Girls Who Code

You can see female interest in programming change in the charts on these pages:

http://blogs.computerworld.com/it-careers/21993/women-comput...

Note that those graphs show raw numbers, not a percentage of the population -- if you adjust for the growing population, female graduation rates in computer science peak in the 1980s. As it says in the text:

"As a share of all CS bachelor's degrees granted that year, females had slipped almost 10 points, from 37% in 1984/1985 to 27% in 2003."

A family anecdote: my mom was working on her Phd in urban planning back in the 1970s and her advisor said to her "You know, in the future, many of these issues of traffic and resource allocation will be resolved through computer simulations, so you should learn to program." My mom thought that was a good idea so she took some classes and learned basic programming. She does not recall feeling like an outsider in those classes: the computer field was still new and felt wide open.

Nowadays a lot of startups talk about the need for "culture fit". This tends to limit the diversity of the gender and race and class of who is hired. For contrast, consider people like Evelyn Boyd Granville, and her acceptance at IBM.

http://en.wikipedia.org/wiki/Evelyn_Boyd_Granville

Here on Hacker News we also discussed the story that Raganwald posted, about another black woman at IBM, though that article is now offline:

http://raganwald.posterous.com/a-womans-story

If IBM applied a filter of "culture fit" then these women would not have been hired. But IBM, and many of the big corporations in the USA, followed very liberal policies that promoted diversity in the work place.

There were some startups from the 1950s and 1960s that broke new ground in terms of diversity. Ray Kroc built up a small startup called McDonalds and in a quiet way he made feminist history in his treatment of June Martino. She was initially hired as the bookkeeper, but she was later entrusted with vast responsibilities and finally, in 1965, when McDonalds went public, she was given shares in the company, exactly like any other cofounder of a startup. This was apparently the first time in history that a woman was treated as a real cofounder and given stock.

http://en.wikipedia.org/wiki/June_Martino

If you look at the numbers, it seems clear that the emergence of the tech-based startup scene, in the 1990s, changed things for women. The startups have not emphasized hiring diversity. The startups tend to emphasize culture fit, and they were doing so even before that phrase came into existence. Why this should be, I am not sure. There have been startups in the past that have emphasized diversity in hiring, so I am not clear why the current generation of startups cannot do so. But what is clear is that it is not a priority for them. The big and boring corporations of the past did a better job of creating spaces for women in tech.

Edit to add: to avoid being overly innocent, we should note how much the talk of "culture" is sometimes a smokescreen to hide power dynamics. Shanley Kane said "In Silicon Valley, and the tech industry in general, a lot of people were giving these talks about what their culture was and it was really superficial and focused on the privileged aspects of the company like free food and massages." Here on Hacker News we have already discussed the post "Google's 'free food' is not free" but it is worth remembering how much the talk about "culture" is just a negotiating tactic.

http://rachelbythebay.com/w/2012/01/21/notfree/

Back of the envelope:

There are ~24,000,000 people aged 18-24 in the US. Assume half are female: there are ~12,000,000 "girls who might code" in the US. Of these, 12.6% are African American. There are ~1,440,00 "black girls who might code" in the US.

There are ~400,000 people who live in Tulsa.

There have probably been several "Tulsa Codes!" events in the past couple years, but either way just stipulate that it could happen. A "Tulsa Codes!" event addresses just 400,000 people. For logistical and practical reasons, that event locks out an overwhelmingly huge number of potential beneficiaries.

But nobody has a problem with "Tulsa Codes!". Threads aren't full of people ranting about "third wave feminism" when they do.

They are not well represented in computer software. That's what we're talking about.

Go to a crypto conference sometime. If, like me, you hail from computer software and not from academic cryptography and mathematics, you will probably be surprised at how much better women are represented there than in software. The technology of cryptography is more demanding than computer software, but very similar in spirit.

What is it about the field of computer software development that keeps so many women out?

I suspect some of it has to do with the way we take observations about what successful developers have looked like in the past --- young white dudes --- and create fast paths for those people. People who "fit the mold" have an easier time making progress in our field. People who don't fit the mold have to contend with more interrogation, more doubt, and slower progress. It's true of other minorities (relative to software development demographics) as well, particularly older people.

It's pretty obvious what pg was saying in this particular "smoking gun" quote: that figuring out a way to get more girls interested in programming at 13 is an incredibly important and hard problem. The willingness for people to quickly swallow and parrot a 1-dimensional narrative of "Paul Graham, sexist pig" based upon a few quotes taken out of context in a highly edited, 3rd hand rage-blog by fucking Vallywag is the type of behavior I expect from political spinmeister hacks trotted out on the 24 hour news channels, not from smart people who I respect.

He was asked whether YC discriminates against women. The obvious subtext of that question is that there's a large imbalance between men & women among YC founders. He could have said "no, we don't discriminate; I don't know why the imbalance among YC founders exist or exactly how to correct it, but here are some reasons I don't think our process is what creates it". He does not say that.

Instead, he composes a small essay about the roots of gender imbalance in the industry off the top of his head, and that impromptu essay manages to reinforce a lot of young- white- male privilege. "If women were going to be good coders", he seems to a reasonable reader to be saying, "they would have found coding on their own". But that's not true: it conflates affinity and aptitude with opportunity and support. It also conflates "living and breathing technology" with "ability to masterfully execute a role"; the correlation between living the life and ability to execute is dubious at best.

It does not help that Paul Graham has said other things in the past that also reinforce privilege.

"Privilege" is a dirty word on HN (I guess it makes me some kind of "nth wave feminist" to use it), but it isn't an indictment. Nobody is saying privilege needs to be rooted out and eliminated. How could you ever do that, anyways? The point is not to brand young white dudes with the word, but simply to recognize that it exists, and be wary about attitudes that reinforce it and allow it to feed on itself. That's the concern I have with the interview The Information published.