Diversity in Technology and Open Source
lucumr.pocoo.org
lucumr.pocoo.org
If you're a male, you have to be careful writing out suggestions like that because to females such as my mother with a career in electronics, it is patronizing. Women can also be a "strong engineers" and they shouldn't be relegated to "open source housekeeping duties" as a dangling carrot to attract them.
Yes, Armin Ronacher means no harm but I think men often fall into the trap of thinking they are championing women's causes when they are actually insulting their intelligence.
It also doesn't (at least in my opinion) help the cause of balancing the industry to be telling engineers that the only way to do so is to emphasize soft skills and deemphasize competition (and subtextually with it technical excellence).
Surely you and Erin both recognize that it's possible that both (a) women, generally (compared to men), 'disfavor zero-sum competition'; and (b) Erin favors 'zero-zum competition'.
[I'm not claiming that it's true or even that there's significant evidence that women 'disfavor zero-sum competition', absolutely or relative to men.]
A small leap I will make is that people of both genders look towards gendered emphasis to validate their choice. If its for their gender then it is a safe choice, and if its not then doubt arise. The effect being a small bias that depending on what one is doing can have a rather extreme results.
When someone says "men prefer X" or "women prefer Y" they are quite obviously speaking statistically.
But people reading it will see the following words that are all in the same semantic space as "gender":
>"men", "woman", "women", "male", "someone" (implied female)
As used in sentences:
>When your team is 4 men, the first woman which joins will make a significant impact.
>When your team is already 20 men you need to get a lot more women on board to have the same impact.
>Tech for recent historical reasons is very male heavy but society is not.
>We don't need more [...] more strong alpha males but people that are good in de-escalating arguments
>So when someone cancels a conference because the speaker lineup after a blind selection was 100% male it just shows how bad the imbalance
Therefore, it is not unreasonable to get the impression that the article is heavy on gender imbalance and not so much on skin color. (No explicit mention of "white", "black", "African American", etc.)
The line under discussion directly contrasts "males" with role diversity. It doesn't explicitly label the other roles as feminine, but it was certainly the impression I got when reading "not more strong alpha males but people that are good in de-escalating arguments".
None of this contradicts your understanding of what you meant with your writing, obviously. And the Unicode point is a good example of how a shortage of how other developer diversity (e.g. in languages spoken) can motivate more useful products.
But for whatever it's worth, I had to read this comment to find out that you weren't writing the entire piece with gender in mind, and writing that diversity of roles line specifically to state what women would bring to the table.
The author is in that point right, that there are more male programmers than female. But to solve this problem you have to start way earlier, with getting the interest to tech things of a child (regardless of gender) in school or even earlier.
How is it surprising that they work in software? It's like finding out that your welding instructor has worked as a welder; surely somebody who contributes to a field in public is more likely to have a job in that field in private, but does that mean that not having a job precludes you from having a skill?
I mean, I've been NEET going on half a year, and I'm contributing to open source projects, do I need a software job to do open source? I was contributing to open source projects before I was ever employed, do I need to have had a job to do open source?
So it's ok now to post a comment with a strong statement without any data?
So any theory of "open source drives away women" has to explain why it has the opposite effect on LGBT people. There isn't going to be a simple answer there and maybe we aren't even asking the right questions yet.
Disclaimer: I don't know how diverse the Gimp developer base is as opposed to a user friendly competing product.
It could be that the different kinds of humans are not identical and given an even playing field different outcomes will result.
It could also be that the lack of diversity in tech is the mirror image of the lack of diversity in other fields such as healthcare and education.
If that is so it could be that the "lack of diversity" is the result of diversity in human inclinations and it would only be stamped out through increasingly authoritarian measures.
For instance, open source contributors need free time, which is usually a consequence of relatively short working hours (i.e. being highly competent and educated) and short commutes (i.e. an expensive house in a convenient location), and not having to spend a lot of time, when at home, on housekeeping, children, old and ill family members, hostile environment, distractions, etc.
I agree with Armin on what he did mention, which is that there is great need in OSS projects for people with greater focus on skills besides just hard boiled coding.
I wish he had discussed the issue that seems to be creeping up in many places now (recent example: cancellation of Electron Conference) which is that there are many activist groups who are seeking to implement diversity quotas. That is a really scary idea and a step in the wrong direction.
Maybe white people like diving computers or lake sonar, maybe women care more about topography or teledildonics; but if nobody cares to even know the accidental characteristics of the people who send the patches, then is it an issue when the numbers bear that out?
But "diversity" is being used to mean diversity of gender, ethnicity, etc., as if it matters, as if it is a reliable proxy for those other things like experience and perspective and skill, when it's not.
When I was starting to program and talk to people on the net, nobody cared about my age or ethnicity or gender. I was just a name and an area code and some skills. The compiler certainly did not care about my ethnicity or gender. It was perfectly fair to me.
I will not go out of my way to try to rope people into my projects because they have different color skin. If people show up to help, I welcome them. I don't particularly care about their genetic code. I don't have any preconception about what the graphs of the ethnicities and genders of contributors would look like. Who cares? We are not our bodies and I won't help build a future where body attributes are considered an important part of identity.
In terms of programming, where the domain is very specific, I somewhat agree that it doesn't matter.
But in a more general sense, since it's life experiences that culminate into who a person is, their perspective will be based on their identity. And it will vary in unexpected ways.
If you're a large company in a diverse market like the American market or a global market, you don't want to appear tone deaf in terms of your marketing and your products.
So it's important to get differing perspectives before that new launch.
Sure, which is why your marketing team has to be familiar with regional cultures and customs.
It doesn't mean you need to account for every target cultural group on your IT support staff.
As always, context matters.
Obviously team diversity isn't the only way to achieve that gain, but it's a good one that's often lacking.
The "yes" argument obviously says compilers don't care about gender or skin color. The "no" argument says that arguments in favor of inequitable staffing are a form of the Just World Fallacy.
You need to do better than just restate those arguments to add something new.
So you can argue about whether that exclusion exists, but you can't shut down the argument by claiming that it's somehow racist/sexist against white men.
Low turnout does not imply someone is being excluded.
Different groups of people tend to have different interests.
> everyone agrees that such overt exclusion, if it exists, is racist or sexist.
> So you can argue about whether that exclusion exists, but you can't shut down the argument by claiming that it's somehow racist/sexist against white men.
I don't even know what kind of fallacy this is, but it's definitely out there.
"If this hypothetical thing we made up does exist, that would make your argument wrong, so you can't use that argument."
Whatever pie-in-the-sky claims you want to make, the direct, practical, and real-world effect of these policies is subverting meritocracy and discriminating against groups of people who are political targets (white and Asian men). It is absolutely both racist and sexist in any sane, non-doublespeak sense of the words.
(I'm ignoring the rest of your comment, since I can't map it onto anything I said or anything I'd want to respond to.)
"Please avoid introducing classic flamewar topics unless you have something genuinely new to say about them."
Yes, b6. It is. Your personal experience is of course unique. While your experience may have familiar notes with many people, it's fundamentally different from someone who grew up as a young girl. Your fears and the options presented to you are going to be different.
Identical people come to similar conclusions. Folks who have raised this idea of "diversity of thought" think that it adequately addresses this idea that diversity is a value add for software by saying, "Well it's all about thought." But those initial conditions our minds are so sensitive to are not easily ignored.
You cannot have it both ways in this. Either you simply don't care about diversity (which is a value decision I am not denying your ability to make), or you do and physical indicators that shape how society treats people are a high quality signal of diversity. You cannot rationally suggest both that diversity is important but that people's traits and backgrounds aren't.
People try to turn this into a value proposition here (as it seems like you may be doing) are taking half the situation into account. For a value proposition, you're looking at your project's upside. For example, a product primarily aimed uniquely at women would probably be best served by a qualified woman. But even in this, bias creeps in (the famous study results suggesting men are so acclimated to dominating the conversation that when women talk for half the time, men perceive it as oppressive). In this specific scenario, women have reported being hired and then simply not given the power to do their jobs, or not being present in larger orgs with sufficient numbers to create the impact they were hired to create.
But the diversity movement is also trying to reduce the downside for individuals who often find themselves on the brunt end of unfair standards, inappropriate behavior and discriminatory teammates. You may talk all day about how "my compiler doesn't care" but this is you looking down at your desk quietly as folks all over the industry go through terrible and unfair treatment. Your compiler doesn't care about that, but you should.
You suggest that changing this status quo would actually be a more subtle form of racism or sexism. This is true, insofar as the top article accidentally falls into a paternalistic role and offers open source housekeeping roles to women as a token offering, thinking this is an incentive. In terms of damaging the fortunes of white men at the expense of offering other people, the truth is that correcting for gender and racial bias in our industry will bring us closer to a meritocracy and offer more potential for talent to emerge than the current state of affairs.
I don't know what is being said here. Are you saying that women or people of color would implement software differently from a white male? How would writing an encryption app differ if a women wrote it? You seem to think that there is something being missed by the Poster you are replying to but I can't figure out what it is exactly.
Fair enough. This is a very broad question though. The answer is "yes", but this doesn't imply that any specific algorithm would be more associated with women or men. It implies that when it comes to deciding WHAT to make, decisions may end up being different.
Software, even libraries, is a lot more than selecting an algorithm and implementing it. Everything from setting a feature calendar to implementation are shaped by people's perceptions of who their "customer" (if you will excuse the product metaphor) are.
> You seem to think that there is something being missed by the Poster you are replying to but I can't figure out what it is exactly.
I think the poster is holding a pretty untenable position but I'm trying to softball it, because I could be much harsher about it.
You cannot advocate for diversity and then suggest that things that would substantially impact someone's experience wouldn't shape their thought. Unless the poster genuinely believes that racism and sexism don't exist in our society or industry, that kind of experience MUST shape people's experiences. And if they do believe that it doesn't, it's a fantasy world they're living in.
But of course I'd rather not pick a fight this way. It's not terribly productive on Hacker News, I've learned, even if it's true. When it comes to matters of race and gender, folks here are much less interested in outcomes and a "tell-it-like-it-is" mindset than with, say, how to architect a small distributed system for growth.
>I think the poster is holding a pretty untenable position but I'm trying to softball it, because I could be much harsher about it.
I don't understand how that position is untenable. It seems to be a meritocratic line of thought, IMO. There could be a problem with how "merit" is calculated? Or you believe that ideas, contributions which have merit are denied / ignored because of the gender/ethnicity of the contributor?
It can be, but is not a universal solution.
> It seems to be a meritocratic line of thought, IMO. There could be a problem with how "merit" is calculated?
A strictly meritocratic world is probably not one people actually want, but even accepting that... do we even have a well-structured and universally agreed-upon definition of merit? I submit we don't, for software.
But that's the whole issue. It's about backgrounds, not traits. You can't tell someone's background from their skin color, it's just a proxy. And if you optimize for the proxy instead of the thing itself you end up selecting black men from affluent families in the suburbs, which minimizes the difference in perspective they have compared to white men from affluent families in the suburbs.
You would end up with more actual diversity if you had chosen someone from Russia or Japan or Germany (or even a different part of the US) who has an even more different perspective. Even if that person is also a white or Asian male. Instead of selecting for the people who add the least actual diversity to your organization but allow you to check a box.
I think this isn't strictly true, but I want to be clear that taken as you meant this (charitably) then it's true! However, a negative version of this bias provably and observably exists in our industry.
So long as that exist, we must remember that this is the case and apply a counterbalancing principle. As it stands right now, people are unfairly denied opportunity by bad actors and then potentially good actors (let's charitably say you are one) often find themselves saying, "Well isn't this unfair?"
You're still balancing the wrong thing. The people underrepresented in this industry are from particular socioeconomic and cultural backgrounds. The people with those backgrounds are not racially balanced, but that does not make balancing on race a solution.
It's not necessarily the case that that there even is a solution. If a particular culture discourages its members from becoming computer scientists then the members of that culture will be underrepresented in computer science. You could make efforts to change their culture, but wasn't the different perspective supposed to be the original point?
If you actually balance on socioeconomic and cultural backgrounds, you will tend to regress towards the base population distribution of your hiring locality. But that's not what we see. Indeed, by selecting for white men in America you actually will amplify classist effects in America, as wealth is disproportionately distributed to white men.
Ultimately this is the observation it feels to me like you're hoping to avoid. If we're genuinely seeking a diversity of perspective we'd expect to see ethinic, class and religious diversity as a natural outcome of this desire. That we don't suggests that in fact people are interested in the OPPOSITE a diversity of opinion.
This particular bellwether is important because it signals a concrete group of people who are excluded just because of how they look and the preconceptions that surround that. This is exactly what you say you don't want anyone to do, but exactly what you refuse to concede might be happening given the stats.
> If a particular culture discourages its members from becoming computer scientists then the members of that culture will be underrepresented in computer science.
What even is this trying to say? What specific group is both discouraging its members to be computer scientists and also publicly demanding more opportunities in computer science?
If you actually balance on socioeconomic and cultural backgrounds, you will tend to regress toward the base population distribution of the qualified applicants in your hiring locality.
Which is basically what we see. It's not as if there are a slew of black programmers that no one will hire because of racism. The outcome is primarily a result of the lack of qualified minority applicants.
> Indeed, by selecting for white men in America you actually will amplify classist effects in America, as wealth is disproportionately distributed to white men.
The causation goes the other way. In order to get into this field you have to come from a culture that encourages young people to learn math and computers, and a family that can afford to send their children to college. Those conditions create much of the racial disparity in qualified applicants.
The economic problem has a plausible solution, you can subsidize higher education, the main impediment being how to finance it. But what do you propose to do about the cultural differences? If children in one culture aspire to be tech entrepreneurs and children in another culture aspire to be professional athletes, you're going to see different outcomes.
> What specific group is both discouraging its members to be computer scientists and also publicly demanding more opportunities in computer science?
Groups like "women" or "African Americans" are broad and internally diverse. You can trivially find contradictory opinions. One black man laments the small number of black men in computer science while another black man encourages his children to join the basketball team instead of the computer club.
There is no way to deny that some specialist disciplines have a paucity of non-white-male populations. But this too is a manufactured situation in many fields. Since schools don't directly train computer science skills, they're acquired in the workplace. Workplaces have a higher pass rate on non-white men than white men.
We've not no problem investing in young white men. Women of all colors? Not so much.
> It's not as if there are a slew of black programmers that no one will hire because of racism.
Actually there are a lot of black programmers who have a hard time finding work (or keeping it) because of workplace discrimination. It's not difficult to go to google and find a non-trivial number of cases. I encourage you do so, since I doubt you'd trust any selection I offer.
You cannot appeal to the pipeline problem to evade the problem of workplace harassment and discrimination. Even if we normalize for the skewed workplace demographics, white men make more.
> The causation goes the other way. In order to get into this field you have to come from a culture that encourages young people to learn math and computers,
It didn't take terribly long for you to start insinuating in a post that african american culture hates scholarship, did it?
> and a family that can afford to send their children to college. Those conditions create much of the racial disparity in qualified applicants.
I'm not sure where college enters into it, but yes, education is dependent on a stable home life.
> Groups like "women" or "African Americans" are broad and internally diverse. You can trivially find contradictory opinions.
I went and looked around. I found a few specific cases where religious objections were raised but they generally don't then demand more access to the field. Seems contradictory and non-trivial.
When we boil people down to their demographic membership, we do exactly that. We decide that a latino son of a doctor from Cleveland has more in common with an Nicaraguan in on an H1B than with a Caucasian son of a doctor from Cleveland.
> You cannot rationally suggest both that diversity is important but that people's traits and backgrounds aren't.
First, in specific cases, you can argue that it doesn't matter. A core dump is a core dump.
Second, you can argue that boiling people down to a handful of traits and background is saying the rest of the considerations are mostly unimportant.
What if you have a candidate whose parents were Amish? That's an incredibly unique candidate, so he should be off the diversity charts, right? But the current model we apply to determine diversity lumps him in with all the white boys who go to tech magnet schools in the Bay Area.
I like this line of thinking because it reveals the absurdity of the current demographics. Given that the US professional landscape is radically skewed away from women and people of color in terms of headcount, salary and access to capital, how do you justify this?
Is it genuinely your belief that there is an predisposition towards success in this field for white straight men, who compose its overwhelming majority?
Were we to NOT do this and actually grade people on their capacity and potential in education, job selection and social standing then I suspect we'd see a much less deviated distribution.
But we don't. We see something that more closely resembles the hypothesis that people are doing exactly what you're describing, to the benefit of white folks.
People transform criticisms of diversity statistics into a criticism of competency standards on hiring and promotion, which is exactly no one wants.
The diversity movement do not care about class. A poor white man is hated as much as a rich white man. I can only conclude that "personal experience" and "fears and the options presented" are not factors.
That's absolutely not the case, it's just that classism is often used as a blind for other forms of discrimination. Classism, Racism and Sexism are all separate axis of potential discrimination. Many intersectional feminists talk regularly about class, but we ALSO make time to distinctly address other forms of systemic discrimination.
I believe so firmly that classism is a major gate to our field that I reshaped my entire career around addressing these issues. So don't tell me I don't care about that or that I'm unfairly excluding that from my rhetoric.
If we look back a few decades, white working class men was a core voter group in left politics in both the US and the EU. Where I live (Sweden), socialist values like equality and an end to classism where very popular in the young male demographic. The social party held a close 50% majority for almost 60 years, only to drop by almost 20% in the last two decades, and the biggest change in demographic is young working class white men that has left. Every data point regarding trust that I have seen is showing that they have lost faith that the socialist movement is advocating for them.
Classism, Racism and Sexism might be all separate axis, but one can't be a advocate for a demographic and at the same time demonize them.
It holds hostage the notion of socialism in exchange for further protecting and excusing bad behavior by a privileged class.
Plenty of white people are not demonized. There is essentially no consequence for bad behavior but dirty looks. And if white men want to punish the upper white class for neglecting them, maybe they should stop pushing rich non-working-class men forward as their proxies.
> And if white men want to punish the upper white class for neglecting them, maybe they should stop pushing rich non-working-class men forward as their proxies.
An eye for an eye only leads to a blind world. How common isn't it to hear that if immigrants want help, maybe they should stop committing crime? The answer by the state and by research is the same: that the negative behavior cause by low social economical group is caused by the low social economical status. The solution is not an eye for an eye (more prison, harsher punishment, deportation, blaming them for their own situation), but rather helping them out of the low social economical status they are in.
I'm pointing out that specific minorities are being unfairly treated and your response is to infer that, not being part of the minority, the folks inflicting this must be part of the majority and then say: "demonization!"
I'm not using the word demon here. I'm saying what's obviously happening. What 20 people just got fired at Uber for doing, and more have been told they can take training or leave over. I'm not sure why recognizing there is a problem for certain people is "demonization" but it seems very fair we cannot call abusers abusers because it will hurt their feelings, but care not at all for the rights of the victims to seek redress.
> Young white men deserve protected status in the same way that young black men deserve protected status.
... What?
> The commonality here is the social economical status
Then why are there more black men in jail then white men even normalizing for economic status and type of crime? You keep suggesting that if we just listened to the stats this would be clear, but observing the stats a picture totally discordant with your worldview emerges.
> The solution is not an eye for an eye (more prison, harsher punishment, deportation, blaming them for their own situation), but rather helping them out of the low social economical status they are in.
I'm happy to help, and I do. But also: folks need to admit they're part of a system perpetrating wrongdoing and work to end that wrongdoing. Everyone has a shared duty to do this as part of our society.
I'm not sure how your response here addressed my criticism about electing super rich white men as proxies if the real problems are caused by the rich unfairly treating the poor, as you imply...
Black people are overrepresented in crime statistics in the US. Calling all black people criminals will not only hurt them, it is racism. Immigrants are also highly overrepresented in crime statistics here in Sweden, and calling them all criminals would be nationalistic and racist.
> ... What?
And this is why both EU and US political landscape has changed in the last 20 years. The inability to empathize with people that need help is in my view the biggest cause of harm in politics and reason why it becomes polarized. A person i need is a person in need, regardless of skin color, gender, sexual orientation, or what other demographic you can put them in.
> even normalizing for economic status
Normalizing for economic status removes the wast difference in crime demographic data. It is the single biggest factor, dwarfing any other attribute. If you provide data and only a single attribute then social economical status is a better predictor than anything else combined. This fact is so established that the Swedish government has taken rather extreme stance where they refuse to issue any new reports on the subject since any non-social economical attributes are just considered noise in the system. The second or third biggest attributes, regardless which ones they are, can only be political abused in order to not focus (and funding) solutions to the biggest one.
> folks need to admit they're part of a system perpetrating wrongdoing
The beauty with addressing social economical status is that you do not need to force black people (or in the case in my nation, immigrants) to admit that they belong to a group of criminals. Rather you create general moral rules which empathize with people in need. Black people are not part of some system perpetrating black people doing crime, and demanding that black people as a group need to take responsible for black people crime is just an other form of collective punishment.
> rich unfairly treating the poor, as you imply
Also called class society and is a core tenet of left politics.
> 'm not sure how your response here addressed my criticism about electing super rich white men
If we want diversity in experience and perspective then companies (and politics) should look at the rich part of your statement, not the white men part for which your comment exclusively talked about. To ignore the social economical aspect is to pretend it doesn't exist.
Said no one ever
Said lots of people just last week.
Dude got kicked out for giving talks on how women have biological predispositions against rational thought. He was a prominent developer that had a profound influence on the community.
It can not get more diverse and inclusive than that.
[1]: http://www.pocoo.org/team/#team
Is that a reflection of your hypocrisy? Or the difficulty of recruiting people that are not white men? I'm guessing it's mostly the latter.
You wrote:
> When you start an Open Source project today, in particular one which is further disconnected from frontend technologies there is a very high chance the organic community development will be everything but diverse.
and given the following, from [this comment][2] in this thread, which seems probably true:
> Major open source projects are disproportionately managed and staffed by people with full-time jobs at major software companies, and the process of obtaining and thriving in one of those jobs is not intrinsically color and gender blind, so this argument isn't persuasive.
[2]: https://news.ycombinator.com/item?id=14488000
we should be tempering our judgements of open source projects, e.g. that they're not welcoming to people that are not white men.
I'm confused as to what principle or principles you think should actually be adopted. Should all open source projects reflect the 'diversity' of the entire world?
Consider the following, from [this comment][3] also from this thread:
> LGBT people are overrepresented - almost double their percentage in the general population in fact (7% vs 4%).
[3]: https://news.ycombinator.com/item?id=14488689
Assuming that the above is true, is this a cause for concern? Would you similarly be concerned if it were true that, say, Asian people, or even just Asian men, were over-represented in technology or open source software projects? Is that not a cause for concern for you too?
It sure seems like the only cause for concern is that there are too many white men. Would anyone ever criticize an open source project for not having 'enough white people' or 'enough men', let alone 'enough white men'?
More from your post:
> What's worse is the longer you wait to try to get people involved in the project that would naturally not try to join the harder it will be. When your team is 4 men, the first woman which joins will make a significant impact. When your team is already 20 men you need to get a lot more women on board to have the same impact.
My problem with this is that you're pretty clearly, tho implicitly, devaluing contributors that don't help your project meet your diversity quotas. Your team is six white men. Have you considered replacing your existing members with women or people of color? When someone contacts you, your team members, or other contributors to your projects, do you ask them to identify their race, ethnicity, sex, or gender so you can discourage white men from contributing? If not, don't you realize that every white man that joins your team or contributes to your projects is making your diversity problem worse? You're also implicitly bashing your team members and contributors for being the wrong kind of people because you're telling them that their homogeneity is:
1. "not healthy for a project or a community to lack diversity" 2. Contributing to an "echo chamber" 3. Increasing the difficulty of future diversity 4. Hurting the project because they are relatively bad at "de-escalating arguments in bug trackers and mailing lists" 5. Hurting the project because they are relatively bad at "[taking] care of documentation" 6. They are not "people that make software work in new cultural contexts (localization, globalization, internationalization, etc.)", i.e. they are unable to understand or work with other "cultural contexts". 7. They are not "people that care about user experience"
---
I'm sure you agree with me in thinking that everyone that wants to 'participate in technology' or contribute to an open source project should be able to do so. And moreover, people that don't even realize that they would enjoy contributing to an open source project should be given that knowledge – all else being equal of course.
But that's the key constraint on how much marginal effort should be expended to recruit people that aren't already participating and contributing – all else is not equal. Everything is costly to some degree.
De-escalating arguments in bug-trackers or mailing lists – let alone even participating in arguments – requires time and energy! And there's only a finite supply of either! And opportunity costs are real and pervasive – arguing with people can be satisfying, but it can also be incredibly aggravating!
Writing documentation – and editing it, or maintaining it, or re-organizing it, etc. – requires time and energy! Someone has to do it and for most open source projects that means someone has to voluntarily do it. And this neglects the fact that 'localizing' or 'globalizing' that same documentation isn't even possible unless one knows at least two languages pretty well!
If you're going to "artificially bring balance" to your open source team, your open source project's contributors, or your conference, you're restricting the supply of possible people and thus raising the relative cost of whatever it is that you want done, whether it be writing documentation or providing user support in your issue tracker or mailing list.
I haven't personally observed any significant and unfair obstacles preventing people that are not white men from participating in open source projects, or 'technology' generally. But I'm sure they exist. Let's get rid of them. But first, let's actually identify them, and let's be careful with implying that every group of people that doesn't near-perfectly reflect the demographics of its wider community or country or whatever is guilty of overt racism, sexism, or other discrimination.
everyone agrees that such overt
exclusion, if it exists, is racist
or sexist.
you can't shut down the argument
by claiming that it's somehow racist/
sexist against white men.
It is, if you stop to think about it, a little silly to argue that the success of Asians means there's no meaningful disadvantages to being African American.
"whiteness" has changed
Yes, the meaning of "whiteness" has evolved -- for sinister political reasons: equating "whiteness" with "successful" serves no useful purpose, other than covering up inconvenient facts like immigrants/Asians doing extremely well, which doesn't fit the left's convenient narratives. Twisting the meaning of "whiteness" really is an incendiary political mobilisation strategy, and you know it. It is a well-worn strategy of political mobilisation going back -- like almost everything on the contemporary political left -- to Lenin. I recommend you reflect on this moral failure of yours: Why not use something like "successful" instead of "white" from now on, otherwise you will continue to be part of the problem. In order to solve a problem, one needs a clear understanding of what causes the problem in the first place.
meaningful disadvantages to
being African American.
Honestly, the failure to address that concern is one of the big reasons why nationalist and populist arguments are so persuasive.
Are you thinking of the term "WASP" in place of "white"? Italians and Greeks are 'W' but not 'ASP', and Irish are 'WAS' but not 'P' — but I think all three would be considered 'white' by people all around the world and I don't think that has changed.
Perhaps where you live where's a different capital 'W' version of 'White' (contrasted to 'white'), but that's not how most of the world looks at these groups. Do you think the people living Italy, Greece, and Ireland believe they are excluded from being 'white'?
For sure some of them are not white. Not that we should care much, unless we are dermatologists or something like that I guess.
That's not really how enlightened discussions work. If you have the opinion that the "compiler doesn't care" and that it's somehow related diversity you have to make an argument to the fact. You can't just say that "it doesn't care" and have it be your opponents job to make an argument that disproves your opinion. At least not if your expect to base your opinions on merit.
The notion that compiler reflect the background and culture of those creating and using it isn't exactly far fetched. Of course something like a hobbyist background (historically only available to certain groups of people) is far more important in certain types of language and that the demographics of people differ between languages.
But that not even the point. What you value in the code you're writing is as much, if not more of a factor, in projects. So it doesn't necessarily matter if the compiler is unbiased since what's good code is going to be judged by people.
isn't exactly far fetched.
Edit: While it might seem confrontational, it's a serious question. This very short discussion is a great example how it's not a meritocracy at all. You get to make whatever insulting opinion you want without backing it up and if someone even slightly dare to talk back it's downvoted. Because people agree with you opinions even though they can't really state why.
The experience of a few people, which you haven't even stated but just used their names for your own cause, doesn't really talk to that of the majority of people. Nor does it explain how the culture of those creating a language wouldn't go into it. This is exactly the problem with opinion based discussion, you never have to answer anyone. So if your believe in merit in any way just answer the original question.
How does the experience of these people make it far fetched that compilers reflect the culture people making or using them?
It's a straight forward question and you've already stated that you hold this opinion, so it shouldn't be hard to give an answer to it.
We detached this subthread from https://news.ycombinator.com/item?id=14488081 and marked it off-topic.