[jholman]

So, it's not at all clear that there are, in fact, enough high-quality female programmers to have a 50/50 split in the field at large, which is clearly the important point.

But I really started this comment to pick a few minor nits. So feel free to ignore the entire comment.

1) They didn't achieve 50%, actually, even with all that work.

2) 127 * 0.22 = 28% . Assuming your industry estimates are correct, you're going to need better than 100% oversubscription to get 50% women, actually.

3) And even that only assumes random sampling. Which is actually clearly a crazy assumption. And I would take this hackathon as proof of that, actually.

4) I'm now on a team that's roughly 13% female. That's a new maximum for my career. So with that in mind, I'm curious where you got that 22% figure.

[roguecoder]

2) It's 227*0.22; you still get the original 100% too. If it were 27% oversubscribed you'd get 28%.

3) It is an assumption, though one that fits with my experience of hackathons (they tend to be 15-20% women.) I wasn't saying it's guaranteed that women will want to attend, just that there are qualified women who could.

4) Department of Labor, 2009 numbers: http://www.bls.gov/cps/wlf-databook-2009.pdf Page 29.