Strange, I thought the opposite: his decision wasn't made in a vacuum. His argument may look unreasonable only if you had no idea of the wider efforts to push similar political agendas throughout the industry and open source communities, and how meritocratic concerns are pushed aside as a result.

Discrimination of any kind is a fallacious argument, that someone doesn't need assistance simply because they belong to a particular group. It's just as fallacious as treating someone like a criminal simply because they belong to a particular group.

I admit, I go back and forth myself on the ethics of group-specific programs, rather than a more inclusive approach helping underprivileged individuals without regard to any particular group to which they may belong. I too might have no problem with Outreachy if it weren't couched in such exclusive language, and if the wider movement in the industry hadn't already been so divisive.

> IMO, it sounds _very_ much like the guy is just tired of working on the project on technical grounds, and is making the unfounded claim that those technical grounds are being caused by political actions he doesn't agree with.

Open source contribution is entirely voluntary. He could walk away for literally any reason at any time. What possible reason could he have to raise a political shitstorm that would only hinder his chance at future career opportunities if he didn't believe what he wrote?

It just sounds like you're looking for a narrative with ulterior motivations so you can easily dismiss his concerns.

I don't think I am, though it's entirely possible. I like to feel like I'm disagreeing with those concerns on their own terms, not using tricks to get out of "legitimately" disagreeing with them.

>Discrimination of any kind is a fallacious argument, that someone doesn't need assistance simply because they belong to a particular group. It's just as fallacious as treating someone like a criminal simply because they belong to a particular group.

Similarly, I disagree with this point on its own terms, I think. The position we are in right now is trying to claw our way out of a situation where we have a lot of marginalised groups who've been harmed not by saturday morning cartoon bigots, but by systemic issues. The problem with claiming to be a purely meritocratic approach is that then the bias is simply in how "merit" is determined. There is no metric which has no bias, and because of this history of systemic oppression where, say, perfectly intelligent members of some minority may be less likely to go to university for a hundred tiny factors, it's very difficult to build a metric which minimises bias and maximises results.

So a very, very rough transitionary solution is to _accept_ that your hiring practices have biases, and introduce targeted fixes for those biases. This looks like discrimination, but is actually _countering_ existing discrimination.

It's easy to look at it and go "this person is unqualified, which is why they needed a "special" path", but there's little evidence that this is actually true, and most suggestions that minorities are simply worse overall than white men at a given task are usually pretty empty.

I mostly agree with you until this paragraph. I can quibble about specific points, but this is the core dispute. Multiple attempts at double-blind hiring practices have more or less reproduced the status quo. So you're arguing that our best known scientific techniques for eliminating bias can't completely eliminate this specific bias. That's a big claim, but possible.

In principle, I think one's beliefs should be falsifiable, so from your side, what empirical data would convince you that this systemic bias either doesn't exist in STEM, or isn't a significant factor in hiring outcomes? I don't know if such data exists, nor will I necessarily look for it, but I think it's a question we should always ask of ourselves so I want to know if you're acting in good faith.

For myself, a double-blind trial demonstrating differences in hiring outcomes would have been pretty convincing. That said, I can still see a few possible sources of bias in such trials since I've done a lot of hiring myself. For instance, a hiring process might highly value open source contributions or publicly available projects/source code to evaluate skills. Since minorities are more economically disadvantaged on average than Caucasians or Asians, they might have to work more to pay for school and so have less time to do that kind of work. As such, they might be less appealing on paper.

But would you really call a hiring quota a "targeted fix" for this bias? What sort of unbiased or less biased metric could replace this sort of evaluation? We all know coding tests are terrible.

Finally, "transitionary solution" seems like a polite way of acknowledging that this is an unethical practice intended to move us towards a more equitable society more quickly than an ethical approach, like studying the problem to find actual targeted fixes. But:

a) It then seems disingenuous to be offended when people question the ethics of compromising one's ethics, even if it's (allegedly) a shortcut to a greater good. Not everyone is a utilitarian, and calling these people evil/racist/sexist for disagreeing with compromising one's ethics is dishonest at best.

b) This is then subject to an empirical evaluation as whether this shortcut actually works. There's considerable debate about whether past attempts at affirmative action were actually successful, so there's some reason to be skeptical.

For myself, I'm no expert on all discrimination issues, but I've read quite a bit on the gender disparities in STEM, the stats, papers and so on. I'm quite confident that hiring quotas won't fix gender disparities one bit. I don't want to get into that debate, but I've received quite a bit of pushback for arguing that politics should be subordinate to such evidence, and not vice versa.

> It's easy to look at it and go "this person is unqualified, which is why they needed a "special" path", but there's little evidence that this is actually true, and most suggestions that minorities are simply worse overall than white men at a given task are usually pretty empty.

I agree, there's probably very little average difference in ability between all human populations for STEM-like work, given sufficiently similar experience. But what can we use to objectively evaluate experience/claims of qualifications?

What does that look like? Do you know of anyone doing something like this today?

My only thought is the symphony auditions where the judges cannot see the applicant (hidden by a screen).

(Speaking as the temporarily self-appointed spokesmodel for the entire left: We'd love an alternative to affirmation action style mitigation steps. No one likes quotas, set asides. But right now it's the lesser evil.)

It's not a simple as what you describe.

Just speculating here, but it might be like applying for a scholarship: you have to justify why you need it more than anyone else. Some set of individuals clearly have the most need, and they should get the assistance. A double blind review process should eliminate implicit bias against gender, race, or other categories.

I bet 99% of those approved would be exactly the minorities all of these targeted groups are trying to support, but there's still that 1% that might just be a "cis white male" with no support and a truly heart breaking story. Such a person has little help available at the moment. I can think of a few logistical issues with this approach, but we're just speculating here.

> My only thought is the symphony auditions where the judges cannot see the applicant (hidden by a screen).

Yes, that's a great example. The point is to remove possible sources of implicit bias and focus on merit.

They've actually tried this at some companies, with double-blind interviews and such, but they ended up with largely the same outcomes as the status quo.

One possibility is that the bias in our industry isn't as significant as is typically portrayed. Another is that the biases happen earlier and lead to minorities either being less qualified or appearing less qualified, on average. But then efforts should be focused on helping individuals in those earlier stages, like training programs or financial support to improve their qualifications.

> No one likes quotas, set asides. But right now it's the lesser evil.

They might be justifiable if they were actually effective, but it's literally impossible for them all to be effective. Taking gender hiring quotas as an example, only about 20% of CS grads are women, and this ratio is about the standard for CS employment across the industry. But every tech company now wants a 50/50 gender ratio. That's quite literally impossible with the current labour pool, and pushing this agenda leads to skewed incentives, fast turnover and unqualified hires.

I'm not sure denying facts and scientific evidence is the lesser evil, because denying evidence simply won't fix anything. Evidence-driven efforts are great, but most I've read about don't fall in this category. Perhaps I only hear about the failures though, so that could be confirmation bias at work. Still, I would think denouncing any efforts that deny evidence would be uncontroversial, but that doesn't seem to be the case.

Except it's the organization that has the need in this sense, not the applicant. The organization has an open position, and a disparity they are trying to correct.

>That's quite literally impossible with the current labour pool, and pushing this agenda leads to skewed incentives, fast turnover and unqualified hires.

It is only impossible _globally_. If you're a company that people have heard of paying the kind of wages companies people have heard of pay, you could absolutely achieve 50/50 gender balance without compromising on quality. That kind of company is _already_ only considering the top 0.5% of on-paper candidates.

Globally, you're right, large companies doing this will starve companies less able to find and buy that talent of equal hiring practices, and _that_ sucks, but capitalism doesn't reward companies for doing things which benefit their competitors in the short term.

I already acknowledged this possibility, but pointed out that it leads to skewed incentives, fast turnover and unqualified hires.

You'd have to start hiring candidates for reasons other than qualifications, or offering pay significantly above market rate for only that gender. This strategy would also make poaching women attractive, thus driving fast turnover and further wage inflation. This demand/wage inflation would then attract plenty of far less qualified women, thus diluting the talent pool, but the skewed incentives simply can't filter out these candidates. I don't see how this would be sustainable if gender parity were viewed as desirable in and of itself, without regard for the overall economics of the business.

Practically, do you really think achieving gender parity would yield such a huge competitive advantage that these factors can be completely ignored? I think we'd have seen something like that by now, but women's pay is still roughly on par with men's in STEM, so clearly gender parity isn't economically viewed as the huge advantage you portray it to be.

The creation of a code of conduct in and of itself is offensive. If your community is bad enough that you need to remind people to not be assholes then your community already has a problem and it isn't going to be fixed by adding a CoC. It doesn't help that codes of conduct are often selectively enforced whenever it's a convenient way to expel people the "in group" doesn't like. I can't think of a time where I've seen a code of conduct actually do what it is meant to: create a better community, providing victims by providing them an escalation chain of authority to seek help from, and expelling actual assholes and people causing internal drama and a negative workplace/community. They're extremely good at creating a divisive community, driving away talented people who happen to be outside the "in group", and being used as a strictly political tool instead of a tool meant to foster a healthy community.

The inclusion of a code of conduct, to me, tells me there are political busybodies that want to seek political control of a group by citing good intentions. Because what kind of asshole would be against something telling people to not act like assholes, right? You can pass anything by painting anyone against it in a bad light. Many overreaching bills pass because no politician in their right mind would be against an act like "Fight Online Sex Trafficking Act" (FOSTA) or even a less polarizing title like "Uniting and Strengthening America by Providing Appropriate Tools Required to Intercept and Obstruct Terrorism Act of 2001" (PATRIOT).

I don't think I've taken part in any community without a set of rules in all my years. Even 4chan has a "code of conduct". The creation of a code of conduct has not been considered offensive right up until a few years ago, where suddenly certain groups of people started objecting to communities having rules even as simple as "don't be an asshole".

>The creation of a code of conduct has not been considered offensive right up until a few years ago

I respectfully disagree. The number of occurrences where a code of conduct is implemented only to swiftly be abused for political reasons has certainly increased in the last few years, which may have something to do with the increase of people sharing my viewpoint. But I've held this viewpoint for at least the past 15 years, so it certainly isn't a new position to hold.