate to “people in Group B have it twice as hard as people in Group A”. A person with the disadvantage has the disadvantage. A person without the disadvantage doesn’t have it. Being a member of a group that is less likely to have a disadvantage does nothing for you if you actually have the disadvantage. Oh, sure, whatever, it might be worse to be in a group that more commonly has the disadvantage, because people expect you to have it and as such treat you differently. On the other hand, having a disadvantage that is rare for your group means there will be fewer support networks and so on in place – economies of scale and all that. I don’t think it is at all obvious that the former of these effects is greater in impact than the latter, and segueing into claiming it is only when you’re called out on making the average-absolute conflation is disingenuous and classic motte-and-baileying.

            Report comment

          • If some disadvantage is suffered by 40% of people in Group A and 80% of people in Group B, that does not equate to “people in Group B have it twice as hard as people in Group A”. A person with the disadvantage has the disadvantage. A person without the disadvantage doesn’t have it.

            This reminds me of a case involving Canada’s rules for accepting refugees. The rules say that if you face persecution in your home country, you can apply for refugee status. A woman from the United States was facing thirty years in prison for doing something that is perfectly legal in Canada (consensual sex with a 16-year-old). Some Conservatives responded by saying, “I can’t believe people from first world countries are using this refugee rule, they have it so much better than people from other countries, they’re so privileged, blah blah blah.” They were apparently blind to the fact that this particular person, despite having the privilege of coming from a first world country, was facing three decades in prison.

            Report comment

        • Zorgon says:

          Also, I should probably mention that white males under 30 are the most likely cohort to be homeless in my country. But that’s still completely irrelevant.

          Report comment

    • rockroy mountdefort says:

      the worst thing that could happen is that we’d all have to spend our lives reading internet comments dripping with snideness and bad faith, like yours

      >What’s the absolute worst thing? That people become more conscious of disadvantaged groups and implement positive action in their favour?

      some people think it matters whether the “disadvantages” that people become more conscious of actually exist, and aren’t actually entirely made up

      crazy, out-there, meninist stuff here, I know

      >That groups who have had hundreds or thousands of years of institutionalised advantage are now marginally disadvantaged?

      hurting the living to punish the dead doesn’t seem like a great approach to anything m99

      Report comment

    • “Really, what is the worst that could happen if… people deliberately or subconsciously misinterpret results to skew their implications towards benefiting a marginalised group?”

      Women are not a marginalised group; they are both a demographic majority and an electoral majority (70.4 million women versus 60.7 million men voted in the 2008 presidential election.) We have many special laws that protect women, and only women, scholarships that are given to women, and only women, special educational opportunities that are give to women, and only women, etc.

      Report comment

      • FacelessCraven says:

        …a third to a tenth the suicide rate, massively lower violence rates, obvious favorable discrimination in legal outcomes, better health outcomes…

        Report comment

        • Liskantope says:

          …a third to a tenth the suicide rate

          I wish this statistic weren’t constantly being quoted without any mention of the fact that women attempt suicide at a much higher rate than men do.

          Report comment

          • FacelessCraven says:

            @Patriarchal Landmine – True, kind, necessary?

            Report comment

          • honestlymellowstarlight says:

            The rankings, as I see it actually implmented, is

            Kind >>> True > Necessary.

            Report comment

          • Muga Sofer says:

            I think very few comments meet the standards of evidence implied in the “true” description, so we’re left hanging on to “kind” and “necessary”. And Necessary is very ambiguous.

            Report comment

          • anon says:

            @honestly
            Mostly because reporting a comment for being untrue goes against the culture unless its obvious intentional misinfo.

            Report comment

          • honestlymellowstarlight says:

            Commenters here accept the ambiguity and expansive range of truth and necessity, but not kindness, IMO. But then again, I am a nasty troll.

            Report comment

          • FacelessCraven says:

            @honestlymellowstarlight – “Commenters here accept the ambiguity and expansive range of truth and necessity, but not kindness, IMO. But then again, I am a nasty troll.”

            I don’t think kindness has anywhere near the ambiguity of truth or necessity. I also think it is really valuable, and this culture’s attachment to it is a big part of what makes it worth hanging around. The whole wide internet is available for screaming at people, but this corner of it is for treating people like reasonable humans, even if we disagree with them. That courtesy is what protects the crazy right-wingers here, so it’s only fair that it should protect the crazy left-wingers as well.

            I don’t actually disagree with the substance of PL’s statement, but I think they’re making their point like a jerk, and think that they would do better to try for either a bit more kindness or a shed-load more truth and necessity.

            Report comment

          • honestlymellowstarlight says:

            @FacelessCraven
            It is nice to see examples of what I’m talking about. I see it more as a sort of AIDS, but for social interation. Stress is good for the body, etc. And the thought of needing someone to “protect” me from what I say gives me hives, it reads as hideously publically weak.

            And of course you don’t disagree, you thought what he said was mean, as if the truth was always cute and cuddly. That’s sort of my point, true/kind/necessary is really only the one in the middle and pretending otherwise obscures things. Unless someone disagrees with me, which I would be interested to hear.

            Report comment

          • Vox Imperatoris says:

            @ honestlymellowstarlight:

            No, the problem is that he said what might have been a legitimate point like an asshole.

            Report comment

          • honestlymellowstarlight says:

            @Vox Imperatoris
            Yeah, and saying things “like an asshole” is assumed to be intentional and/or bad, like we’re all running off the same social scripts and emotional vunerablities.

            I point out I have no disagreements with Scott’s implementation of the policy, only how it’s cited by commenters against other commenters.

            Report comment

          • anonymous says:

            If it gives you the hives, you welcome to quit trolling here and go back to 4chan.

            Report comment

          • honestlymellowstarlight says:

            @anonymous
            Was that two of (true, kind, necessary)? lolololol

            Report comment

          • hlynkacg says:

            @HMSL

            I’m going to wager on true and necessary.

            Seriously though, you need to step up your game if you’re going to keep hanging around here.

            Report comment

          • FacelessCraven says:

            @HMSL – “It is nice to see examples of what I’m talking about. I see it more as a sort of AIDS, but for social interation. Stress is good for the body, etc.”

            Charity and politeness are like AIDS?

            “And the thought of needing someone to “protect” me from what I say gives me hives, it reads as hideously publically weak.”

            politeness and charity don’t protect you, they protect the conversation. I would like to be able to talk to people I disagree with. people being jerks makes that harder.

            Report comment

          • HeelBearCub says:

            @FacelessCraven:
            Did the comment that started this sub-thread get deleted? Or is just in the wrong place?

            Report comment

          • jaimeastorga2000 says:

            @HeelBearCub: Deleted.

            Report comment

          • My understanding is that *attempted* suicide is often a cry for help. The fact that women attempt suicide more often, but men *actually* suicide far more often, suggests that most female suicide attempts are cries for help — that deep down, perhaps unconsciously, they’re hoping to get help from those around them rather than to actually end their lives.

            I suspect that you don’t see so many of these attempted-but-unsuccessful suicide attempt among men because, unlike women, they have no expectation that anyone will have any sympathy for them and offer any help.

            Report comment

          • Liskantope says:

            unlike women, they have no expectation that anyone will have any sympathy for them and offer any help.

            Possibly. Or, men are socialized not to ask for help, to hide their emotions, etc. (Only a slight variation on what you said.)

            Report comment

          • The Nybbler says:

            One of the primary ways men are socialized not to ask for help and to hide their emotions is to give them abuse when they ask for help or show their emotions. So unless the society has changed between the time this socialization took place and the later-on when they might consider a suicide attempt as a cry for help, those are two sides of the same coin.

            Report comment

        • Muga Sofer says:

          >[women have] a third to a tenth the suicide rate …

          But somewhat higher suicide *attempt* rates.

          Report comment

          • John Schilling says:

            We unfortunately use the word “suicide attempt” to refer to two very different behaviors, one of which is more common among men and the other more common among women.

            Report comment

          • Muga Sofer says:

            In my (admittedly limited) personal experience of depression, it’s more of a spectrum.

            EDIT: Of course, that same experience would lead me to disclaim that neither suicide attempts nor successful suicides necessarily reflects some sort of general factor of “quality of life” or “oppression”.

            Relevant: http://slatestarcodex.com/2015/01/15/depression-is-not-a-proxy-for-social-dysfunction/

            Report comment

          • FacelessCraven says:

            @Muga Sofer – “In my (admittedly limited) personal experience of depression, it’s more of a spectrum.”

            Perhaps so, but the disparity in success rates seems to indicate a discontinuity in that spectrum.

            “Of course, that same experience would lead me to disclaim that neither suicide attempts nor successful suicides necessarily reflects some sort of general factor of “quality of life” or “oppression”.”

            Why does the question of who kills themselves three times as often not reflect a general factor of quality of life or oppression, but Git PRs or women in tech or any other of a hundred oft-debated issues do?

            Report comment

          • Muga Sofer says:

            Because suicide is more often caused by psychiatric disease than by anything that could be considered “oppression”.

            This isn’t universally true – oppressed groups may find it harder to get medication (although I don’t believe that’s true in this case), and trauma can lead to depression and subsequent suicidality – but in general, it seems to be a public health concern.

            Certainly, both these figures are alarming from a public health standpoint, if only because they naively imply that it should be possible to lower women’s suicide attempt rates to match men’s, and lower men’s suicide success rates to match women’s. But they don’t in any way imply discrimination.

            (Biased GitHub referrals, on the other hand, are a form of discrimination. Albeit a small one.)

            Report comment

          • Zorgon says:

            “But they don’t in any way imply discrimination.”

            If the shoe was on the other foot you would be screaming to the high heavens that it ABSOLUTELY DID IMPLY DISCRIMINATION!!!1!

            The only reason SJ people get to enjoy this particular form of prevarication is because there is no good research into exactly why this specific disparity in suicide attempts/completions exists. And there won’t be as long as that sort of research is controlled by SJ people with an iron fist.

            But as a male with a history of depression… I can posit a strong theory of my own, and it most certainly has a lot to do with discrimination.

            Report comment

          • Muga Sofer says:

            > as a male with a history of depression… I can posit a strong theory of my own, and it most certainly has a lot to do with discrimination.

            As a male with a history of depression myself, I’m more inclined to hypothesize that this is about *depression*, not oppression – perhaps to the point of bias, if I’m honest.

            To be clear, when I thought that men *were* more likely to be suicidal, a number of obvious explanations presented themselves. I just don’t think that’s the case.

            >If the shoe was on the other foot you would be screaming to the high heavens that it ABSOLUTELY DID IMPLY DISCRIMINATION!!!1!

            I admit, if *both* statistics pointed to women having it “worse off” suicide-wise, I probably would be inclined to favour the oppression hypothesis.

            As it is, however, I think that the fcat I’m not crying “DISCRIMINATION!!!1!” about the fact that women have much higher levels of apparent suicidality is evidence enough that I wouldn’t do it if men attempted suicide more.

            Report comment

          • Zorgon says:

            >As a male with a history of depression myself, I’m more inclined to hypothesize that this is about *depression*, not oppression – perhaps to the point of bias, if I’m honest.

            Likewise… and more specifically about differential access to mental health services.

            Report comment

          • Agronomous says:

            @Zorgon:

            If the shoe was on the other foot you would be screaming to the high heavens that it ABSOLUTELY DID IMPLY DISCRIMINATION!!!1!

            Can all of us agree to stop mind-reading? I see way too much of it around here lately, even from people whose positions I largely agree with (like Zorgon). It’s pretty much the opposite of charity, and tends to drag discussion down to personal attacks and defensiveness.

            “If the shoe was on the other foot, wouldn’t you be claiming that it DID imply discrimination?” would be a slightly less-inflammatory, but still objectionable way to make the point. The only way I can think of to put it acceptable is to go directly to meta and ask for general criteria for when something indicates discrimination.

            Report comment

          • FacelessCraven says:

            @Muga Sofer – “Because suicide is more often caused by psychiatric disease than by anything that could be considered “oppression”.”

            You are probably correct, but I’m pretty sure you can’t actually prove that. Further, I’m pretty sure that the claim “male suicide is proof that men are oppressed” has a lot more actual evidence behind it than a great many claims frequently advanced by the Feminist movement, starting with the “women in tech” narrative. Certainly it is far more defensible than the Wage Gap, which is prevalent enough that I would not publicly contradict it under my actual name for fear of professional consequences.

            “Certainly, both these figures are alarming from a public health standpoint, if only because they naively imply that it should be possible to lower women’s suicide attempt rates to match men’s, and lower men’s suicide success rates to match women’s.”

            …That would require a recognition that the fact that men kill themselves three to ten times as often as women is a problem worth paying attention to, a position I’ve never heard advanced by anyone but filthy MRAs and people who made the mistake of listening to them. Speaking personally, when I was presented with the actual suicide statistics about a year and a half ago, I was initially convinced that they were false, since I “knew” that the female rate was higher and that suicide was primarily a women’s issue. This despite the fact that, at the time, I’d spent several months deeply depressed, too miserable to sleep barring extreme exhaustion, and fantasizing about how nice it would be to just not be alive anymore, and how I could make that happen reliably and with a minimum of mess.

            For what it’s worth, I do not think men are actually oppressed. I think life is painful in enough ways that every individual or group can be confident that their agonies are personal and unique. I do think, though, that women have an effective lobbying group, and men simply do not. Consequently, women’s problems receive attention and response in a way that men’s issues do not.

            Further, it seems obvious to me that Feminism, as a broad movement, is entirely happy with that state of affairs and actively works to maintain it. I find that contemptible.

            Report comment

          • Nita says:

            @ FacelessCraven

            I “knew” that the female rate was higher and that suicide was primarily a women’s issue.

            I’m sorry you were misled that way. Is that a typical misconception? I’ve considered myself a feminist as long as I’ve had any ideological affiliation at all, and I haven’t believed anything like that at any point in time. On the other hand, I’m not American, so…

            MRAs who actually talk about men’s rights/issues as such (as opposed to how feminism is the root of all evil) are doing good work, IMO. It’s unfortunate and wrong that many feminists have come to conflate men’s rights with anti-feminism and even misogyny. But some people in both movements tend turn every issue into a battle in the Great Gender War, and I’m not sure how to change that.

            Report comment

          • FacelessCraven says:

            @Nita – “Is that a typical misconception?”

            I have no idea. Looking back on what I remember of my beliefs at the time, I wasn’t under the impression that it was a particularly Feminist issue, just that it was a problem that hit women more. All the ads I saw for depression/SSRIs seemed to have a female, examples of tragic suicide in media seemed to be women, where male suicide usually came across as justice or an honorable way to die. I think it would be pretty easy to come up with a just-so story about how this is all society oppressing men, but that’s all it would be. To be clear, I am quite confidant that neither the suicide rate nor my impressions are created by Feminism, but it seems pretty clear that the movement as a whole isn’t real interested in helping either.

            “MRAs who actually talk about men’s rights/issues as such (as opposed to how feminism is the root of all evil) are doing good work, IMO.”

            That’s generous of you. Karen Straughan seems like that sort of person (the little I’ve seen of her material seemed fairly reasonable, if contrary to all received wisdom), but a lot of the other ones in the movement seem intent on simply gender-swapping feminism, including all the flaws. I think they raise some good points, but the movement as a whole seems pretty self-defeating.

            “It’s unfortunate and wrong that many feminists have come to conflate men’s rights with anti-feminism and even misogyny.”

            My guess is that as long as society interprets things according to the general oppression framework, this is inevitable. If everyone is either a victim or an oppressor, and if victims’ suffering matters more than that of oppressors, the only way to live is to seize victim status. That’s what the MRAs are trying to do, it seems to me. The framework isn’t supposed to devolve into oppression olympics, but it seems clear to me that it does more often than not, and that no one has a way to fix it. The model is fatally flawed.

            Obviously others disagree, though, so I guess we’ll see how it goes.

            Report comment

          • Orphan Wilde says:

            FacelessCraven –

            I suspect it may be true that “men have it worse”. At least in the first world nations. I don’t think it matters that much.

            I play it up, however, because when you argue it effectively, people are far more willing to dismantle the oppressed-oppressor dichotomy (because they are more willing to dismantle the dichotomy than to treat men as oppressed, which is… interesting in a rather alarming way, but at least works for my ideological objectives of getting people to treat people as individuals rather than shoving them into classes).

            Report comment

        • FacelessCraven says:

          @Liskantope – “I wish this statistic weren’t constantly being quoted without any mention of the fact that women attempt suicide at a much higher rate than men do.”

          @Muga Sofer – “But somewhat higher suicide *attempt* rates.”

          …An open question to either of you: I put forward that men have three to ten times the suicide rate of women, and you both immediately reply that the *attempt rate* for women is higher. This is entirely true. Since we all agree on the evidence, could either or both of you explain your interpretation of it? Do you hold that a smaller disparity in suicide attempts outweighs a *larger* disparity in successful suicides?

          Report comment

          • Muga Sofer says:

            >Do you hold that a smaller disparity in suicide attempts outweighs a *larger* disparity in successful suicides?

            No. I just think it suggests that this is a much weaker example than it appears to be in the absence of context.

            The obvious interpretation is that men attempt suicide in more immediately-lethal ways, probably for the same (societal?) reasons they tend to commit murders in more immediately-lethal ways. This could reasonably be considered gendered issue, much like the fact that many more men tend to die in violent crimes.

            On the other hand, the obvious interpretation is that this implies whatever causes suicide is more common in women; which, considering the statistic in isolation seems to imply the opposite is true, is a fairly important point. This could *also* be considered a gendered issue.

            At least, I think it is? I used to quote that statistic too, but stopped when I learned it was misleading.

            Report comment

          • Cauê says:

            This is the “they’re not equally competent at it” interpretation, and I really don’t think it’s the obvious one, compared to “they’re not attempting the same thing”.

            Report comment

          • Liskantope says:

            I think it’s unclear without further research on my part what the best interpretation is. I agree that there may be interpretations where different intentions behind suicide attempts are correlated with gender. But I do believe one factor at play here (admittedly not the only factor) is that men choose more violent means (for instance, guns) to attempt suicide, and these methods have a much higher chance of success. That in itself would tell us something interesting about gender and violence, although it’s not the main point.

            The main point is that merely saying “men have a higher suicide rate” in most contexts I see it quoted for (appearing to be arguing that men are more oppressed than women) seems to imply that men have it worse because they are more frequently driven to suicide. The fact that the female attempt rate is much higher at the very least makes that implication fairly dubious. Of course, the fact that more men succeed at suicide still does in itself indicate a disadvantage for men, but only inasmuch as it’s better to be suicidal but unable to kill oneself than to actually die of suicide.

            Report comment

          • HeelBearCub says:

            @Muga Sofer/@Liskantope/@FacelessCraven:
            Shouldn’t we at least be willing speculate a bifurcation of near end of natural life suicides vs. others?

            What I’m trying to get to here are long planned suicides vs. impulsive ones brought about by short term peaks in stressors. I don’t know what the split is, but I’m guessing planned suicides have a much higher rate of success across the board.

            The impulsive ones would seem to depend on the immediate availability of means that lead to success.

            Those are all more questions than answers, obviously.

            Report comment

          • John Schilling says:

            Planning is not the issue. Just about everybody has the means to kill themselves very reliably on almost no notice – jump out of a tall building if you’re in a city, else drive a car into a solid object at full speed. Just about everybody has the means to conduct a “suicide attempt” that is very likely to fail on short notice – barbituate overdose if you’ve got some, else slit your wrists (across, not lengthwise). And I think most people understand that jumping from a ten-story building almost always makes you dead whereas downing a handful of pills usually doesn’t.

            Intent is the issue. Most suicides reflect an unambiguous intent to die. Most suicide attempts do not.

            Report comment

          • Cauê says:

            But I do believe one factor at play here (admittedly not the only factor) is that men choose more violent means (for instance, guns) to attempt suicide, and these methods have a much higher chance of success.

            Surely choosing a method with a higher vs. lower chance of success has something to do with the intentions behind the attempt.
            (it’s intuitive to me that it should have much more to do with that than with gender, being such a big decision)

            Report comment

          • Vox Imperatoris says:

            @ John Schilling:

            Both male and female suicide attempts tend to be irrational and impulsive.

            By the time you’ve gotten to the top floor of the Empire State Building (or whatever), you are very likely to have reconsidered. Not to mention that it’s not that easy to get to the roof of most tall buildings.

            One big difference is that men are much more likely to own guns, which are much more likely to work on short notice.

            I think one of the major reasons women choose poison is not that they know it doesn’t work, but that poison in their minds offers a painless way out. (Often, what they choose is not painless at all, but they think so.)

            I’m not denying the phenomenon of fake suicide attempts, but I’m hardly sure that you can explain all of the difference that way.

            Report comment

          • Chalid says:

            Right, if the attempt vs success difference was primarily due to intention, you should see little effect of gun ownership rates on suicide success rates (for example). But of course you do see such an effect.

            Report comment

          • Nita says:

            @ Vox Imperatoris

            Yes.

            Also, they expect poison to be cleaner — you fall alseep and don’t wake up, your corpse doesn’t look too revolting, cleanup is easy. Choosing to splatter your blood and brain all over the carpet, or your guts all over the sidewalk, feels ugly and self-centered in contrast.

            Report comment

          • One argument I’ve seen for why women prefer pills for suicide attempts is that women are strongly acculturated to avoid making messes– and, I’d say, against looking ugly, which eliminates hanging.

            I don’t have a feeling for how much less effective pills are compared to shooting oneself, though obviously there are plenty of stories about people being taken to a hospital and having their stomach pumped. On the other hand, there are warnings all over the place about risks associated with pills.

            I’m going to go with the assumption that people aren’t reliably competent, and especially not about things they don’t usually do, so I’m not going to judge the seriousness of a suicide attempt by the effectiveness of the method if the method is at all plausible.

            Report comment

          • John Schilling says:

            But of course you do see such an effect.

            Right. We see gun-free Japan having twice the suicide rate of the United States, with Europe being roughly in the middle. Or was that not the effect you were referring to?

            Bringing guns into a suicide debate, or really any debate where they aren’t needed, tends to be mind-killing.

            And they aren’t needed here. Even if we take out every suicide ever attempted or completed with a gun, the effect under discussion remains overwhelming. Women, at least in the United States, are more than twice as likely as men to “attempt” suicide using methods that are unlikely to actually kill them. Men, again in the United States but taking the gun-wielding men (and women) entirely out of the equation, are more than twice as likely as women to actually kill themselves using highly lethal techniques. The relevant methods, mostly handful-of-barbituates and hanging respectively, with a side order of wrist-slitting and jumping and an unknown number of deliberate car crashes, are equally available to all genders.

            Please to explain this without any variation on “it was the deadly guns!”. Or we can derail the thread with yet another thousand posts on guns, which I thought we were done with and isn’t relevant to what we were talking about up to now.

            Report comment

          • Pete says:

            I tried to kill myself using drugs when I was a teenager (I’m male). I genuinely wanted to die. I regretted it soon afterwards. I told no-one for years (so I don’t believe it was a cry for help).

            Why drugs not another method? Taking pills isn’t as scary as other methods. I may have wanted to die, but I can imagine the pure, immediate terror jumping off a building and the thought is unpleasant. Hanging sounds extremely (and importantly immediately) painful. I had no access to a gun, but if I had, I imagine the immediacy of the pain and the death would have stopped me using it.

            Taking pills causes no immediate pain, and our brains are very good at discounting pain that will happen even in 20 minutes, compared to pain that will happen right this second. For me at least, wanting very badly to no longer be alive, didn’t reduce my fear of death, or pain, so I took the easier option.

            On another point, how do we know that more men commit suicide? Because we have actual dead bodies and causes of deaths. How do we know that more women attempt suicide? Because they tell us in surveys? Is it hospital reports? Surely an obvious potential answer for the discrepancy is that men attempt suicide as/more often as/than women, but admit to it less for whatever reason. Actual physical evidence (dead bodies) are surely better evidence than human testimony.

            Report comment

          • John Schilling says:

            @Nancy: The lethality of suicide rate by drug overdose is roughly 11%, and that’s almost certainly an overstatement as many of the failures will be reported as accidental overdoses (or not at all). This study suggests 2% lethality for suicide-by-drug, but is admittedly estimating the nonfatal suicide rate from ER admissions.

            Lethality of suicide-by-gunshot is 92%, from the first source. The most common non-gunshot methods of actual suicide were hanging and carbon monoxide poisoning, both at 78% lethality.

            Suicide attempts by drug overdose(*) and wrist-slitting are an entirely different order of thing than suicide attempts by gunshot or asphyxiation. And I do not think that this is terribly obscure; as you note, “take handful of pills, wake up in hospital with lots of people caring for you” is a cliche at this point.

            * Not counting premeditated physician-assisted suicide.

            Report comment

          • John Schilling says:

            @Pete: We know that women attempt suicide more often than men in part because they tell us about it, yes, but they are telling us about it from hospital beds where they were brought after having rather obviously inflicted serious injury on themselves. Pretty much every count of suicide attempts I have seen, starts with ER visits for self-harm, and those are pretty unambiguous.

            There may be a marginal difference in the rate with which men and women will pass off a weak suicide-by-overdose attempt as a merely reckless accident, but I believe that by the time you get anywhere close to killing yourself that way you are also scheduled for three days with an overworked psychiatric resident who will have a say in how it gets recorded and has heard all the usual stories. Same goes for passing off a failed suicide-by-gunshot as an accident.

            Out of curiosity, was your case recorded as a suicide attempt at the time?

            Report comment

          • Adam says:

            Men are more likely to present risk factors unrelated to chronic mental illness, like being war veterans, recently released prisoners, even unemployed (but not as much as a few years ago), though as far as I can tell, the effect size is so massive it does not even come close to disappearing if you control for these things.

            Report comment

          • Chalid says:

            @John Schilling

            So you don’t want to discuss the guns, but only after you write much more about them than I did, and in a *far* more inflammatory way, too. But ok, I guess that’s how you communicate.

            I actually thought it was a point of general agreement that guns and suicides were associated; it’s one of the conclusions in Scott’s “Guns and States” post and I recall many comments discussing the claim in a generally supportive way (though I admit I did not read *all* the comments in both threads). If it were true, then it ought to have significant implications for how we can interpret intention in suicides, should it not?

            If guns and suicides are related, as Scott and most of the commentariat seemed to think, it is absolutely relevant to this discussion, and frankly adds far more information than just handwaving about the availability of other methods, speculating about suicides’ mental states and expertise in suicide techniques, etc.

            If you don’t believe in such a link… well, lots of people do, and maybe having a connection pointed out will inform their thinking.

            Please believe me when I say that I had no intention of turning this into a discussion about gun control.

            Report comment

          • Jiro says:

            I actually thought it was a point of general agreement that guns and suicides were associated; it’s one of the conclusions in Scott’s “Guns and States” post

            Gun suicides are important with respect to gun deaths, but unimportant with respect to suicides. (Where “important” means “significantly affects your conclusions about”.) This is not contradictory.

            Report comment

          • John Schilling says:

            @Chalid:

            1. There is a universally-acknowledged and strong correlation between guns and suicide by gunshot. Guns are the preferred suicide method of Southern-ish men who own guns and want to commit suicide. The correlation between guns and suicide in general, if it exists at all, is much weaker and likely to vanish in a fog of cofounders. And Scott deliberately pulled away from that discussion at the “guns correlate with gun suicides” level to focus on homicides.

            2. A one-sentence drive-by of the form “guns correlate with badness” will get just as inflammatory a response as a one-sentence “feminists correlate with badness”, for about the same reason. I would rather not see either one here. I would not expect either to go unchallenged.

            3. How is the alleged and unquantified “link” between guns and suicides relevant to the discussion of the gender differential in the rate of suicide attempts vs. actual suicides, when that differential remains huge even when you eliminate all gun suicides from the dataset, or even when you look at gun-free Japan? You have a hypothesis that does not even begin to explain the anomaly at hand. What else do you have?

            Really, it looks like you are making the same forest-for-trees kind of mistake as the promoters and even authors of that github study. There is a substantial anomaly screaming from the dataset – deanonymized submitters are less likely to have their code accepted, women are more likely than men to attempt suicide but less likely to succeed. But rather than try to explain that, they focus on a relatively minor differential and shout “it’s the sexism!”, you look at a small differential and shout “it’s the guns!”.

            Report comment

          • John, I had no idea the effectiveness of drugs for suicide was so low, and I do have a lot of random information.

            Report comment

          • Adam says:

            Anything that has to be digested gives you a pretty big window in which to voluntarily or involuntarily spew it back up before that happens. If people tried to poison themselves by injecting frog venom or drinking bleach, I’m sure it would work a lot better.

            Report comment

          • Chalid says:

            @John Schilling

            To 1 and 2: read the conclusion of Scott’s “Guns and States” post. One of his points is that “an Australian-style gun control program that worked and had no side effects would probably prevent about 2,000 murders in the US. It would also prevent a much larger number of suicides.” And if you search “suicide” in the comment thread you’ll find that a pretty large majority of comments that take a stand seem to believe that guns increase suicide, including one by Scott. Note the study by Alex Tabarrok too. Maybe you disagree but I do think you are the minority opinion here.

            3: The relevance is this. You have made the claim that, in general, people who succeed at suicide *really mean* it, and those that fail at suicide don’t. If that were true, then you would not expect the “trivial inconvenience” of finding a reliable non-gun method would affect suicide rates. And yet (most people believe) it does. Note that I was replying in support of Vox, who was making this point except without data.

            I don’t think a gun/suicide link explains the male-female differential and I have no idea why you think that I think that. It merely makes a particular suggested mechanism behind the differential less likely.

            Report comment

          • Cauê says:

            Right, if the attempt vs success difference was primarily due to intention, you should see little effect of gun ownership rates on suicide success rates (for example). But of course you do see such an effect.

            What we’d be interested to see here is gun ownership having an effect on male vs. female suicide rates. Do we see that?

            Now, I just eyeballed Wikipedia lists, but that doesn’t jump out at all. The US and UK have the same male/female ratios, for instance.

            Report comment

          • Chalid says:

            @Caue

            Whether one “needs” to see that depends on what point one is trying to make with the data.

            I am merely making the point that if trivial inconveniences affect suicide rates, then it is problematic to attempt to infer much about intent from success rates. No male/female discrepancy is particularly required for this point.

            @everyone

            I suppose I should explicitly say that I have no idea what the “real” cause is, and I am absolutely not trying to darkly hint anything. if you look at the abstract Nancy Lebovitz found, you see “Theories that attempt to explain this finding focus on gender differences in suicidal intent, socialization, emotions, interpersonal relationships, orientation and access to methods, and neurobiological factors,” and hey, one can construct superficially plausible theories around lots of those! Which is why it’s useful to try and rule some out, e.g. with intent and the guns/trivial inconveniences issue.

            (Conveniently, Nancy’s paper is also a test of the intent hypothesis. It claims to rule out intent via a psychological autopsy study (basically, interviews and reading through health/psychiatric history): “Although women were significantly less likely to use a violent method than men, there was no difference in the lethality of their suicidal intent.”) FWIW.

            Report comment

          • Montfort says:

            @Nancy, Chalid:

            The paper’s on libgen.

            It seems odd to me that they try to distinguish between suicidal intent by measuring the suicidal intent of successful suicides. I am not surprised that successful suicides measure about the same on the psychological autopsy, but maybe I’m missing something.

            Edit:
            I suppose under a basic model where “really suicidal” people try really hard to kill themselves, and succeed with probability X, and “less really suicidal” people try less hard, but still succeed with probability Y (where X>Y, but Y is still significant relative to population size), you would expect to see some difference. But it still seems like a more direct comparison would be a random people attempting suicides (not of suicide attempts – repeat attempts would confound things).

            Report comment

          • Just a little something for the Inkling fans: there’s a bit in a Charles Williams novel (probably Descent into Hell about a man who committed suicide with a suggestion that if he’d taken as much care about what he’d done in rest of his life as he did about his suicide, his life would have gone better. This wasn’t played for laughs.

            Report comment

          • Chalid says:

            @Montfort

            Your edit is of course correct. Keep in mind that X can be quite a lot larger than Y.

            I expect that doing these profiles on unsuccessful suicides is prohibitively difficult – harder to get access to medical records since you’d need the patient’s permission, family less likely to talk to a nosy researcher, and of course there are probably a lot of ethics concerns.

            Report comment

        • I assume that some fraction of the difference is the result of men mistreating other men rather than favoritism towards women.

          Report comment

          • anon says:

            In the context of discussing which gender is more marginalized (the marginalization olympics?), does the distinction matter?

            Report comment

          • vV_Vv says:

            What is the difference?

            Report comment

          • The difference is that the discussion seems to frequently get framed as either women are awful or women are unfairly favored. If part of the situation is that men should treat each other to some extent the way men treat women, then this should get mentioned.

            Report comment

          • onyomi says:

            I was thinking about the “women are wonderful” effect, and also the “goddess or whore” dichotomy and thought: maybe much of what we call “sexism,” against women, or, indeed, against men, is just one more instance of the outgroup homogeneity effect.

            Report comment

          • vV_Vv says:

            I think we all agree that Western men tend to treat women better than they treat other men.

            This can be framed as men giving unfair advantage to women or unfair disadvantage to men, it depends on what do you consider as the baseline of fair treatment. In practice, this looks like splitting hair.

            It is expected, and to some extent inevitable, that men give preferential treatment to women, at least as long as they can’t treat them as chattel. And even in very patriarchal societies like Saudi Arabia or the ISIS, where women are indeed considered pretty much like chattel, men still try to shelter them from physical danger more than they shelter fellow men.
            After all, from the point of view of a heterosexual male, women are reproductive resources while other men are reproductive competitors.

            The usual criticism of modern feminists is that they keep complaining that women in the first-world are treated worse than men and they ask for special benefits to “level the playing field” while in fact first-world women are generally treated at least as good as or better than men.

            Report comment

          • dust bunny says:

            They’re perhaps nicer and more considerate, but they’re also more dismissive and less inclined to take women seriously. You may disagree on what being treated well entails, but having experience with both lack of respect and lack of kindness, I consider the lack of respect a greater threat to my wellbeing by a wide margin.

            Old and ugly women exist too, by the way. They might not even concede your point on the kindness.

            Report comment

    • Maware says:

      People will start to hate the marginalized group even more, because it’s obvious they are being promoted unfairly. Said marginalized group will also be angry at being patronized and needing constant deck-stacking in order to achieve normal results, because you know they are oppressed. They need our help, you know. Systematic injustice. It’s not like they’d ever get hired solely on the quality of their work alone.

      Report comment

    • NN says:

      What’s the worst thing that could happen if you discriminate against a group that has had hundreds or thousands of years of institutionalised advantage?

      The Rwandan genocide.

      Less extreme but still quite horrific examples include the “fast-track land reform” of white farmers in Zimbabwe and the expulsion of Pieds-Noirs from Algeria after the end of French rule. See also every Communist revolution ever.

      No, I don’t think outcomes anywhere near extreme as those are likely to happen in the West, but pretending that discrimination against “privileged groups” is harmless is not only wrong but actively dangerous.

      Report comment

      • Muga Sofer says:

        The French Revolution was also pretty clearly aimed at “privileged groups” – i.e. rich people – as were much of Maoism’s purges.

        Report comment

        • Simon says:

          Having read the two published parts of Frank Dikotter’s trilogy to be on Mao’s actions during The Great Leap Forward and the Chinese Revolution I think you are crediting the Great Helmsman with too much organizational structure in the purges. Notionally they were against property owning classes and industrialists but as things uncoiled they grew swiftly indiscriminate.

          When the mud and straw buildings of ordinary farmers are being crushed for “fertilizer” to “improve” soil and yields, leaving huge numbers without shelter, things are not operating on normal rationality.

          Recommended reads BTW.

          Report comment

        • Not to mention the Kulaks, who Stalin murdered for being only slightly less poor than their neighbours.

          Report comment

    • hlynkacg says:

      Tom Bull asks: what is the worst that could happen?

      Is that a serious question?

      Nobody not Hitler, nor Lenin, nor any despot you could name, ever came forward with a proposal that read, ‘Now, let’s create a really oppressive and evil society.’ Hitler said, ‘Let’s take the means necessary to restore our national pride and civic order.’ And Lenin said, ‘Let’s take the means necessary to assure a fair distribution of the goods of the world.’

      “In short, it is your responsibility, men and women of the class of 2010, not just to be zealous in the pursuit of your ideals, but to be sure that your ideals are the right ones. That is perhaps the hardest part of being a good human being: Good intentions are not enough.”

      -Anton Scalia, commencement address to University of West Virginia

      Report comment

    • Daniel Kokotajlo says:

      For the record, I agree with you Tom Bull.

      The worst that can *realistically* happen isn’t that bad, relatively speaking. Which is why I’m way more concerned with being an effective altruist and making the world a better place than with fighting the culture war.

      Report comment

      • eh says:

        I present this list of left-wing mass murder deniers as evidence that, when genocides begin, movements similar to the social justice movement can both drive them to greater extremes and hamper attempts to stop them.

        In a world where half a million or more die from malaria each year, it’s understandable that you might want to focus on those deaths, rather than vague future possibilities of genocide. However, a disturbingly large number of countries contain ongoing genocides, and those countries are by and large the ones which effective altruists have to help. Regarding privilege, it’s obvious that the position of the Tutsis relative to the Hutus before the Rwandan genocide – rich, educated, middle class, and of a different ethnic group – was very much a privileged one. What’s not so obvious is that the DRC could probably afford more mosquito nets and deworming tablets if it hadn’t been plunged into two decades of racial violence in the aftermath of Rwanda.

        It’s very easy to believe that ideologies can’t have terrible consequences from the safety of America, and it’s very easy to build an ideology around criticising American whites, because as a group they’re impervious to harm, and because they have an undeniable history of slavery and colonialism. What will have real consequences is the application of Social Justice terminology and tactics elsewhere in the world. I’m worried that Americans, and other first-worlders, don’t see that socially-permissible racial vilification might lead to horrific consequences.

        Report comment

    • David Moss says:

      “Really, what is the worst that could happen if there is an institutionalised social justice movement like you say? Where people deliberately or subconsciously misinterpret results to skew their implications towards benefiting a marginalised group?”

      You’ve assumed the very thing which Scott was expressing doubts about: that the institutionalised social justice movement is simply about helping up the relatively disadvantaged.

      The worry is that “institutionalised social justice” ends up doing things which are nothing to do with helping disadvantaged groups e.g. taking a case where women are on the whole decisively advantaged relative to men and demanding more advantages; lowering the status of ‘enemy’ cultures just cos; providing an excuse to mock the weak and low status; denouncing the more progressive candidates (Corbyn, Sanders) because they’re not approved by the social justice clique, etc.

      Report comment

    • Really, what is the worst that could happen if there is an institutionalised social justice movement like you say? Where people deliberately or subconsciously misinterpret results to skew their implications towards benefiting a marginalised group? What’s the absolute worst thing? That people become more conscious of disadvantaged groups and implement positive action in their favour? That groups who have had hundreds or thousands of years of institutionalised advantage are now marginally disadvantaged?

      If you skew your interpretations because you think a group is marginalised, and you think that group is marginalised because of other people’s skewed interpretations, and you convince other people to skew their interpretations…how sure are you that the group you think is marginalised is really marginalised?

      More importantly, people aren’t skewing their interpretations towards benefiting those groups. They’re skewing their interpretations towards creating moral outrage. That may or may not benefit the groups. For instance, girls reading these news stories about the supposedly massive discrimination they’d face in tech might respond, quite rationally, by avoiding careers in tech. How is that to anyone’s benefit?

      Furthermore, where’s the “hundreds or thousands of years of institutionalised advantage” for 1st and 2nd generation Chinese immigrants? They’re the ones being overwhelmingly passed over for Ivy League admissions even though they’ve just arrived from a far poorer country.

      Report comment

      • InferentialDistance says:

        That may or may not benefit the groups. For instance, girls reading these news stories about the supposedly massive discrimination they’d face in tech might respond, quite rationally, by avoiding careers in tech. How is that to anyone’s benefit?

        It benefits websites who get paid advertising fees for everyone who clicks their articles. The more they mine the outrage, the worse the underlying problem gets, and the more justified their outrage appears.

        Report comment

    • Dave_M says:

      To be effective in creating positive change in the world you first need an accurate model of reality. Most versions of feminism and other SJ activism actively suppress critical thought, actively suppress the truth if it doesn’t fit their preconceived worldview. They often do more harm than good by pushing their simplistic solutions to complex problems.

      Report comment

  25. Murphy says:

    I have a question about the error bars/confidence intevals/things.

    Which are they? I was under the impression that they scraped the whole of github so confidence intervals don’t make sense. it would be like explicitly counting that you have 102019 bricks in a pile then sticking a confidence interval on it for no damned reason.

    Report comment

    • Decius says:

      Some contributors were not genderable?

      We have 102019 (bricks and cinderblocks), but there are some masonry items that we can’t figure out if they are bricks or cinderblocks.

      Report comment

    • Consider a random process in which PRs are generated and accepted. In this random process, women have an acceptance rate A_w and men have an acceptance rate M_w.

      Now think of historical github as a sampling of this random process.

      The confidence intervals/error bars/etc apply to that random process, and should alter your beliefs about *future* samples from that random process (assuming the world doesn’t change). I.e., after reading this study, you should change your offered betting odds on whether an individual randomly selected *future* female pull request will be accepted.

      Report comment

  26. Decius says:

    Your summary indicates that an automated program scraped usernames, the presence of a profile picture, and other available information to determine apparent gender.

    Did it use image recognition to attempt to determine a gender of the profile picture? Does a random sample of the gender assignment given by the program correlate well with manual scoring?

    Report comment

    • Deiseach says:

      Your summary indicates that an automated program scraped usernames, the presence of a profile picture, and other available information to determine apparent gender.

      Since for my Facebook account my profile photo is a screencap of the God Shiva* I took from an online upload of a Hindi TV serial, I would love to see this programme try identifying my gender based on that 🙂 (For a start, would they count the third eye or not?)

      *If you think I’m going to stick a real photo of my ugly mug online for the world and his wife to gawp at, you badly underestimate my level of reclusiveness.

      Report comment

  27. Dave Berry says:

    Most of the articles that I’ve read about discrimination against women have noted that female approvers/reviewers are pretty much as biased as male approvers/reviewers, from which they conclude there is a systemic bias. So I’m not surprised that this survey notes that women reviewers are as hard on submissions from women. The coverage in the press is another matter entirely!

    Report comment

    • Muga Sofer says:

      Yeah, this is a pretty standard finding, it’s just inexplicably absent from the popular narrative. Similar findings exist for racism.

      Report comment

      • dust bunny says:

        I seriously suspect being prejudiced against a group into which one belongs is simply too difficult a concept for most people to handle. It can be explained to them, and they understand it; it’s not that it’s too complicated. The next week they will have forgotten all about it, or their recollection of whatever was explained has mutated beyond recognition, or their understanding of it is entirely context-specific. The idea that people are consistent and rational is too deeply ingrained in them. They can’t bring themselves to really believe/accept/comprehend that people are capable of holding multiple contradictory beliefs at the same time. People who appear to do so must always be lying.

        Report comment

  28. To your list of pre-peer review comments I would add:

    10. If in the end the analysis fails to find any evidence of gender bias, submit the heck out of the paper anyway until it is published somewhere, anywhere. (Publication bias is a bias too!)

    Report comment

  29. d says:

    I was about to post a slightly snippy correction to a Metafilter comment about the SSC analysis, but actually on closer reading I think they may have a point. They were complaining that Scott’s characterization of women’s submission as being “less likely to serve project needs” wasn’t right, as the report said “less likely to serve a project’s immediate needs”. I was going to say that Scott actually used the word “immediate” in the preceding sentence, so it was hardly a big shift in meaning.

    But then I noticed that Scott summarized these submissions again later as “less useful changes”. I don’t think that’s right. Patches that mention Github issue (or user report) tend to address immediate problems (i.e. they’re bugfixes for specific bug reports or feature requests). Patches that don’t mention a specific issue are more likely to be new features or systematic rearchitecting of the software. They don’t necessarily serve _immediate_ needs, but it’s a bigger jump to say they’re less *useful* or not serving the project’s needs more generally.

    I realize that in both of these contexts, the broader picture is that women appear to be getting submissions accepted which are usually harder to get accepted, so yay women; but the characterization might be a little off. I think you can stick another immediate in the first mention and not lose anything. “Less useful changes”, I don’t know how to fix. Maybe describe them as “less obvious improvements”? Or just fall back on the “less immediately useful”?

    Report comment

    • Kyle says:

      The way the study assumes a “need” is highly flawed. If both the issue and the pull request come from external sources (if they were from internal sources then why open an issue or PR instead of just fixing it?) then most likely it’s the same person opening the issue and offering a fix. Thus it may be a need to the submitter but not particularly important to the project.

      What I’d consider more important or needful are issues that have been open for months or years and are full of “+1 please fix” type comments. Some projects even put bounties on their most needed or hardest issues so who is collecting those bounties?

      (Maybe I’m misremembering, but until like 3-5 years ago didn’t GitHub automatically create an issue for every pull request?)

      Report comment

  30. Deiseach says:

    the researchers wrote a program to automatically link contributor emails to Google Plus pages

    Annnnd crap like this is why I don’t have a Google + page or anything like it. Would it have killed them to ask contributors for permission to search their gender, or ask them to submit their gender in the first place?

    If I want you to know personal details, I will provide them. If I’m contributing something anonymously, then I don’t want you to know that stuff.

    By comparing obviously gendered participants with non-obviously gendered participants whom the researchers had nevertheless been able to find the gender of

    How good was this programme to find gender? If the researchers could find out a contributor was actually Sally Smith from other sources and were able to check that against the result given for the ‘guess the gender’ programme, that’s fine, but if they couldn’t find out who it was and made the best guess as to gender, how good was this best guess?

    If someone wants to tell me that no, they did ask permission and yes, it is a hot-stuff programme that is really very accurate for finding out real gender, then I’ll happily take correction.

    Report comment

    • anon says:

      Asking every contributor would have certainly led to a far smaller dataset.

      Report comment

      • Deiseach says:

        I’ll yield on far smaller, but surely more accurate than “we’ll try a gender-guessing programme” on the non-gendered names? If your point is to try and investigate possible gender bias, then trying “This was submitted by a female, if my interpretation of the username ‘killerbeez666’, profile photo of an octopus unscrewing a jar lid from the inside, and link to pastel-coloured biker leathers on Pinterest is accurate” may perhaps lead you astray in deciding which pile the particular submission goes in for counting.

        Report comment

    • honestlymellowstarlight says:

      This was my first reaction too, but if you post something publically, you post something publically.

      Report comment

    • Muga Sofer says:

      a) Yes, it would have significantly reduced the effectiveness of the study (which is founded on large-scale, automated data gathering) to personally ask each contributor for permission, probably to the point where they wouldn’t have bothered. That’s the reason the used public datasets in the first place.

      b) It’s not as if they were deanonymizing these people. They were making a statistical analysis of public information. Why is the risk of having your public data used as a datapoint in a study so terrible as to be “the reason you don’t have Google+”?

      Report comment

  31. Mark Atwood says:

    This whole topic is quite literally exactly in the middle of my profession.

    I have a lot to say, but it has to gell a bit (or a lot) in my head, before I start writing text. And then decide what true things will still be counterproductive to say.

    However, I can say right now: that the students that wrote this paper don’t know what they don’t know, and also a significant number of commenters here are likewise blessed with a comforting blissful unawareness of this particular slice of reality.

    It doesn’t help that this is a field with huge expanses of known unknowns, unknown knowns, and unknown unknowns, all spiced a large and unknown number of “things that are believed, that aren’t actually so”.

    If I’m really lucky, the fallout of all this will not make my next month and year more complicated. However, I doubt I’m going to be that lucky.

    Report comment

  32. Kyle says:

    It’s pretty easy to find the GitHub profiles of most of the authors. It appears that they worked in these two repos:

    https://github.com/DeveloperLiberationFront/GenderBiasInPushRequests
    https://github.com/DeveloperLiberationFront/ExploringGHTorrent

    Going by discussions on the second repo it seems like the first repo contains the data. So why not make it public?

    I bet this is part of the tool they used to guess a user’s gender: https://github.com/DeveloperLiberationFront/genderComputer

    Also note that Developer Liberation Front’s avatar on GitHub is the clenched fist.

    Report comment

    • Mark Atwood says:

      They have deleted the GenderBiasInPushRequests repo.

      Did anyone grab a clone or fork of it?

      And anyone who names their GitHub org account “DeveloperLiberationFront” has already predeclared their biases and their resistance to hatefacts.

      Report comment

    • Deiseach says:

      Also note that Developer Liberation Front’s avatar on GitHub is the clenched fist.

      Is that a male clenched fist, a female clenched fist, or a gender-neutral clenched fist?

      Report comment

      • The Nybbler says:

        It’s the Students for a Democratic Society clenched fist. Which is sort of amusing for a group which seems to have only one black member out of 11.

        Report comment

        • Kyle says:

          The clenched first is more of a general symbol of the hard left that goes back 100+ years. Though some anti-communist/anti-fascist libertarian groups also use it but make a point of only using the left fist.

          Also of note is the name of the group: “Developer Liberation Front”. “X [Liberation] Front” is a common name for many communist, fascist, and terrorist groups. Groups named in that format are often associated with violence.

          Report comment

          • Broggly says:

            Yes, that is what the refrance

            They’re clearly not a violent terror group. The issue is whether the joke is that they’re an apolitical group using a name and logo suggesting left wing terrorism, or whether it’s that they’re a left wing group jokingly portraying themselves as more radical and dangerous than they are.

            Report comment

        • vV_Vv says:

          Which is sort of amusing for a group which seems to have only one black member out of 11.

          In fairness if they are Americans one black person out of 11 is not inconsistent with proportional representation.

          Report comment