Hacker News Reader

Comment by cperciva | Hacker News Reader

Humans are herd animals. They tend to look at what other people like them are doing, and do likewise.

It's unfortunate that people look at trivia like gender and ethnicity in such circumstances; but it's a fact of human behaviour. If potential contributors look at a project and don't see anyone who "looks like them", they're unlikely to contribute; so projects which want to maximize contributions are well advised to seek ways to "look inclusive".

(And this goes beyond open source software, of course. My university's computer science program has seen a significant uptick in applications from females after going out of their way to select photos of female students for recruiting material.)

In short, it's a matter of overcoming cognitive biases.

View on HN

I view this issue differently, and I would love your opinion on my view.

First on sex/gender: I feel that hormonal differences create psychological differences in the norms between genders. This is why men are generally more aggressive and violent(see higher incarceration rates) and women are more nurturing. These differences lead to people in general choosing careers that reflect their psychology, and forced diversity hurts those that do best in their field.

Then on ethnicity, I feel many of the differences come from both differences in opportunity(which should be overcome) and differences in cultural and social norms, which lead to people choosing different careers.

In my opinion, forced diversity hurts those who would normally gravitate towards those fields because they are displaced by someone who doesn't necessarily want to be there, but just sees opportunity.

Again, I'm not looking to fight. I'd love you opinion on why I'm right or wrong.

In the context of open source contributions, no one is displaced.

In a wider context, yes, it's a bit more complicated, but Free Software? As a movement, there's power in numbers, and looking to increase that number via a focussed effort on minority groups is simply a smart move.

Open source software isn't a zero sum game. When one person joins, that does not displace someone else who would normally gravitate to that field because open source can scale efficiently to any number of contributors.

On the contrary, not having enough diversity can and does put off those who would otherwise make great contributors.

Are there differences between genders? Sure. It's an open question how much of the differences are due to direct biological causes and how much are due to societal norms (I find the parallel between open source software and indigenous gift cultures to be quite interesting, for example), but there are certainly differences.

As such, I would oppose any attempts to establish quotas or even aspirational targets for diversity; as you say, forced diversity has negative consequences -- not just for the organization, but also for the individuals involved.

But as much as we shouldn't be trying to bring in individuals who aren't really interested, we need to make sure we're not keeping out individuals who are interested... in a sense, it's more important to look diverse than it is to be diverse.

Hormonal effects has a long history of being overstated and overused when looking for cause in human behavior. It has been used to argue that women aren't fit to vote, then later that they are not fit to hold professions where a stable emotional state is required, to later conclude that estrogen effect on emotional states is complex and what people at the time attributed to estrogen was primarily cultural expectations rather than actually brain chemistry (concluded with some really interesting studies). One conclusion that a profession said that always made good sense to me is that sex hormones are intended to regulate effective reproductive success, not to regulate emotions in women or violence in men.

For example, there was a study done on violent men in prisons and testosterone. The theory that was tested was if testosterone levels were indicator of more violence, for which the data revealed was not the case. Further studies showed that so long testosterone is within natural occurring levels, aggression was not effect by testosterone.

> First on sex/gender: I feel that hormonal differences create psychological differences in the norms between genders. This is why men are generally more aggressive and violent(see higher incarceration rates) and women are more nurturing.

There's strong evidence that genetic differences account for some behavioural differences between men and women. The ones you suggest do make it clear why you'd expect a skewed distribution of genders in fields such as the army or midwifery. It's much less clear that genetic differences should lead to such an underrepresentation of women in tech fields; in this case I suspect that culture and social inertia are a large part of the explanation.

I have one big objection to this idea about potential contributors not seeing people who "look like them":

We're talking about open-source/Free software here, not working in a corporation. Volunteer FOSS work is all done remotely. If you asked me what the contributors to, say, The GIMP looked like, what sex they were, and what their ethnicities were, I could not tell you other than to merely make a wild guess. I suppose I could go look up their names and then make very good guesses based on those, but that's a lot of work; it's not like the contributor list pops up every time you start the application. FOSS volunteers frequently never see each other in person, and simply send in code contributions. They're not interviewed in-person, and they don't have to show up in-person to an office full of other people. If anything, it should be the most egalitarian form of work worldwide!

For most users, FOSS software is made by people who are utterly nameless and faceless. So I really don't see why this should be a factor to potential contributors. It should be the one place where they can feel safest about not being singled out and treated unfairly for their sex/age/ethnicity/other factor. There's not even a requirement that they use their real name! No one's stopping them from making a new email address at fake.name@gmail.com and contributing with that if they're really paranoid about being treated unfairly.

Reply on news.ycombinator.com