Hacker News new | comments | show | ask | jobs | submit login
How feminism helped me solve one of file systems’ oldest conundrums (valerieaurora.org)
67 points by zdw on Oct 6, 2014 | hide | past | web | favorite | 116 comments

OP here. To all the people flustered that I'm calling "listening politely" and "respecting people" feminist acts, I'll point out that, at the time, the only place you could reasonably expect that behavior in the Linux community was... a feminist collective, LinuxChix.

To this day, women in $COMPUTER_THING groups tend to be overrun by men searching for a civil place to have a technical conversation. It happened with the #debian-women IRC channel too. Just today, another man told me how volunteering for Women Who Code taught him to be more welcoming to newcomers in his own open source project. And have you seen the stickers on Guido Van Rossum's laptop?

So, yeah, it's possible to have these values without identifying as feminist, but in open source software today, explicitly feminist communities are usually the only ones that put these values into practice.

Regardless of any gender discussion, kudos for the atime hack. It's so much easier to say atime is a dumb feature or that apps relying on it are stupid. But you found a great compromise.

> Regardless of any gender discussion

The gender discussion (the insights it brought to the table) was central to the development of relative atime says the OP.

I don't care to get into that, it is a more complex subject than a single work like OP's article or the entirety of this thread can do justice towards. I am merely sayng a good hack is a good hack and discussing it on its merits.

> discussing it on its merits

Again, the OP makes it clear that taking seriously some of the basic tenets of feminism was operational in realizing the code. So, it is one of the merits. Otherwise, arguing that

> I am merely sayng a good hack is a good hack

and refusing, against your own argument, the merits of the code, is just a simplistic attempt to de-legitimize feminism as if it were a topic unworthy of serious debate. That, in turn, is just a redo of the very behavior (of not taking people and their stories seriously) that the OP (and the OP over at G+) was talking about in relation to the failures of the open source communities.

> is just a simplistic attempt to de-legitimize feminism

Please refrain from making assumptions about my motives. You don't know where I stand on this issue, I merely said it was complex and I was not attempting to solve it in a single comment.

I just wanted to comment on the idea itself and personally congratulate the author. I chose to not make your preferred topic the focus of my comment. I think you can live with that without making this kind of assumption about me, or making personal attacks.

I'd say the biggest success story so far in making a linux community that listens politely and respects people has been the Ubuntu project and I didn't think that was explicitly feminist. I also think it has flaws in how it implements policy sometimes, but it is so big I would be surprised if it never had any issues and I think in general it has changed the culture of linux community for the better.

Some of us, many of whom are involved in the Ubuntu Women project, spent the better part of a decade working to make Ubuntu that way. I'm glad our work paid off :)

You've done it pretty well, thanks for making my life easier.


This comment was killed by user flags.

Well, I should hope so. That is part of why we link to them.

Your comment suggests a strong confounding factor: feminist communities are likely to be both smaller and more focused on user conduct than technical communities at large.

Without controlling for these factors, attaching the behaviors you do to "feminism" is nothing but saying "good things are done by those who agree with me, and bad things are done by those /others/".

Your argument does nothing to explain why you think this happens, and simply admits you're appropriating them without knowing why the correlation exists in communities.

Classic propaganda.


I'm glad I've been downvoted and apologized for because I raised a methodological issue with trying to ascribe behaviors to a community, instead of actually showing me where she discussed why she doesn't think it's either of the (relatively well known) effects I asked about. I couldn't find anywhere she addressed this topic, but I'm entirely open to being corrected.

Here's the wikipedia article on the fallacy I'm claiming is being committed here: http://en.wikipedia.org/wiki/Correlation_does_not_imply_caus...

Ed 2:

My latin is terrible these days, lol

> feminist communities are likely to be both smaller and more focused on user conduct than technical communities at large

Is Python small? I'm sure there were other small linux communities (the term seems almost redundant) who were much less welcoming to women.

I don't know what to say to convince you that open source communities are traditionally hyper male and sexist. It's not so hard to imagine that you might get more contributions from women in an explicitly women friendly space within a larger women unfriendly (to say the least) community.

Here's some reading, I encourage you to read it if you think I'm wrong.


> Also, these communities’ openness means that a minority of difficult members (including, for example, a sincere misogynist or an insincere troll) can disproportionately affect the tone and dynamics of interactions. Finally, the ideology and rhetoric of freedom and openness can then be used to (a) suppress concerns by labeling them as “censorship” and, to (b) rationalize low female participation as simply a matter of women’s choice.

> I argue that some otherwise commendable features of the free culture movement also contribute to the gender gap. That is, the geek stereotype and discursive style can be unappealing, open communities are especially susceptible to difficult people, and the ideas of freedom and openness can be used to dismiss concerns and rationalize the gender gap as a matter of preference and choice.

I'm making a second reply purely to point out that the article you cited talked about confounding influences as being one of the main sources of the problem, while you attacked my comment for talking about confounding influences being part of the solution.

That seems absolutely insane, and suggests you didn't actually respond to my comment on the merits, but rather, out of anger someone didn't agree with you.

> It's not so hard to imagine that you might get more contributions from women in an explicitly women friendly space within a larger women unfriendly (to say the least) community.

No one (at least, not me) was talking about this.

What I said is that ascribing certain common behaviors to being a result of being feminist or being part of a feminist community without examining other causes is such a common fallacy is has a Latin name and wikipedia entry: http://en.wikipedia.org/wiki/Correlation_does_not_imply_caus...

It's also a classic propaganda maneuver.


Been too long since school, can't Latin on the fly.

OP provided evidence from personal experience: "In these communities, I observed this." You speculated about both qualities of feminist communities and qualities of user group size and composition, and then used that to drive the premise of your argument. That is unsound criticism.


That in a community you experience something doesn't necessarily imply that it's related to the core ideology of the group being different - there often are other factors which cause the change and would independent of that core ideology. "With this, therefore because of this" is such a common fallacy, it has a name in Latin: cum hoc ergo propter hoc. [1]

Further, my experience with online communities, which dates over a decade - and the experiences of my friends managing both large and small internet communities - implies that there are two correlations: a larger group will tend to have more problematic actors, who tend to be overrepresented in the number of comments, and that a group which focuses more on the behavior of members will tend to have better behaving members (if only because they ban the others).

Without controlling for two well known internet community effects, arguing that your community is somehow special is the height of improper reasoning.

[1] http://en.wikipedia.org/wiki/Correlation_does_not_imply_caus...

Ed: Terrible Latin.

> my experience with online communities, > which dates over a decade

Translation: "I am over twenty"

Seriously though, I love logic too, but it doesn't always make you right. Sometimes it makes you that guy that stops everyone making progress. I've been that guy before, and I've fired that guy before.

Translation: I don't approach very serious questions with the best tools to come to the correct answer, I just go with whatever feels good to me so I can be seen Doing Something!(tm) even if it ends up not helping, and I'll punish or censor anyone who disagrees with me!

Seriously, I don't understand this view when taken in response to serious topics: aren't sexism and the challenges women face in technology topics serious enough that they deserve us at our best, not us at our most base and animalistic? aren't serious topics the ones we should most realize we're emotional beings and strive to approach with rationality, knowing that they're complex and full of correlations caused by confounding variables, which are known to be deceptive to our instincts and intuitions? aren't serious topics the ones we should realize our own propensity to shout down dissent for less-than-good reasons, and instead strive to discuss as mature individuals our different experiences and views with each other?

Social topics are large, complex, and full of all kinds of confounding variables that make sussing out the true cause of things a complex task. That you think now is the time we should abandon the tools we invented just to do that - like questioning if a correlation is actually causal or not - is just kind of weird to me.

I'm open to being wrong, but so far, no one has taken the time to point out what they think I'm wrong about here (it would have just been a quick cite if she'd address that point!), and instead have attacked me personally for asking a methodological question about a sensitive topic.

So I'll be blunt: I don't think you actually care about women, sexism, or making things better. I think you just want to feel good about it, about Doing Something!(tm), rather than having to deal with the real work of sorting out a complex social issue.

That's all I hear from you when you say "Let's not be logical here, let's just go with what I feel is right", that you don't care enough about the topic to even be concerned if you're right or not.

The only logical approach is to demand rigor in situations that demand it. For me this is not one of those situations, because the cost (to me) of the OP being mistaken in her conclusions or actions is zero, as it should be to you. So we logically we should support her because if she's right we all benefit.

For some reason something about the OP's position has made you angry and defensive, as demonstrated by the lengthy replies you've made to a number of people, in my case containing some quite personal comments. Up to you, but I'd suggest taking some time to explore why that is.

I'm actually very supportive of the effort to make safe spaces for women in tech.

Rather, I'm unhappy the OP is going out of her way - to the point of making a separate post about it - trying to appropriate values common to other world views as being inherently feminist. I don't think that's empirically true, and I think that's pretty clear from the fact no one factually corrected me and attacked me for asking about it.

Rather, her comment classic "othering" of people you disagree with, and is not how we should conduct ourselves on sensitive topics such as sexism.

> Rather, I'm unhappy the OP is going out of her way - to the point of making a separate post about it - trying to appropriate values common to other world views as being inherently feminist

So you felt "othered" and didn't like it. That's natural, and fine. I feel the same way sometimes. It's clear that others here do too.

Where you're hitting turbulence is rationalising that emotional response and building hard logic on top of that soft foundation.

It's actually fine for feminists to appropriate values. It doesn't stop anyone else using them. Sure, we can resent the implication that we don't hold those values, but the way to deal with that insecurity is by proving it to yourself. If you're human-centred, be human-centred. If you know, other people will know too.

It's hard. I suck at it too.

I think we both know that if I had made the implication in the reverse direction, everyone here would be lecturing me in response on how I shouldn't be othering people I disagree with.

I think the response to me has shown that people weren't objecting to the question I asked and instead because I dared to call out such behavior from a feminist, instead of Approved Bad Guys(tm). Read up to where someone posted a study agreeing with me to tell me I was wrong.

I have two objections to her comment:

1. It's empirically untrue, and that does us no service if we're concerned with actually solving the problem. As the study cited to me points out, many of the problems related to sexism are actually caused by confounding factors in the community, and not necessarily just because a few problematic actors are sexist. Similarly, my point was that solutions are often about addressing these confounding factors, such as reducing the undue influence of problematic actors (eg, posting way more than most people).

2. No, I don't think it's okay for feminists to use propaganda tactics regularly used by groups to "other" people they don't like just because we agree with the goals the feminists have, and I think such poor behavior should be called out when anyone does it - including people we agree with. We may just have a difference of opinion here on what's acceptable, but you've been incredibly dismissive of me simply having a different opinion on what's appropriate conduct and attacked me personally for it. I regard this, too, as poor conduct.

I think at this point, everyone responding to me has shown two things:

1. They're not concerned if "feminism" (no one even seems to be clear what that means in this context) actually fixes the named problem of having safer spaces for women in online communities, they're supporting the claim for ideological reasons, not because it's true. All the data actually suggests that changes to the community orthogonal to the concerns of feminism have the dominant effect on the comfort level of women in the community, because they serve to amplify or diminish the role of problematic actors (of several stripes).

2. People who are supporting her comments aren't interested in actually engaging with others, and will use all kinds of fallacies, irrationalities, and outright abuse to try and shout down opinions that don't fit with with their chosen ideology.

I think we should expect better from people.

I guess it's not surprising that you feel like I've been attacking you, but I'm really not. At no point have I criticised your character or said anything that should be insulting (unless your self-esteem is bound up with being right, which is justifiably not my problem). I admit that me not engaging with your arguments must be frustrating, but I hold firm to my position because you have consistently missed the point in your replies.

Unpicking this stuff is hard, especially for the person under the looking glass. I guess it was arrogant of me to try to get you to buy into this perspective, especially in a HN thread.

Best of luck to you.

You're simply being dishonest if you're going to say you weren't subtly attacking my credibility to even speak on the topic in your replies. An implied ad hominem is still an ad hominem.

They have been nothing but condescending responses with the implication I'm being unreasonable, but without actually engaging with what I'm saying in response in any manner, pointing out what specifically I've said that's unreasonable, or anywhere that any of my statements haven't been accurate.

You've been extremely rude, and I wouldn't be surprised if you often get negative responses if you interact with people in this style.

None the less, I wish you well and hope that you learn to interact sincerely with people about topics, rather than feeling the need to preach at them.

I believe your original post might have been taken differently had you omitted the last line:

> Classic propaganda

I stand by that comment in light of the fact no one factually corrected me, and attacked me personally in a fit of irrationality, going so far as to cite a study that agreed with me it was confounding factors that largely influenced these kind of community behaviors as if it was correcting me.

I think this shows that people's support of the claim is not because feminist communities actually cause that (or rather, that it hasn't been shown in any clear fashion), but because they want to adhere to the ideology and claim successes for it, even when those successes may be caused by other factors.

Propaganda intended to other people who aren't in their party, start to finish. And I think we should expect better of people than that, when talking about serious and sensitive topics.

This is only true if OP's arguments have no content and their qualities are attributed solely to a particular group.

Given that OP's argument does have content that we can evaluate your argument is irrelevant.

So that propaganda thing is just jumping to conclusions.

First, Let me just apologize for commentors like ObviousScience[1].

Second, Guido's laptop for those that don't wanna search for it right now - https://adainitiative.org/wp-content/uploads/2014/09/fword_g...

Third, you definitely have an uphill[2] battle but there are those of us who get the point you're making.

1. http://i.imgur.com/6FOoDLA.png

2. https://news.ycombinator.com/item?id=8411699

First, it's incredibly patronizing to apologize on behalf of another commentator without first engaging with the commentator directly. Not detailing what exactly you disagree with doesn't enhance your argument and is frankly, a bit of a cop-out.

Second, thanks for the link, Guido's just an all-round awesome guy!

Third, to reproduce an excerpt from the other HN discussion you linked (as per your comment): >> _The whole tech scene needs to collectively take a sensitivity training course to regain its humanity._

That's a pretty heavy statement. How about the fact that the OP has mentioned at least on 3 different occasions in the article that she was taking the feminist "AKA" human-centered approach to solving the problem. Many would read and be left with the impression that the OP is somehow under the impression that non-feminist approaches are clearly not human-centric enough. There's far too much convenient conflation weaved in there, and if we're making a huge brouhaha about being politically correct, I would you encourage yourself and the OP to take the sensitivity training before brandishing such accusatory remarks on the community that doesn't necessarily identify itself as feminist.

> the impression that the OP is somehow under the impression that non-feminist approaches are clearly not human-centric enough.

Feminism is well-founded on the premise that personal experiences matter ("personal is political"). That is what the OP is refering to in that statement.

> politically correct

Political correctness (a symlink to hypocricy) has nothing to do with this. OP is telling the story of how feminism's teachings on how one should pay specific and careful attention to individual stories of their experiences pays off significantly in the long run for social change.

Almost no other critical political approach pays more attention to personal experience than feminism. The ones that do borrow their attentiveness from feminism. Henceforth, rather than being offended for some reason for OP's comments on feminism, I'd suggest you check out feminist approaches to methodology.

> As recently as 2013, a leading file systems developer was still arguing that file systems didn’t have to save file data reliably by mocking users for playing computer games.

Tso neither mocks users for playing games, nor argues that filesystems should not be required to be reliable. He says:

"[Lortie, that filesystem operation that you said was guaranteed to be safe actually isn't safe. You mis-read the docs. It's racy, and here are the two races that you could lose. Losing both could lose your data. Here's an analogy to illustrate the races: ]

The failure scenario would probably be something like the user who plays tux racer all the time, and uses crappy proprietary drivers that crash the system every single time an OpenGL application exits. If they think that's normal, and are willing to live with the crap proprietary drivers, and they are also the sort of people who carefully position all of their windows to be precisely just so, and if the !@#!?! desktop libraries are still bogusly rewriting the entire contents of every single registry file, regardless of whether the application changed anything --- then eventually, said user will whine about how the hours she spent obsessively setting up their window layout got lost after Tux Racer creashed (sic) their system again."

Notice the carefully constructed persona used in the illustration:

* Tolerates terrible, crashy, OpenGL drivers.

* Makes use of the OpenGL drivers by playing a video game.

* Spends a lot of time meticulously positioning application windows.

* Uses a WM that stores application window position in a single file that gets updated all the damn time for no reason at all.

* Whines when the enormous amount of work put into positioning windows gets wasted when the poorly-designed software they're using crashes and takes some of their data with it.

It was safe to rename() a file and expect the data it contained after crash to either be (1) the data before the rename(), (2) the data after the rename(). What changing the ext4 default logging mode to "data=writeback" did is add two more options, (3) nothing (0 bytes), and (4) whatever random garbage it found on disk, potentially including a copy of /etc/shadow. No-one misread the docs, the behavior changed.

But hey, if you're angling to be hired by a YC-funded storage startup, this is certainly a post.

> No-one misread the docs, the behavior changed.

I write this without a hint of snarkiness or ill will. Please keep this fact in mind as you read on.

If you read the message by Lortie [0] that Ts'o responded to [1] you find that Lortie was specifically talking about the guarantees made by ext4 in regards to rename behavior. In that message Lortie says "[the documentation of ext4 in Documentation/filesystems/ext4.txt] says to me "replace by rename is guaranteed safe in modern ext4, under default mount options"." Ts'o opens his message with "It's not _guaranteed_ safe."

I understand that people are frustrated about ext4's default options. In the referenced exchange, neither Lortie nor Ts'o were talking about the merits of the differences between the rename robustness guarantees of ext4 and other filesystems. Lortie was double-checking his understanding of the documentation of the guarantees that ext4 provides with its default options. Ts'o was correcting Lortie's misunderstanding of that documentation.

[0] http://www.spinics.net/lists/linux-ext4/msg38774.html

[1] http://www.spinics.net/lists/linux-ext4/msg38778.html

Trying to figure out how to apply Charles' Rules of Argument to the situation, I realized I wrote an entire article on this topic (and linked to it from the OP):


Short version: ext4's behavior under default mount options changed. You can't blame it on someone mis-reading the docs.

The actual behavior is orthogonal to the guarantees in the documentation.

> But I was part of a culture – a feminist culture – in which I respected people like my friend and programmers that attempted to use fully defined, useful features of UNIX in order to implement features efficiently.

That is one hell of a stretch of the definition of feminism.

This article seems like some weird interweaving of two unrelated stories about sexism in tech and relative atime. The stories independently seem reasonable, but this weird fusion does not really make sense to me other than as a promotional advantage.

OP seems to have conflated taking a "human-centered" approach with a "feminist" approach. This constant reinforcement in the article about taking the "human" point of view leads me to believe that OP's perception of non-feminists is that they're in the camp of people more likely to take a "non-human-centered" approach. How bizarre.

It's not conflation, she used approaches developed in another context. Jumping from feminist to human-centred doesn't seem like a bizarre stretch to me at all. Both (from the perspective of the status quo) involve letting go of "I know best" and attempting to empathise with others, understand their problems, and often try to drive change.

Am I the only one that feel like this article is a whole "I know best" than non-feminist group?

Agreed. It's one thing to say that taking the human-centered approach (making discussions more open and inclusive to all people with a varying depth in knowledge/experience) is what we should ultimately converge to because the entire community benefits from people being more civil and understanding; it's a whole other thing to say feminism is really THE way to converge to this. By virtue of saying this on 4 different occasions in the article, OP's regard of feminist approaches to problem solving comes across as more supercilious than she may have intended.

Here's my simple takeaway. The article could have been a reasonable read if the war story was painted with the context of "oh hey, this is how more discussion and safe spaces helped fix this bug and here's why all of us should do it". What I got instead was "Wow, things/ideas/methods in the community are so broken. Feminism is obviously the only contender to fix this properly, and people who aren't in the feminist camp can't be as effective in reaching these goals. Btw, did I mention that I got groped? And also, everybody who disagreed with me is so clearly anti-feminist"

No, I don't think you are the only one. My personal take is that the OP meant no such thing.

I think she's just saying that women bring different approaches, and those different approaches can be useful.

Interesting that you use the word 'just' there. I think it's a much stronger statement to say that women bring different useful approaches than to say that feminism does so. It's an assertion of differences at a significantly deeper level.

It'll seem much less bizarre if you spend a week or two reading LKML :/

It seems to me that the author is conflating feminism with her helpful community of mailing lists, and further with good engineering practice.

I don't see anything particularly feminist about reporting bugs upstream.

She's saying that the creation of a "safe space" for women to discuss linux development without fear of dismissal was key to her being able to solve the atime issue.

But why limit it to a safe space for women, why not a safe space for anyone who doesn't want to be insulted?

It seems sexist to me to assume men can't be in a non abusive community.

It's a pretty accepted fact that men interrupt women and women do better when they have women authority managers or teachers. It's why there are women's colleges.

The more we work to make every space safe for women, the less need women will have for a safe space.

edit: There are enough exclusively male places for me to be feminist in, if I feel there's some sort of shortage of places to express myself.

The point is that this is a polemic piece, it's Valerie Aurora after all. The opposite of polemics would be apologetics. Valerie is implying that feminism is equivalent to civility, civility in the Latin sense of being mindful of the rights and responsibilities to human society.

That's where she loses me.

And what's with this dehumanizing technique of not naming those she disagrees with? You see this often.

They name names, it becomes a radical feminist attack piece. People would ask, why did she attack so and so? They are so nice and do so many things for the community.

What's with begging the question by saying "it's Valerlie Aurora after all"? It doesn't do anything for someone reading your comment who doesn't know what to think of her.

I'm not a gender studies major, but to me it seems safe to say feminism is rampant with sexism. Not necessarily in the hateful way, but in the way that affirmative action is fundamentally racist.

So, while I may not always agree, it is never a surprise.

"Feminism is the radical notion that women are people." - Rebecca West

There's nothing sexist about feminism; it's simply wanting equality for women -- that does not mean that women and men are the same (as is often claimed), but that they are treated the same. Comparing feminism to affirmative action is incredibly misleading.

Rebecca West does not speak for all feminists. She wasn't even alive for the mess of the third wave we're still in. I'm just replying to say your comment on a first reading comes across as naive in the subject of feminism. But it can also be seen as disingenuous, for it's a classic propaganda technique to hide behind a reasonable claim made by a proponent of X_1 to further along some goal of X_2, all while using the undifferentiated label X. (See https://www.mtholyoke.edu/acad/intrel/orwell46.htm on 'democracy'.) I also can't ignore that, left unqualified, an uncharitable reader may interpret your claim for men and women to be treated the same as a claim to be treated the same in all contexts, which is obviously a stupid idea, because men and women aren't the same.

I'm not trying to argue with you, I'm just trying to encourage you to be less sloppy if you're going to reply to comments like sliverstorm's with intent to change minds. At best, your reply is just an argument over a definition (http://lesswrong.com/lw/np/disputing_definitions/), let's all try to do better.

The core ethos of feminism is that simple, yes. But how do we get there?

That's where things like affirmative action routinely work their way into the discussion, and I hope you don't mean to pretend otherwise.

Pretending that the definition you gave is actually what drives all the "feminism" movement isn't being honest: it's certainly true for most people, but there is (and always has been) a not-small branch of feminism that is staunchly sexist.

Kinda like Hacker News. CHECKMATE.

Feminism, in general, is biased against men in the same sense that giving a short person a box to stand on at a concert so he can see the stage, and not giving a box to a tall person who can see just fine already, shows bias against tall people.

This is a great metaphor - on a concert, it doesn't matter who sees the stage - the important thing is the music, there is really no difference between short and tall people, and noone needs a box. The only reason the small person might need a box is because (s)he feels inferior compared to the large one.

This is just like a lot of feminism. Feminists arbitrarily decided that men's lives were better than women's, and wanted to make it possible for women to lead men's lives. Personally, I consider that being at home is better than being at work, but feminists hardly ever fought for the ability of men to be stay-at-home dads.

I'm not condemning box-giving. I'm just pointing out that only giving boxes to short people is by definition discriminatory (I hope you agree or we need to have another conversation).

It's positive discrimination (which, coincidentally, is how affirmative action is described) but discrimination nonetheless.

I'm not judging. I just thought I would share because I find it useful to remember.

Hmm. How would you categorize, say, giving powered wheelchairs only to people who have difficulty walking, and not to people who can walk fine by themselves?

This is the most patronizing thing I've read in a while. Women don't magically have some (ableist) disadvantage that prevents them from participating.

The thing that keeps them from participating is men bullying them and forcing them out.

You're reading way too much into my analogy. The point is just that lifting up someone who's been pushed down does not also require you to lift up someone who's doing just fine in order to be "fair." This applies whether the disadvantage is due to outside bullying (as in the case of feminism and affirmative action) or an actual disability (it's not "unfair" to make special allowances for an autistic office worker, for example).

That is wrong, that suggests women are inferior to men and need help permanently.

In reality the real goal should be making the short person grow to be equal height.

That's my point. Don't overthink the analogy. Or, feel free to substitute "give a ladder to a person stuck in a hole, but not to a person on solid ground," if you prefer.

And what if the small-person-on-box now blocks the view of the tall person?

For purposes of this overstretched analogy, it's much more common for the short person to be beside the tall person and not blocking their view at all, but the tall person starts screaming that they deserve a box anyway.

This is brilliant. I'm going to have to steal this whenever possible.

How misogyny helped me beat the market

I had a problem - figure out how to make money off the stock market. Luckily I was part of a culture - a privileged mysoginist racist evil culture - which recognizes that there is an objective reality that can be approximated by concrete models and empirical corrections. So I did some bayesian stats and came up with a moneymaking stat arb strategy.

That's about as reasonable an argument as the one contained in the OP. It occasionally brushes against reality (e.g. feminist critiques of analytic philosophy and objectivity [1]) but really just attributes intellectual success to an unrelated ideology.

That said, I'm upvoting this in the hopes that someone can steelman her argument, which could be interesting. The aforementioned steelman should hopefully also reject my argument.

[1] http://plato.stanford.edu/entries/femapproach-analytic/#FemC... http://plato.stanford.edu/entries/feminism-epistemology/

One thing to note is that the particular problem was known, but all the solutions involved suffering: "“Switch to a slower directory-based method,” or “Use a file size hack that had bugs,” or any number of other unhelpful things".

Maybe this comes from a macho sensibility. Another example is the problem that it's easy for a noob or fat-fingered typist to type "rm -rf *". The current solution is "lump it".

Maybe women are more likely to weight things in favor of not blowing things up (data from actual women needed), and feminism is required to give equal weight to that viewpoint.

Many maybes here. I am spitballing too.

You want to allow rm -rf * (shell globbing happens first, so rm cannot know it was called with ).

You also want to allow rm -rf ./ but you don't want to allow rm -rf / *. And that's where `--preserve-root` comes in, which is default now I believe.

EDIT: How annoying. The asterisk makes emphasis weirdly.

rm -rf * is a power tool. Because of its usefulness, you don't remove the blade from a circular saw. You do, however, put a hand-guard on the blade to prevent catastrophic accidents. Note that rm -rf / requires --no-preserve-root to remove / on most (all?) distros.

The shitty performance of atime is shitty performance, rather than a useful attribute of the tool.

Steel man. Interesting: http://wiki.lesswrong.com/wiki/Steel_man . Sort of an idea in the vein of (but stronger than) the Principle of Charity.

Yes, the steelman is a stronger variation of the typical principle of charity. It goes beyond interpreting your opponent's argument as strongly as possible, all the way to actually fixing your opponent's argument. It's a great technique, one that I'm not as good at as I'd like to be.

On the one hand, arguments supporting feminism that I see tend to be vacuous and borderline dishonest (case in point: this article). But there are sufficiently many smart and honest people who buy into the general theory that it's hard for me to completely reject it. That's why I'm so curious in threads like these, and why I'd love to see a steelman.

Some good steelmen would allow me to draw a reasonable conclusion about the theory itself, rather than merely it's vocal proponents. I could then safely either accept or reject it.

ISTM that term could stand some improvement before being used in a feminist discussion.

The OP responded. Basically, linux needs feminist spaces because FOSS communities are traditionally sexist spaces where their contributions are not welcome. The point isn't that no one else could have had the idea, that's absurd. The point is that no one had until that point, so it wasn't trivial. It's the banal idea that if we have spaces where marginalized groups are able to contribute without all of the extra baggage associated with being a part of the group. In the FOSS community, you face extra criticism as a woman.

I'm confused - did the traditional "go upstream to Linus" community reject her patch? Or did she somehow get a patch into the kernel via a sidechannel through linuxchix?

Near as I can tell, this was merely a case of the "hyper male sexist" community letting good code speak for itself.

I understood the article to be saying that the values that the author identified with the LinuxChix community assisted in bringing the issue to their attention, helping them understand that the use case didn't require the performance penalty that atime represented at the time.

Here you are yet again "just asking question" on a post about women.

It is never ever the other way around. You simply never make questions the other way.

I have to point this out because you claim to be unaware of it and you claim to be even handed in your questioning.

I don't understand. What is the "other way"? Are there posts claiming that other political/philosophical ideologies (I guess one that is "opposite" to feminism) result in feats of systems programming? If so, the only reason I haven't asked is that I've never seen them. This is in fact the first article I've seen online making a "political ideology X => programming feat Y" assertion.

I have, on one occasion, asked someone in person a similar question about his Objectivist principles. His justification was about as weak as the one given here.

Last time you claimed I never asked questions the "other way" on a different topic, I pointed out explicitly cases where I did: https://news.ycombinator.com/item?id=8372354

I don't really see how this is feminism. I do think that supporting women in tech is a noble goal, but "considering user needs in the design of a system" never struck me as a uniquely feminist concept.

Why would this have to be "uniquely feminist"?

If it's a feminist value, and that's how OP comes by the value... isn't that good enough? Seems like that's a fine way to arrive at human-centric design?

OP's point is actually broader than just human-centric design too. She's not just talking about consideration for users, but also a manner of collaboration w/ other developers too.

Supporting women in tech is not a “noble goal”; it’s a baseline for civilised behaviour. And while “considering user needs in the design of a system” isn’t necessarily feminist, it is something that is more likely to come up within a space that starts from respect rather than disrespect.

(Blaming users, male or female, is inherently anti-feminist.)

Actually, it sounds like an issue completely unrelated to gender at all.

Much of feminism - particular the branch that is vocal within the tech community - is not gender-specific. It's a social (and in many cases, political) worldview.

Then why call it feminism? That is just misleading.

I guess I could parse the attitude of developers who don't "consider the user's needs in the design of a system" as patriarchy... but in the gender-neutral "parents telling children what's good for them" sense of the term, not the gender-politics one.

Err... I don't think there is such thing as a gender-neutral patriarchy.

A patriarchy is not "a power system" or "a system of authority". It is a system in which males are in power.

"Unfortunately, as a result of my work, several more Linux storage developers came out publicly in favor of harassment and assault."

I really would like links to these public statements. I'm not a fan of shaming, but if people are advocating violence, some finger-pointing would be entirely appropriate.

Will you volunteer to filter my social media and email for rape threats, and compensate me for the lifetime earnings loss that is the result of naming someone who controls whether my code gets merged or whether I get invited to the top Linux invite-only summits? I've learned not to name names unless I'm ready to face the consequences of speaking out to protect other people.

Of the three options noatime, relatime and atime, I think relatime is the worst. Why? Because it is the most complex. Ordered by complexity, noatime is the easiest and relatime the hardest. Updating a timestamp based on an algorithm is much harder than never doing it or always doing it.

Complexity is bad, complexity even if it solves user problems is bad. The bash shellshock bug is a consequence of complexity.

If I where Linux dictator, file access time would be deprecated starting from today. POSIX be damned, but they can also do the right thing and deprecate atime in their standard. Mutt would suffer but I'm pretty sure I could write a patch for it to make it independent of files atimes. It saddens me that it seemingly "cant be done" because of legacy. Linux is full of Unix cruft in every corner (dotfiles anyone?) it needs to be cleaned up.

Although you're right about it being relatively the most complex, it's absolutely a very simple algorithm.


  if false:

  if true:

  if a > b:
Doesn't seem very complex to me.

I believe the original check was if mtime was greater than or equal to atime. The current Linux implementation isn't that simple though, see: http://lwn.net/Articles/244829/. If you allow some extra complexity, it's easy to then allow some more, and then some more..

I'd like to see that patch because it sounds pretty damned tricky to me unless you're proposing to read the entire file in each time you want to check if it's been updated.

As a man I have found the linux community extremely toxic, RTFM etc, over time this has gotten better, but I would hardly call it sexism.

It's not about sexism, it's about decreasing "geek machismo" that has existed in the community, and how this helps more people be comfortable putting themselves out there and having a technical discussion (where there is always a nonzero chance one is wrong) in public.

fwiw the Kernelnewbies list and irc is another good place to start for civil kernel discussion. I would suggest that common to both it and Linuxchix is the belief that no matter what the questioner or respondents' level of expertise, the only possible response to a question is a positive, helpful one.

This entire article could communicated everything it intended by focusing on the motivations behind creating intermediate "safe spaces" to have an open judgement-free discourse. Instead, it was pitched as a pro-feminist piece coming across as Feminism = InclusiveProgress. I'm in agreement with ObviousScience when he says that it came across as a propaganda piece.

Also, I would rephrase the "geek machismo" to "geek hardassery" since machismo seems to outline masculine pride. I do, however, agree that people tend to be hardasses with the whole RTFD/RTFM angle and that's something we could collectively improve upon.

I think this is a biproduct of what open source is. People contributing their spare time for little to no financial gain, and when new people expect them to hold their hands they take offense and lash out. I think it could be a slippery slope to expect people to act a certain way when contributing to the common good (open source and the like) because really there is no obligation to use the work these people produce.

Conversely this sort of behavior when dealing with vendors where money changes hands is almost non existent. And I suspect this is because there is a economic incentive in place for people to act civil to each other.

Complaining about this dynamic wont change it only changes to the economic model will change peoples behavior.

Great read, thank you for posting that! There's an interesting shift towards empathy in programming that I think will make better software. The empathy to ask - what is the easiest for the user, what are the expected actions and expected results from the user's point of view, etc. This same approach can be used in actual software design as well - treat each components / class / API as a person with motivations and goals, figuring out what's the least amount of info this person needs, best way to get to their goal etc. I've found that way of thinking yields cleaner design overall, and is easier to communicate to others. This method of approaching problems should be encouraged, heck, if not demanded, from all!

The problem with women's groups in technology is that there are very few women in tech to join the groups.

And I highly doubt that male chauvinism/sexism/harassment is the reason women aren't going into tech careers.


"The" reason? Certainly not. There are far fewer girls getting exposed to tech early in their education. Far fewer girls see tech in general as a viable career path.

"A" reason? Absolutely. Other commenters have have linked to specific incidents where female tech employees have left jobs and even left the industry as a result of they way there were treated.

I don't say this to downplay your position - which I understand to be that there are in fact other factors driving the gender ratio is tech. I assert without reservation that treatment of women in our industry is not the primary factor causing the imbalance.

In my opinion, the social conversation within our community is toxic to both sides. I try very hard not to become involved in it. That said, my own political views lead me to see individuals as individuals first and foremost, to treat others in a way that encourages cooperation, and to engage with my peers in voluntary and mutually beneficial ways. In almost every way that I've been able to objectively nail down, I am a feminist. I just don't share some of the viewpoints of the particular variant that is vocal in our community.

That's fine. I've worked both online and in person with many people whose beliefs in this matter differ from mine. Human beings are perfectly capable of identifying areas of common interest and working together to improve both of their positions, however incompatible their desires end states happen to be.

I try to share my own values by working with young people and encouraging them to see tech as a valuable and valid career path - and as a passion that's worth pursuing regardless of their profession. There are many people who don't see it that way, for reasons that vary depending on their background.

My sincere belief is that a world where every young person sees tech as an option is superior to today's world, where a large portion of children and young adults see it as inaccessible or unworthy of their attention. You don't have to self-identify as a feminist (or anything else) to share that belief :)

>>And I highly doubt that male chauvinism/sexism/harassment is the reason women aren't going into tech careers.

Then you're wrong - http://fortune.com/2014/10/02/women-leave-tech-culture/

That article has serious problems. It is basing much of its analysis on the idea that the pipeline into tech no longer is highly unbalanced. The sources cited by the article actually show that this is not correct--the pipeline is still very unbalanced.

The pipeline is unbalanced, yes, but the meat of the article is in how it surveys the reasons women leave.

The number of females entering tech is a fraction of the number of males. Therefore even eliminating all harassment of females would not bring us closer to equal representation.

That's not to say we shouldn't work to eliminate this treatment or an attempt to minimize it in any way, of course.

What I believe we should ask ourselves is this: "Is the tech community responsible for the fact that far fewer females see this industry as a reasonable and viable career path?"

Until we can answer that question with a confident "nope", then the social conversation around feminism and its role in our community need not be settled. We can work toward a more equal future without addressing the underlying philosophical and political motivations for doing so.

Free individuals, voluntarily working together to achieve a common goal. That's what it's all about, people. Forget the lables, find something in common with your neighbors, and make it happen.

Well, if women are leaving faster than men and often cite reasons of being bullied or otherwise forced out by men... why would women desire to enter that workforce?

That article doesn't actually claim it's discrimination.

Rather, it complains that women don't get enough positive discrimination (ie, discrimination in their favor or against men).

> Some women felt that their work environments were discriminatory, but most reported something milder: the simple discomfort of not fitting in in an otherwise homogenous setting.

That is, the only actionable thing that it suggests is that you should preferentially hire women so they don't feel so lonely, not that there is any actual discrimination to address.

How would you explain that there used to be more women than men as programmers? (http://gender.stanford.edu/news/2011/researcher-reveals-how-...)

What do you think changed?

  > I highly doubt that male chauvinism/sexism/harassment is the 
  > reason women aren't going into tech careers.

  > How would you explain that there used to be more women than men 
  > as programmers? 
The women who went into "programming" back then for the most part played a role somewhat similar to the role a compiler does for modern programmers, and so the fact that it seemed like an attractive role to some women of that era has very little bearing on its appeal now.

At that time "programming" meant the relatively unskilled job of translating someone's detailed pseudocode onto punch cards or worse. The need for such unskilled labor diminished as the abstractions and automation improved. Ensmenger speaks for himself about his research on this in the following article:

  Goldstine and von Neumann spelled out a six-step programming
  process: (1) conceptualize the problem mathematically and
  physically, (2) select a numerical algorithm, (3) do a numerical
  analysis to determine the precision requirements and evaluate
  potential problems with approximation errors, (4) determine scale
  factors so that the mathematical expressions stay within the fixed
  range of the computer throughout the computation, (5) do the dynamic
  analysis to understand how the machine will execute jumps and
  substitutions during the course of a computation, and (6) do the
  static coding.
  The first five of these tasks were to be done by the "planner," who
  was typically the scientific user and overwhelmingly was frequently
  male; the sixth task was to be carried out by coders.
  Coding was regarded as a "static" process by Goldstine and von
  Neumann -- one that involved writing out the steps of a computation
  in a form that could be read by the machine, such as punching cards,
  or in the case of the ENIAC, plugging in cables and setting up
  switches. Thus, there was a division of labor envisioned that gave
  the highest-skilled work to the high-status male scientists and the
  lowest-skilled work to the low-status female coders.
  As the ENIAC managers and coders soon realized, however, controlling
  the operation of an automatic computer was nothing like the process
  of hand computation, and the Moore School women were therefore
  responsible for defining the first state-of-the-art methods of
  programming practice.

This one story -- about a small group of women chosen for their exceptional capability as human "computers" who previously calculated by hand numerical tables like the ones the ENIAC was designed to calculate -- has been spun into a narrative suggesting that women dominated early software development. Nothing could be further from the truth, and I don't think Engsmenger sees it that way. He says elsewhere that "Although many of the earliest programmers were women, by the beginning of the 1960s programming was generally considered a male profession. Certainly the elite ranks of ‘‘systems men’’ [i.e., those responsible for what we would today consider software design] were, quite literally, men." http://homes.soic.indiana.edu/nensmeng/files/ensmenger2003.p...

Of course, all of this shows that male chauvinism (and very very very probably harassment too) has always been a huge problem for women who wanted to participate in this field, but the idea that "there used to be more women than men as programmers" just isn't true, at least as we understand the word "programmer" today, and the evidence it's based on has no bearing on why women aren't entering tech fields today.

You can't just bag all reasonable behaviour and call it feminism. Feminism might be a broad church due to there being many historical revisions of it, but if you make it so wide as to just basically mean people being nice, then it loses all meaning.

Also, if you make the argument that a feminist environment was in some way necessary rather than merely conducive to the development of this particular solution, then were the experiences gained in a hyper-masculine and highly misogynistic organization based on the needs of state violence required for Grace Hopper to invent compilers, given that she was a Rear Admiral in the US Navy?

edit - I fully support the creation of comfortable safe spaces and organisations that bring more people into programming as society in general seems far too accepting of abusive bigots, so these places and organisations serve a good and useful purpose. I also think that the creation of spaces like this will in time help raise the quality of software by having a wider range of perspectives looking at the problems, so it isn't just a good idea because it is being nice. I just think that the author here is putting an undue amount of weight on feminism as a major cause of this particular solution to the atime problem. I can't see anything about it as a solution that couldn't have been worked out by, say for instance, a hermit who hates everyone.

Er, why not? Are reasonable behaviors a scarce commodity?

If the argument is that

1) Feminism is a tradition of empathy and humanism 2) the application of OP's feminist ethos led OP to prioritize the needs of users in a particular way

then isn't this particular manner of operating, by OP's definition, feminist?

You can argue that Feminism is indistinguishable from other sorts of ethos, but it'd be hard to argue that it's not an ethos at all.

Ok, you can, but I just like well defined terms as it makes it easier to know what you are discussing.

Feminism is very broad already, a good example being the conflict between those who advocate for equality in political representation and those who advocate for working towards a utopian matriarchy.

I just think it is confusing and weakens the argument being made by overstating it, to make the definition broader still.

The reason it seems to you that broadly defined reasonable behaviour gets all thrown in the feminism bag is because feminism just asks people to act reasonably with women; it does encompass a broad set of reasonable behaviour. As the quote goes, "Feminism is the radical notion that women are people." Yet this is so hard to achieve.

The idea that women are people is not confined to feminism. You could of course argue that all outlooks that regard women as people are de-facto feminist, however this would still not make them feminism. A good example would be ancient Sparta. Their society definitely regarded women as people, however Spartan society is not feminism.

The fact that Sparta treated women better than the average Mediterranean city-state makes it ipso facto feminist.


This comment was killed by user flags.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact