Ok. If the issue is "problems faced by women in FOSS" then where is the list ? [LWN.net]

Ok. If the issue is "problems faced by women in FOSS" then where is the list ?

Posted Sep 30, 2007 12:31 UTC (Sun) by nix (subscriber, #2304)
In reply to: Ok. If the issue is "problems faced by women in FOSS" then where is the list ? by evgeny
Parent article: To Sir, with Love: How To Get More Women Involved in Open Source (O'ReillyNet)

The only facts I can see in this thread consists of requests from you that
other people should prove a negative. This is of course impossible.

And as for `just look around': can you find *any* other field of human
endeavour with a percentage of female participants as low as free software
development? It's damned rare (perhaps pure maths?), and software
development is not something which appears to be intrinsically male-only:
even if you go in for the `extremes of ability are male' hypothesis, free
software development does not *require* extremes of ability. It's not
*that* hard, and with a global talent pool to draw from we'd expect more
than the well under a hundred female free software developers total that
we see.

Women don't 'choose not to participate' in most fields. It behooves us to
determine why they choose not to participate in this one, and if possible
fix it: not by forcing anyone to participate, not by massive and enforced
changes, but simply because doing things that cause half the human race to
choose to avoid us is a bad way to ensure that we have many eyes to make
bugs shallow and come up with neat ideas. It's obviously a good idea, just
as trying to make small changes that avoid us driving away Indian or
Japanese participants is a good idea.

(And they *do* participate in proprietary software development in larger
numbers. 1/5th of the *programmers* on the thing I work on in my working
hours are female: still not a high percentage but way higher than in free
software.)

to post comments

Ok. If the issue is "problems faced by women in FOSS" then where is the list ?

Posted Sep 30, 2007 13:08 UTC (Sun) by tzafrir (subscriber, #11501) [Link] (1 responses)

I can't find any decent statistics right now, but I suspect that you'll find striking similarities to the rate of males/females which have graduated computer sciense with honors.

I also suspect that this applies to some minorities as well.

(At least this was rather true when I graduated almost 10 years ago).

And this is before one extra factor comes into play: "devoting extra time to the family". Like it or not, women are more likely than men to reduce the work load in order to raise the kids. Unless you claim that much of the free software is developed by "students" and thus this factor is not as strong.

(You can like it or not. You can think of ways for changing that. But those are outside the scope of the current discussion)

Ok. If the issue is "problems faced by women in FOSS" then where is the list ?

Posted Oct 2, 2007 2:45 UTC (Tue) by njs (subscriber, #40338) [Link]

>I can't find any decent statistics right now, but I suspect that you'll find striking similarities to the rate of males/females which have graduated computer sciense with honors.

I don't have statistics on graduation with honors, in particular, either. It's not a well-defined term between institutions, even. So here are some other statistics for the US, from the same report that we discussed a few days ago in the other thread: http://lwn.net/Articles/251569/

Bachelor's awarded, CS: between 28% and 37% female over the last two decades.
Enrolled graduate students, CS: between 20% and 30% female over the last few decades
Master's awarded, CS: between 25% and 34% female over the last few decades
PhD's awarded, CS: between 15% and 19% female over the last few decades.

All of these numbers are wildly larger than the percentage of females in FOSS coding, which is in the 1-2% range.

>And this is before one extra factor comes into play: "devoting extra time to the family".

This is like the 5th "extra factor" you've made up as previous ones have been shot down, but anyway. This one, for a change, is almost certainly real, but I still dispute its significance.

Suppose that the "natural rate" for women in FOSS would be 15%, because that's the smallest of the numbers mentioned above, i.e. we can be pretty sure it's a low estimate. (In fact, having children is a reasonably common reason for women to drop out of PhD programs, so it already includes some of the effect that you are claiming accounts for the difference.)

Suppose that the "actual rate" for women in FOSS is 2%, because that's on the high side of the statistics I've seen. With those assumptions and if I didn't screw up the arithmetic somewhere, that then means that your "extra factor" needs to explain the loss of *87%* of potential female contributors -- probably more in reality.

Here I also don't have statistics (anyone want to chime in?), but I'm pretty sure that far fewer than 87% of techie women even have kids, never mind having a sufficiently disproportionate share of raising them to keep them from being hobby programmers. I base this intuition on my intuition that way fewer than 87% of the male FOSS contributors have kids, and presumably kid-having is equally distributed between the male and female portions of the population.

...so, uh, do you have a 6th try at explaining why it is like totally not the men's fault not even the ones who like to harass women for fun?

Now we are back to square one

Posted Sep 30, 2007 13:17 UTC (Sun) by khim (subscriber, #9252) [Link] (4 responses)

Even if we'll manage to find as many female FOSS developers as there are male ones - it'll only increase "talents pool" two times. Not a big deal.

Lack of female developers is not a problem worth fixing. It's can very well be symptom of something sinister in the FOSS community - that's for sure, but disease worth fixing by itself it's not.

And, BTW, what the big fuss is about: it's free software. You can fork any project you want - and if you'll be successful and your project will have >50% female participants you'll have the power to punish "male jerks" with impurity... Hell - even if just lead developer will be female it'll be enough to establish firm rules...

Why you are concentrating on changing existing, "broken" system ? Free software was not created when RMS convinced proprietary software developers that "software is like sex: it's better when it's free" - it was created when he left MIT and started coding. Why can not "women-friendly" FOSS communities be created this way ???

Starting a women-focused group

Posted Sep 30, 2007 23:19 UTC (Sun) by SelenaDeckelmann (guest, #47907) [Link] (3 responses)

At the risk of opening up a thread that was winding down...

I am the author of the Women in Technology article quoted at the top of the page.

Free software was not created when RMS convinced proprietary software developers that "software is like sex: it's better when it's free" - it was created when he left MIT and started coding. Why can not "women-friendly" FOSS communities be created this way ???

While much of this discussion has ranged outside of what I set out to talk about, I wanted to mention that I am part of a group of people (men and women) who started a software users group whose goal is to get more women involved in F/LOSS. This is a group specifically about programming. We meet in-person, in Portland, OR.

We are in our infancy, but are fortunate enough to have experienced programmers, and enthusiastic volunteers driving the effort. Over the next few months, I hope to have more to report.

Those of you here who have not read my article, I am very interested in what you think about the specific points I made in it. (here it is)

My perspective, in a nutshell, is that I have had an exceptionally positive experience both as a user and a developer of open source software. My goal in writing the article was simply to encourage more women to join us and share in that fantastic experience and community. I see F/LOSS as an important social movement, not just about the code.

It is self-evident that we cannot have F/LOSS without code. But we also cannot have code without people. I think that we all benefit from actively inviting into our communities people who are different from us - different genders, different ethnicities, different cultures or political beliefs. The discussion that occurred here is remarkable because of the willingness of people of polar-opposite opinion to (mostly) rationally discuss a very difficult problem. If there is one advantage the open source community has relative to the rest of the world, it is an abundance of people willing to listen to other points of view and be persuaded.

And with that, thanks for listening. Feel free to contact me directly. I am fairly easily found through google at this point.

Starting a women-focused group

Posted Oct 1, 2007 0:39 UTC (Mon) by evgeny (subscriber, #774) [Link] (1 responses)

> While much of this discussion has ranged outside of what I set out to talk about

Indeed.

> I am part of a group of people (men and women) who started a software users group whose goal is to get more women involved in F/LOSS. This is a group specifically about programming.

I wish you good luck (though somewhat skeptical about groups formed to follow more than one agenda, and mostly orthogonal ones at that).

> Those of you here who have not read my article, I am very interested in what you think about the specific points I made in it.

OK, I did re-read it, and I guess the only point which I disagree with is ...

> I think that we all benefit from actively inviting into our communities

... that word "actively". Perhaps we mean it differently, but to me, it associates with "forcibly". I hate it when _I_ am "actively" being involved into something, so expect others might find it rather unpleasant, too. I believe the internal barriers (== discrimination, doesn't matter against women or any other subset of the mankind), in cases when they exist, should, if possible, be dealt with instead.

PS. Oh, and the title. Perhaps you believed that "To Sir, with Love" would ring the bell and make the connotation obvious. It didn't for me - not until I googled for it (and that happened only after I had read the article, confused by an apparent lack of connection between the title and the body).

What I mean by active

Posted Oct 1, 2007 1:21 UTC (Mon) by SelenaDeckelmann (guest, #47907) [Link]

Thanks for having another look.

... that word "actively". Perhaps we mean it differently, but to me, it associates with "forcibly". I hate it when _I_ am "actively" being involved into something, so expect others might find it rather unpleasant, too. I believe the internal barriers (== discrimination, doesn't matter against women or any other subset of the mankind), in cases when they exist, should, if possible, be dealt with instead.

Fair enough. I meant active as the opposite of passive. Speaking for myself, I did not think anyone would want to hear my opinions on technical subjects in a user group - until a friend asked me to speak up. My point in the article was that asking someone directly for their opinion, or letting them know that their participation is wanted, can encourage a person to participate. That's very different from forcing people.

Honestly, I don't think that it would be possible to force women into IT. It is certainly possible to invite them. I've "recruited" a few women from other disciplines to get CS degrees myself.

PS. Oh, and the title. Perhaps you believed that "To Sir, with Love" would ring the bell and make the connotation obvious. It didn't for me - not until I googled for it (and that happened only after I had read the article, confused by an apparent lack of connection between the title and the body).

I'll try to pick a better title next time :)

Good luck

Posted Oct 1, 2007 5:16 UTC (Mon) by khim (subscriber, #9252) [Link]

Wish you success - if you'll succeed it'll be sample which can be pointed out as "you don't need to tolerate jerks and misogynists to achieve success". Just don't forget that for most men "creation of place where both women and men can participate equally and where women do not feel ostracized" is not a success by itself (you successfully created yet another way to waste time? congratulations! we sooo don't need this). Something used by all major distributions or something touted as "major new feature" in one of them - this will be a success...

P.S. May be "just" 200% increase in number of Linux desktops in area can be considered success (but here I'm not sure: since it does not have global impact it's easy to say that such increase was a reason why your group survived, not a consequence of it's existence).

Ok. If the issue is "problems faced by women in FOSS" then where is the list ?

Posted Sep 30, 2007 13:30 UTC (Sun) by evgeny (subscriber, #774) [Link] (4 responses)

> The only facts I can see in this thread consists of requests from you that other people should prove a negative. This is of course impossible.

What is impossible? E.g. LKML archives are there for everyone to look & search. Show me examples of female participants being treated rude based solely on the gender issue. Have problem, eh? On the other hand, the flame level against males there is very high ;-)

> It behooves us to determine why they choose not to participate in this one, and if possible fix it:

Good. And what does one get trying to pinpoint the possible causes? An extremely helpful anything between "look around yourself" and "you are the problem". If anything, this could only worsen the deal.

> And they *do* participate in proprietary software development in larger numbers.

And certainly you couldn't imagine there are other factors beyond the sucked-out-of-the-thin-air *isms... Like "participation in proprietary software development" provides immediate revenue in the form of cash - which perhaps helps tolerating assumed abuses. Just one of the zillion of possible explanations.

Ok. If the issue is "problems faced by women in FOSS" then where is the list ?

Posted Sep 30, 2007 14:12 UTC (Sun) by nix (subscriber, #2304) [Link] (3 responses)

The problem I see is newbies being treated unnecessarily nastily. This
drives all sorts of people away: it is plausible that it preferentially
drives away women (also Japanese and others from cultures where formality
is prized).

I don't see much sign of women being treated nastily *because* they are
women, but that may in part be because there are so few around, so you
rarely see them treated in any way at all: but then we've already had some
pop up on this very thread saying that they were treated nastily (although
as far as I know not on free software mailing lists).

I've seen several projects nearly die because their maintainers drove
everyone away. In some cases the project failed: in some it limps on with
one or two maintainers: in some the project forked. In a sense this is
Darwinian and self-correcting, but notice that the criterion here wasn't
anything technical: it was social... and some of those projects *did* die
(I'd have to dig a bit to fish out their names, though).

Ok. If the issue is "problems faced by women in FOSS" then where is the list ?

Posted Oct 1, 2007 12:59 UTC (Mon) by tzafrir (subscriber, #11501) [Link] (2 responses)

Where are Women/Japanese/newbies being treated harshly? On LKML? that is a technical mailing list, so newbies should indeed do their homework before posting there. And if they don't , they will be flamed for wasting the time of everybody.

If they don't know how to post, they should go to #kernelnewbies instead. That's why it's there.

I try to be nice to beginners. But as you are probably well aware, some of them simply don't read basic documentation or do basic search (Even after being told to. Nicely). Sometimes being harsh is the only way to get the message through. Sometimes even this won't help, and you have to ignore requests for help, or (im)politly answer that you won't answer. And it is not always easy to tell if someone who is asking a question has made an honest effort first.

If I would spend 30 minutes double- and triple-checking before answering to newbies, newbies simply won't get answers and will get frustrated. And answering their questions won't be fun enough for me, so I won't do that.

It's not matter of doing the homework

Posted Oct 1, 2007 14:47 UTC (Mon) by khim (subscriber, #9252) [Link] (1 responses)

It's clash of cultures. Japanese (and to the lesser degree a women) expect to get all criticism in private - never in public. That's just how they were raised. But you can do tons of homework yet still get flames on LKML (it's kind of hard to do everything perfectly from the first try no matter how much homework you've did). Often it's enough to "close the story": they'll go away and will try to invent closed source API to avoid this public embarrassment (among other factors). Of course hackers culture (very much alive on LKML) says the direct opposite: everything should be discussed and all mistakes should pointed out with brutal honest in public.

You can say that it's Ok to ignore quirks of women and Japanese (and thus drive them away) - it's your right, but don't try to say that they are wrong, not you. Actually the sad truth is that there are noone who's clearly wrong - it's clash of cultures and representatives of both honestly believe they are in the right...

It's not matter of doing the homework

Posted Oct 2, 2007 20:42 UTC (Tue) by nix (subscriber, #2304) [Link]

Excellently put. (It goes to show, even if someone *is* intrinsically evil
because they disagree with me, they can still be right. ;} )

Proprietary developers are in different position

Posted Sep 30, 2007 14:33 UTC (Sun) by khim (subscriber, #9252) [Link] (5 responses)

Don't know about your employer but in my case I was forced to study "code of conduct" rules and signed paper which signifies that I've read and accepted them - and I can be fired if I'll violate them. There was sizable chunk about "how not to offend female co-workers". I've not seen anything too restrictive from my viewpoint (I do not use soft porno pictures in my presentations, for example), but I can easily imagine guys who'll hate these rules, but will accept them: employer is employer, who pays sets the rules. But now these same rules are pushed in FOSS world - and I see why it does not fly: "LinuxChix is not my employer and not my mom, why should it set the rules for me ?"

When you can give better answer to this question than "we deserve to be treated with courtesy and respect just as any human being" - you'll be ready to set the rules. Because right now most "jerks" are offering you the same amount of "courtesy and respect" they offer "just as any human being": none. Their peers are accepting it, why can not you ?

P.S. It all looks like a story with topless bar: when feminist organizations started to complain male waiters were also deprived of the top... Here we have something similar: the same principles applied to male and female work differently so it's hard to achieve discrimination-less position...

Proprietary developers are in different position

Posted Sep 30, 2007 15:12 UTC (Sun) by nix (subscriber, #2304) [Link] (1 responses)

Your employer is a lot more... excessive about it than I am. (Perhaps one
advantage of working in the UK is that employers assume common courtesy
rather than slamming down what is by the sounds of it a ridiculously
overdone spiked iron fist and mandating it?)

I find it interesting that you consider that treating other human beings
with courtesy and respect is *optional*. I suppose it is, if you don't
mind driving most people away and having them call you nasty things when
you're not there. It's a `should do' on the same level as `remember to
shower regularly': if you don't do it, most people will dislike you. (But
perhaps you don't care about having everyone else dislike you and stay
away from you.)

Your comparison of software development to working in topless bars doesn't
even deserve a response. If you seriously think that the two occupations
are comparable, you're beyond help.

(It's amusing to me that I, a diagnosed autistic who had to *learn* this
stuff over several decades and am still very bad at it, am now having to
explain it to other people. I'm normally the one who makes social faux pas
and has to apologise for them: but perhaps the difference is that I care
if I'm doing so, and try to avoid it. Your words strongly imply that you
think that if you piss off other people through being unnecessarily nasty
to them, it is in some way *their* fault. I hope that attitude doesn't
land you in court or jail someday, is all...)

Anyway, I've had it with this thread.

Finally! Breaktrough...

Posted Sep 30, 2007 17:31 UTC (Sun) by khim (subscriber, #9252) [Link]

My employer is US company. They need this paper for accidents. The "no discrimination" hysteria in US is such that company can be sued if it can not prove that it did everything it could to avoid "unjust discrimination". So if they got the paper and someone called female worker "bitch" and than said "bitch" sued the company - it can show that it's employer's fault: he signed papers so was warned.

As for courtesy and respect - yes, that's the problem. Respect is not a right. It must be earned. And optionality of needless courtesy is often considered a feature among FOSS developers.

And the topless bar story just shows that the same rules when applied in the exact same way to male and female can be acceptable for male and offensive for female. If the situation in FOSS is not like that then my comparison is wrong, of course, but then the question arises: what's the hoopla is all about if rules happily accepted by males are driving females away ?

Proprietary developers are in different position

Posted Sep 30, 2007 22:32 UTC (Sun) by mepr (guest, #4819) [Link] (2 responses)

The person who said something along the lines of "I can't think of examples of females being treated badly for being females, but maybe it's just because they are so few" is on to something.
The people who said that women are socialized out of programming were also largely correct, I think.
There is a woman who became a member of the LUG in my town. Her story is interesting. She never learned to play chess because nobody ever suggested to her that it was a good idea. She never learned to program for the same reason. At the same time, her brother was encouraged in these same things, and went on to be a programmer.
In her early twenties, she tries chess and finds out that she's good at it. Then she tried programming and found out that she's good at that. Within a couple of years she was a Debian developer, and currently maintains a fairly well known Debian package.
Unfortunately, she quit going to lug meetings, because despite repeated, direct comments on the subject by her and others, there were a few of the males in the group that insisted on making highly unwelcome, highly sexualized and demeaning remarks.
What is so difficult to understand that there are very few people willing to put up with such things for very long? I have long held the theory that women avoid working in IT because they would rather work in one of the many fields where they will be treated with a modicum of respect and not live through incessant pissing contests at work.

And that is small model of the story of "women in FOSS", right ?

Posted Sep 30, 2007 23:11 UTC (Sun) by khim (subscriber, #9252) [Link] (1 responses)

She quit going to lug meetings, because despite repeated, direct comments on the subject by her and others, there were a few of the males in the group that insisted on making highly unwelcome, highly sexualized and demeaning remarks.

Yup. And when time come to choose between "one intelligent woman" and "few male jerks" majority choose "few male jerks", right ? If behavior of people does not change and it's not acceptable - they must be removed from the group. That is the question which should be discussed first: are members of community (LUG, LKML, gentoo-dev, etc) ready to kick out few male members to make women fell comfortable and who exactly can be kicked and who will not be kicked - no matter what. Then you can decide if you want to try to "make women feel welcome". It's tough question - but without answering it you'll only get flames and steam...

And that is small model of the story of "women in FOSS", right ?

Posted Oct 1, 2007 21:09 UTC (Mon) by mepr (guest, #4819) [Link]

Yup. And when time come to choose between "one intelligent woman" and "few male jerks" majority choose "few male jerks", right ? If behavior of people does not change and it's not acceptable - they must be removed from the group. You're right, that's essentially what happened.

Given the dynamics of the group, it would be quite difficult to kick out a person for such a thing. The local LUG is largely run as a very informal, loose association. It was difficult in the particular (non-isolated) incident I am thinking of, because one of the people involved was well liked by most of the group, even though most people there were not happy about it.

However, I am glad to say that the kind of behavior we are talking about is not tolerated on the LUG's mailing list. The rare unacceptable comment generally receives a very strong response and we have never had to kick someone off of the list.

BTW, I hope that my comments (and the summation of this thread) don't make it seem that people who care about F/OSS software are bad people. In fact, a great deal of the reason why I have been involved has been the idealistic community spirit shared by many of us, and the unusually high density of principled people and deep thinkers.

In fact, I usually find that people who love F/OSS software are usually wonderful people that have wonderful significant others. Also, I think the majority of us recognized that we are diminished by having too few women around.

However, I wonder if a culture that is perennially stocked with immature males will take a long time to grow up.

Also BTW, WRT groups that promote the involvement of females in F/OSS, they are performing a needed service. It is just the way of things that people need role models and mentorship, and it is also the usual way of things that people need a mentor that is like them. Isn't the wizard/apprentice one of the oldest stories in hackerdom? If it is more likely for that bond to form between people of the same gender, then such groups provide a place to go for women that look in their immediate environment and don't find a same-gender role model.

Ok. If the issue is "problems faced by women in FOSS" then where is the list ?

Posted Oct 2, 2007 3:50 UTC (Tue) by njs (subscriber, #40338) [Link]

And as for `just look around': can you find *any* other field of human endeavour with a percentage of female participants as low as free software development? It's damned rare (perhaps pure maths?)

Pure maths -- no way. In the US as of 2001, women get >27% of math PhD's, hold >17% of math postdocs, and >16% of tenured faculty positions in "mathematical sciences". None of these are quite pure pure maths, but. (Source: nsf04317).

In the US there are a higher percentage of female *construction workers* than FOSS contributors -- 2.5%, which is slightly higher than the *highest* count I've ever seen for FOSS. (Source: figure 19c on this page of "eLCOSH"; note that unlike 19a and 19b, 19c is specifically excluding managers, secretaries, etc. -- these are the people actually out doing the work.)

Apparently we are at least doing better than, umm... welders. If we look only at welders, than we're doing great. Makes you feel better, don't it? :-/