"Gender bias in open source programming "

Yeah right FUCK OFF

I would say just one thing: why is that notion so offensive to you?

That's stupid. For 99.9% of the open source projects I use, not only I don't know if the creator was female, but also I don't even know their name or who they are. And even if I do, who gives a *$&£?
I could be wrong, but that's my experience. Any developer woman who will share their perspective on this?

Somebodies missed Sociology 101

@lordmirziteh if that is sociology then I hope everyone misses it.

@bioDan yeah studying how society works should totally be missed by everyone in society.

What i really like about devRant is that people don't bring their political and social discussions here. Please, let's keep it that way?

Yeah, it's a thing.

Consumers of the software don't care, but the communities that build them are often really toxic to women.

A friend of mine was an avid open source developer, and when it got out that she was a woman (and a trans women at that), she got hit with tons of incredibly inappropriate conduct from other developers-- everything from sexual harassment to death threats. When she reported it, she was banned from the community.

Look up the Hypatia Game Engine (it has another name now). There's a talk my friend gave at MIT about the experience and an org she started to help improve the situation.

@lordmirziteh true if 'studying society' is as vague as "gender bias in open source"

Imho,
Stochastic models > Sociology

@starless I feel what you're saying.. But don't you think you're overly generalizing and exaggerating by saying that communities that build software are often really toxic to women?

@bioDan no she is not. Just google «silicon valley sexism». You wil get articles from the economist, forbes, bbc, atlantic, guardian, all quoting research and experiences about silicon valley being super hostile towards women. «60% of women experience harassment». Not discrimination, HARASSMENT! And i know i said «take political and social discussions elsewhere», and this is just me pointing you in the right direction.

@disolved thanks for the information, I was unaware of the hostility towards women in silicon valley.
I know silicon valley is the capital city of startups and many titan corporations are based there. Surely perpretrators should be persecuted on an individual basis according to the law which is already in place against gender discrimination.

But I don't think it represents the views of both male and female developers around the world.

Programmers just care about code. Its beauty, its logic and manifestation. I fail to see how it's related to gender.

@bioDan just as you don't have to understand why people are racist or agree with racism to believe that racism is real, you don't have to understand why people are sexist or agree with sexism to believe that sexism is real. Sexism is very real, and some places in tech it's even rampant. Maybe you have a romanticised view of developers/programmers? Remember that every country has different cultures and sub cultures. Tech people in Norway, India, California and Spain (for example) are all completely different animals.

@disolved I know, I lived in the US, and in Israel, collaborated with developers from Mexico, Australia, India and Europe.
I've been in business for more than 10 years.

And I appreciate you're trying to lecture me about sexism and racism. I know they exist and I know they are a problem in general.

Specifically taking about our industry though, from my personal experience and conversations with many female collegues, we agree that we are measured by our individual skills and characteristics.

@bioDan being respected in programming circles often involves 'looking like a developer'. This could mean linking Linux and techno, wearing jeans, all kinds of things that have nothing to do with writing code. If you check enough boxes, people you meet will believe you're a developer. If you don't, they don't. Being male is worth a couple of boxes.

It's basic human behavior and developers are still humans. I can link you to a couple of essays on this and it's origins, if you'd like.

Also, traveling to Israel as a developer was so so nice.

@bioDan Yeah I was implying our industry directly, not just talking about sexism is general. It's nice that you have experience from several teams where the females agree that sexism is not a problem. Good for them! But I still don't think its warranted that you have such a hostile reaction to research indicating that there is sexism in the open source community, which is what this post was originally about. The world is a big place, and just because your personal experience does not align with something does not mean you should brush it off as nonsense like that.

@starless I actually grew up in Israel.
And those little boxes you mentioned are just hollow shells. Superficial. Unsustainable over time.
"Fake it till you make it" mentality doesn't often work, and if it does - then only in the short time.

People buy stickers and stuff for the hype and because they love or are proud about a particular software or community.
You can't really build a career out of that.

Respect is earned by:
1. Being respectful.
2. Being decent.

( Or by humiliating and instilling fear, but that form of respect is despicable. )

Not by what you wear or what stickers you have.

@disolved
"such a hostile reaction to a research indicating that is indicating that there is sexism in the open source community which is what this post is about"

What rsearch? I thought it was a rant which i totally can relate to since I also believe gender has nothing to do with code.

No matter how you flip it, computer science doesn't care about your gender. It cares if your bit is flipped or not 😉

@bioDan you outline how it should be.

When people don't pay attention to your contributions, it's difficult to get recognized.

Some areas are better than others, but overall, it's still often detrimental to be a developer who happens to be female. We can change this together, but in order to improve this, everybody needs to drive it, not just the women.

@Lisanna you should have only one ++ not 4. It was almost perfect...
We will try again when this reaches 64. :(

@starless I agree. But I hope you also agree that there are cases where being a woman dev can also be an advantage.

I.e. this meme:

@starless wow. That's crazy. And they banned HER? Wtf.

@Joserc87 he was talking about the product, OP was referring to the process of developing it. OSS developers often treat female developers worse. There are github stats that back up discriminatory patterns in merges, etc.

@bioDan yeah, occasionally, it helps.

Often, if it's helping, I need to be careful that no one will try to coerce me into fucking them later, but it can be helpful. If nothing else, I'm pretty memorable.

Also, if you're an ugly woman, you generally don't get those advantages. Getting technical help shouldn't depend on how 'fuckable' you look. Imagine if in order to get your password reset, you had to hit on your web browser until the link worked, just for it to do its job .

Usually, it's kinda shitty. Sometimes, it's OK. It definitely doesn't balance out,and I'd trade in special treatment for people believing the first time any day of the week.

@Joserc87 she was 'causing trouble', because there weren't any death threats before she started contributing.

@starless thanks for sharing your views. They we're very insightful for me. 👍
I loved the analogy with 'how fuckable you look'.

On the other hand I'm inclined to believe that most men have the same advantages as the ugly women do. None.

They need to gain their reputation by the fruits of their work, as should all do.

@bioDan I think they don't start out with extra 'points', but they're more likely to have an easier time earning more.

For example, a male executive or leader will recognize a younger colleague that reminds them of themselves and mentor them. If there are more men than women in the company, female developers are less likely to get that kind of mentorship because they're less like the current leadership. Then, when someone else is looking for a new leader, the male developer already knows how.

There are some women-specific mentorship programs to help counter this effect.

It's no big thing that does this stuff, just the accumulation of lots of little ones. Sometimes it matters, sometimes it doesn't. I think it really depends on the area.

In my experience, Israel is a really good place to be a female developer. No one ever reacted with "Really?!", which is my usual reaction in the US. Instead, they just asked "What languages?", like you're supposed to.

@starless Nobody made death threats until she came? xD I'm sorry but that's just funny. Sounds like "Nobody created any concentration camps before the Jews came, so obviously it's not the Nazis fault; the Jews are trouble makers". Great logic. Bravo

For everything there's people who are assholes because of it. Of course there's a gender bias.

There's bias because of the fucking text editor you use, and bias because you drink tea instead of coffee and bias because you drive a bike and because the number if hairs that grow out of your nose is %2>0.

Why shouldn't there be a gender bias?

@Zaphod65 because it has happened in other places. It starts off with that. Later the community is hijacked by people that come in that want special treatment based on their genitalia. If you don't provide, you're labeled as toxic, and banned from the community. The news start to cry that the community is full of toxic individuals, etc.

When all we wanted to do is code, and all that we saw were the brains of the creators of software. Genitalia is literally the very last thing on my mind when looking at any code, and I'm sure that is the case with the vast majority of people in this field. Yet right now we are already being labeled as sexists. You have nothing against it?

@bioDan I really agree with your values and views of how things SHOULD be, but in order to actually achieve this dream world the first step is to admit that it's not necessarily like that now. You say that your female coleagues told you there are no problems, but have you considered what would happen if they DID say there were any peoblems? These females have good reason to believe that people like @letmecode (reffering to his earlier comment here) would come crawling out the woodwork and call them feminazis and sjws. Being a female in tech can be a constant quiet battle to be taken seriously, and calling someone out on sexism (concious or otherwise) is gonna make people take you less seriously.

I really don't give a damn who makes the code, it either works securely or not. The gender push in any field is utter bullshit. Gimme a ring when they fight for more women in construction jobs and more men in nursing.

@starless what you're trying to express (attributes that make someone look like a developer) is a psychological bias called representativeness, first defined by Amos Tversky and Daniel Kahneman. It's going to exist in every sphere where the majority can be observed to exhibit certain attributes over a longer period of time. You can always expect this bias to influence people in a certain way, but in every developer community that I've been part of, I've seen women get preferential treatment. Women can ask ridiculously stupid questions and not get completely destroyed in the replies. Women that are actually good at it often are seen as leaders of opinion, and get more respect in the field than the equally experienced men.
Of course this is anecdotal evidence on my part, but as long as there is no study that takes into account representativeness as a bias, and determines a systematic sexism in the field, I will call bullshit, because that's the only true way to accept information.

@disolved you misrepresent @Letmecode.
(also made a typo in the reference)

I happen to agree with his views which are actually a consequence of the actions and views often held by third wave feminists and sjws.

@apisarenco from what I can make from your reply, you're saying that "no one has done a study of the impacts of propagation of unofficial 'mentees' in the workplace" and "without a study like that, sexism doesn't exist".

I must be misunderstanding you. Care to clarify?

I wasn't actually trying to start such argument. Maybe if people stopped complaining about such random facts ? Isn't open source software already OK? Why try to cause havoc because of gender bias?

@starless in order to make a positive claim "something specific exists", evidence must be brought forward. Evidence comes in the form of aggregated observable data that shows that despite all other biases, there is a discrepancy that suggests another bias (sexism in this case). Anyone who claims that there is this specific type of sexism, without any hard evidence, is a bullshitter. With the same success I can bring my own anecdotal evidence of the contrary, and we'll be sitting in a stalemate, where you'll throw me your examples and I'll throw at you mine, while neither of them matter in any way.

@apisarenco thanks, I also know how science works. My assertions are based on statistics from github, as well as peer reviewed studies on systematic gender bias in academia.

Studies on gender bias in the workplace are more difficult to conduct, so to the best of my knowledge, don't exist.

We can use Occam's Razor, and suppose that since we don't have any evidence that sexism in industry CAN'T happen, it's most likely that it does and is likely propagated the same as in academia. Furthermore, there are some pretty clear stats from github on the matter that suggest its happening in development in isolation, too.

There is a strong scientific backing for the presence of sexism in development workplaces. There is no need for me to resort to anecdotes, unlike what you've demonstrated so far.

Pls stop this has gone far enough

I like how people here are saying "This has never happened to me nor my female colleagues and therefore it's not a problem".

If someone says they're hurting in cancer and that it's a big problem in the world, do you apply the same logic? "I've never had cancer and therefore I don't consider it a problem for anyone".

Of course not all men are sexist, and by purely stating that sexism is a problem in our industry is not saying that you are a sexist. Not at all, it's about SOME asshole people being ignorant fuckers making life hard for women in tech for no good reason. No need to take this personal (unless you are a sexist, then you can fuck off!).

I also agree that it's the code that counts, gender does not matter to me. But that's apparently not the reality for everyone in this world. So please, let's just accept that this is a problem for a lot of women and let's all try to make the world a better place for everyone ❤️

Pls don't stop I'm enjoying this to much 🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🇰🇵🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿👍🍿🍿🍿🍿🍿🍿🍿🍿🤗🍿🍿🍿🍿🍿🍿🍿🍿🍿😀🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🌝🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿🍿

I quite agree with what @simeg, and @starless said. Because you haven't experienced it doesn't mean it doesn't happen. Where I am from, you don't get see many female developers though this being addressed. I believe to stop it, there must be effort from everyone. I was once at a tech meetup where someone was making what I would consider a sexist joke. I told them that it was sexist. The lady who was the target of the joke though didnt seem to mind. We know a lot of us don't care about who a person is, only how good they are. But we should also speak out when we see such things happen. I personally don't give a rat's arse about your religion, political affiliation, race, gender, ethnic group, sexual orientation e.t.c If you can code, that's all I need to know. Bias based on OS, IDE, text editors, indentation style, repository e.t.c I believe are just for fun though I don't judge people based on that.

@simeg since you like it how.. can you find me the person who said it never happened to him or his colleagues?

@Letmecode you sound like a guy that values argument grounded in reason and rationality, and yet you uncritically blurt out that a claim of gender bias in open source programming is a feminazi SJW thing, even when there are statistics documenting it.

@Letmecode im not bashing on you for using the words feminazi or sjw, I'm pointing out thay you automatically spew those words on reflex, just because you see a claim of gender discrimination that you don't personally agree with.

@Letmecode Also, when I call you out on your behaviour, you for some reason try to make this a discussion about the words feminism, egalitarianism nd feminazism. Dont. We don't care. It's getting old, we've heard it all before. This discussion is about discrimination of females in programming, which for some reason you and others try to discredit and refuse to admit is real even when there are statistics and research shoved in your face.

@g-m-f «Social Justice Warrior» is a derogatory term for people who fight for social equality and justice.

@g-m-f just google it... and then burn the machine, pls.

@Letmecode what are you talking about? Whether you did it on reflex/blurted it out or not doesn't make a difference to the validity of my argument. Actually the fact that you didn't do it on reflex actually makes the implications about you even worse for your credibility. Are you telling me that even when you KNOW there are OBJECTIVE statistics out there pointing towards gender discrimination in OSS, you still just stick to your opinion and flame people who argue based on science because for some reason your subjective experience and belief is worth so much more than statistics and research? And you still have the audacity to say that everyone that don't agree with you are irrational?

@disolved I don't know what's your argument. Can you please state it?

Like @Letmecode said, no one is for discrimination against women, you can stop trying to make him look sexist. Obviously he's not.

Broadly speaking, the statistics are not the problem, it's the interpretations for their reasoning & conclusions people derive from them.

@bioDan @Letmecode Ok, Im sorry if I've been unclear, and if I've missunderstood or missrepresented your opinions. I am definitely not calling anyone here sexist, and I do not doubt your values when it comes to egalitarianism and gender equality. My argument is this: There are studies and statistics showing that in general, OSS contributions made be anon females tend to be accepted more frequently (or as frequently) than those made by anon males. On the other hand, when the gender is identifiable this trend changes, and female contributions are rejected more often than male. Considering this evidence of gender bias, it is almost irrelevant to the discussion whether males like yourselves or even females have anecdotal evidence from work environments or in OSS environments to the contrary. The discrimination is not visible to you, it is a statistical significance, not necessarily anything explicit. This does not however make it any less real. Are we on the same level?

@disolved yes. Thank you.

My issue with @Letmecode was the way he pulled feminazism and sjws into the discussion, and as I understand it attempting to discredit the claim that gender bias is a problem is OSS. By attributing a claim to feminazis it immediately looses credibility, and I think he knows this. As i understand it, gender bias in OSS does not fit in with either of your world views and personal experiences, and therefore it makes perfect sense for you to explain away studies and statistics pointing to gender bias in OSS as being humbug. In my world view, studies and statistics are more reliable than human opinions and perceptions (maybe because I've studied bias, reason and cognition) and that's why I discredit your personal experiences and opinions so easily when I know there are more objective and trusted sources of evidence out there.

@Letmecode ...I can't make you read it, but here is the actual peer reviewed article outlining the study. https://peerj.com/articles/cs-111/

@starless what starts from GitHub? Did you account for other biases to conclude that it can't be anything else? Did you account for the base rate even (how many women join the industry)?
There are also no good studies (with a large statical population, that would account for biases, whose results were reproduced) that would show this sexism anywhere in the first world. Peer review isn't what it used to be. Right now contradicting papers get published every day, and papers are publishing even "troll" articles, ESPECIALLY if it features gender stuff. So for me, evidence would be a scientific study, the would look like one, that would have a big enough sample size so that the law of large numbers would be considered, that would account for the phenomena observed by psychologists in the last 50 years about human biases, and the results from which were replicated. Because that's how science works. It doesn't depend on an approval committee.

@apisarenco you seem pretty confident that it's impossible that evidence of sexism exists.

Have you ever actually sought out evidence, though? A Google search of "github gender bias study" and this (http://bbc.com/news/...) was the top result. Try a bit harder next time, yeah?

There's a lot of really loud people who bicker about shit that doesn't matter in the name of gender equity. They get a lot of press. It's pretty shitty. It's also kind of rediculous to discount every other bit of evidence you might see for this phenomenon because of a loud minority. This can't be the first time a woman has claimed about sexism to you.

Thanks to that loud minority, feminism, "the radical concept that women are human beings", as Gloria Steinem put it, is a dirty word. In my experience, most guys don't know this and just assume it's some kind of equal opportunity program on steroids, and so continue to treat people badly because they won't learn better.

@starless look, I don't want to be confrontational, but if you get personal - I will. Quote from the article that you referenced but failed to read "The paper is awaiting peer review".

What's shitty to imagine a boogeyman that makes you look as a victim and behave inadequately to anyone who questions that bullshit.

Bring evidence, or STFU

@apisarenco here it is, peer reviewed and pretty darn solid: https://peerj.com/articles/cs-111/

@apisarenco I don't need to imagine a boogeyman, I've experienced plenty first-hand.

I do think there's an issue with 'victimhood' and purely experiential politics. What's bullshit is that when someone says something like "rape can be really traumatic", it's an invitation for hostile 'debate' until the speaker says something like "I should know. I've been through that." and suddenly, people treat the issue like it happens to human beings.

I think the issue is that people respect victimhood more than the victims themselves. Once that changes, I think people will stop claiming it as an identity.

This rant is getting heated up. You guys might wannna take a chill pill. Argue constructively and not take thing personal please. I would hate to have to use the -- button for offensive comments

@disolved thank you for a link, finally. Now after reading it entirely, I must applaud the comprehensive work done by them, but at the same time observe a few flaws:
1. It is shown that women are doing better in Open Source community.
2. It also shows that BOTH genders do worse, if their gender is in the profile, in BOTH insider groups and outsider changes.

One would think that the 2nd issue would be important, and segmentation would be done to determine, say, experience or contributions based on how much time they've wasted beautifying profiles on GitHub.

Also it's important to take note about the sheer difference of sizes of statistical populations that are compared. It is obvious that there is a strong survivor bias at play here, given that, obviously, a lot more male hobbyists will turn to modifying a few things in open source projects, than female hobbyists, since the later will probably not "survive", given how few members there are.

@apisarenco to your point number 1: yes, ANONYMOUS females are doing better, while gendered female pull requests are declined much more often than any other group, including gendered male pull requests. The fact that anon females are doing better than anon males strengthens the evidence that there is gender bias at play when gendered profiles are discriminated. Your point number 2 is actually invalid in a statistical sense. It does not matter if gendered male profiles are also less likely to be accepted as outsiders, because female gendered profiles are *even less* likely to be accepted, despite the fact that female profiles without gender on them get accepted much more often as outsiders, as you can see from the graph and numbers. There is a 12% drop for females, but only a 3.8% drop for men. Is is this 3x difference in acceptance rate drop that is important. I am not quite sure what you mean with the survivors bias in this case or how it's relevant.

@disolved "The fact that anon females are doing better than anon males strengthens the evidence that there is gender bias at play" - wrong. It only suggests that this might be an explanation. The far more plausible explanation is that the few females that do end up as coders are far better than the script kiddies.

A point cannot be invalid in a statistical sense. Ignoring an obvious discrepancy is failure to segment and identify the variable responsible for this shift.
It is bad enough that the study is an observational study, and not a set-up test with a control group, but here we have you defending ignorance. How can I express the severity of NOT HAVING A CONTROL GROUP? A control group is the statistical group that differs from the study group only by a single variable - the one that is being tested, and are otherwise statistically uniform. That's the only way to reliably determine correlation without accounting for thousands of variables. Otherwise you must account, and not ignore.

@disolved so why is it important to account for that variable? Because you're comparing apples to oranges. The groups are hardly comparable. A ton of assumptions was made in the whole study, like the assumption that the volume of code translates to quality, or that females who completed their profile don't differ at all from females that didn't, and are exactly the same as the males that completed their profile, aside from the difference in sex.

These assumptions must be proven first, by comparing obvious males from non-obvious males, and the same for females. Only if it's proven that the obvious members make EXACTLY the same contributions as their non-obvious cohorts, then you can infer that the accept rates should be the same.

Because what's more likely: That code reviewers look at profiles instead of code, or that people are different, and their actions are testament of those differences?

@apisarenco you're laying an undue burden here. First, you want any evidence, then, you wanted peer-reviewed evidence, and now that's not good enough for you, either.

If you wanted to convince me that sexism didn't exist, that there is absolute parity between the treatment of developers, regardless of gender, how would you do that? Where are the statistics?

@starless and evidence was brought, and I tackled it. Since the evidence showed a very little discrepancy, AND failed to account for one major source for discrepancies that just laid there in front, I mentioned that.
There are a multitude of things that CAN BE. Discovering a discrepancy and arbitrarily attributing it to one thing is stupid. It's not science. Normally, such articles should conclude with "might suggest gender bias" and "more research is needed" because they clearly did not conduct it in-depth. Great work, don't get me wrong, but when things like this can be found - follow up studies are necessary BEFORE jumping to conclusions.

I don't have the burden of proof, because you still haven't met yours. I'll admit that there is gender bias against women in most of Asia and Africa. You have to prove however that people in countries like France, Germany, USA, look at genitals instead of code. Because that's the stupidest idea I've ever heard.

@apisarenco it's great to hear that you don't think sexism is an issue in the western world. Where do you work? I'd love to apply.

In all seriousness, though I get the feeling you haven't read many experimentally-driven papers. It's very normal for an individual paper to determine that more study is needed, and generally, one never finds a single, perfect, self-encapsulated paper that irrevocably proves one thing either way. That's just not how science works. Fields move ahead slowly, inch by inch, and occasionally, a breakthrough occurs and they move ahead a little more.

Despite a mountain of evidence that, when combined, paints a picture that strongly suggests the presence of a small, sexist bias, you seem to insist that your personal analysis of the information trumps that of scientists who work in that field.

You are not an expert in this field. You are making an extraordinary claim "no sexism in development in the west" and me believing you requires extraordinary evidence.

@starless "though I get the feeling" - that's System 1 talking. No need to voice what it says. It's a fast assesser, but often makes stupid conclusions.

"very normal for an individual paper to determine that more study is needed" - yet they did what? Correct! They jumped to conclusions. Confirmation bias or availability heuristic at its finest. Because who needs thorough analysis and rational assessments of observations?

"that irrevocably proves one thing either way" - good research should always tackle the obvious. It is obvious that identified males get their code less accepted than non-obvious males. Yet that was ignored. Because of an obvious bias.

"mountain of evidence that" - evidence of differences. That also suggests a bias in favor of women. Depends how you look at it.

"suggests the presence of a small, sexist bias" - as one of many explanations.

"you seem to insist" - I insist that facts that are THERE should not be ignored. You seem to advocate for ignorance.
(part 1)

@starless (part 2)
"You are not an expert in this field" - you made an unfounded conclusion, yet again.

"You are making an extraordinary claim" - no. Nothing is extraordinary in the idea that people are treated equally today. You insist that bias exists. No evidence that points specifically to that bias was presented. Your claim therefore should be ignored until further developments.

@theScientist in the US, I often need to somehow 'prove' that I'm a developer. It might be telling the person three times, asking really sharp technical questions, or whatever. In Israel, when I said I was a developer, they just believed me. I think it might be partially because of how they militarized some stuff and draft straight from high school, but am not sure. Culture is friendlier for sure.

There's also some really cool data analytics stuff going on out there.

May be you didn't read this paper: Gender differences and bias in open source: pull request acceptance of women versus men