7 ms·
FWIW, the startup I was at seemed to have a soft affirmative action for females. They would reject males that didn't do much outside of college work, but would
by crassus 13y ago
FWIW, the startup I was at seemed to have a soft affirmative action for females. They would reject males that didn't do much outside of college work, but would be willing to consider females with the same resume because "it would be nice to have more female coders".
Culture fit was about a modicum of friendliness. You could get rejected for culture fit by, say, looking away from your interviewer and mumbling to yourself. There could have been some great (and disadvantaged!) candidates with Eyeball Avoidance Syndrome + Tourette's that were rejected unfairly by the process, but so it goes.
The company was still 90% men in engineering, because that's what your talent pipeline looks like. I'm sure they were secret sexists, or something. Valleywag would write a scathing article about them.
In your example, I bet IBM was still predominantly male despite the anecdotal hiring of a few women that you mention.
- leahculver 13y agoNope. IBM was >30% women engineers when I interned there in 2004/2005. About half the interns were women and there was plenty of racial diversity as well. My boss was a woman, my mentor was a woman, and frankly it was the most diverse place I have ever worked.
- crassus 13y agoGoogle is reportedly over 30% women, too. I wonder if it is just something about startups - higher risk, lower guaranteed reward, worse training programs, higher pressure. The startup world compares well to the original California gold rush - a risk for a small shot at a high reward - and that was mostly male, too. Also, keep in mind that it's a zero-sum game. If Google and IBM each hire 1,000 female engineers from top schools, that means that startups get less diverse.
- x0x0 13y agogoogle may be 30% women overall but I would be shocked if that where true of engineering; Rachel of rachel by the bay definitely didn't have that experience
- jfarmer 13y ago> I wonder if it is just something about startups - higher risk, lower guaranteed reward, worse training programs, higher pressure. This is explained equally well by inexperienced managers hiring people similar to themselves culturally, racially, and socioeconomically. Likewise, if I were a woman, I wouldn't want to work on a team with five dudes oblivious to their dude-dom.
- hacker789 13y agoAcross all cultures (perhaps across all of history), women are more risk averse. In fact, feminist activists didn't push to get more women into tech until tech was a higher paying, lower risk endeavor. Only then did it become scandalous that young women were choosing different career paths (but interestingly, the fact that young women are earning more than young men and graduating from college in far greater rates than young men isn't scandalous).
- jmagoon 13y agoDo you have any evidence to back up your broad-brushed claims? And who are these "feminist activist" strawmen(straw-women?) you have set up?
- newnewnew 13y ago5 minutes of googling: "MEN, WOMEN AND RISK AVERSION:EXPERIMENTAL EVIDENCE" - http://aysps.gsu.edu/isp/files/ISP_Ind_3.pdf http://aysps.gsu.edu/isp/files/ISP_Ind_3.pdf "Sex Differences in Everyday Risk-Taking Behavior in Humans" - http://www.epjournal.net/wp-content/uploads/ep062942.pdf http://www.epjournal.net/wp-content/uploads/ep062942.pdf Reality fits egalitarian beliefs quite poorly.
- jfarmer 13y agoYou've obviously made a serious study of this and that's why you have such justifiably strong opinions on the matter. Can you do 10 minutes of googling next time and share more? Or maybe 20?! Think of how informed you'd be after 20 minutes!
- judk 13y ago30% is probably including sales org, it is high for engineering. Anyway, consider the difference between 2-3 women of 12 people in an startup, vs 200 of 1000 in a big company, where women may cluster on some teams, or have "dotted line" relationships outside their official hierarchy.
- frodopwns 13y agoI think he was referring to a different time frame (1960s-1980s). But he didn't provide evidence so your objection may still be valid.
- PakG1 13y agoI have to say that I dislike it when someone says Bigcorp was like this, that is what Bigcorp's culture is. Having worked in several bigcorps, they are huge, diverse, and it's impossible for one employee to say that their experience is reflective of the entire company. There are differences from department to department, from team to team, and for the multinational bigcorps, from continent to continent. A warehouse worker for Amazon cannot say that they had the same experience as a merchandising agent, and we can't even start to compare those experiences with that of an AWS devops engineer. When someone says, "My experience at Bigcorp was like this!", we should always tone it down to read it as "My experience on the whatsit team in the whazzat department at Bigcorp was like this!" Sorry, just had to get that out there.
- icambron 13y agoThat was in response to: > In your example, I bet IBM was still predominantly male despite the anecdotal hiring of a few women that you mention. Responding to someone's completely unsupported generalization with an anecdote to the contrary is still a step up in the conversation. If you're genuinely interested in improving the rigor of the conversation, perhaps you should reply to the grandparent with, "how do you know?"
- joe_the_user 13y agoThe thing about startups and bias is that if you have a multitiered informal hiring process, a little bit of biases can be invisible and can be multiplied over filtering stages. The company may at one stage indeed "let's consider her" but unless the company's overall hiring is formally specified, unless your documenting and so keeping reasonably track of what you are doing, you may not know the way that biases are creeping your decisions. Another thing about startups is they often involve a "join our company and our lifestyle" approach. That also means that you will windup with people like you.
- crassus 13y ago"Prove that you're not a witch" is a very old tactic - and it's the opposite of the modern standard of guilt.
- joe_the_user 13y agoThe "mismatch" in dialog about this stuff is perhaps telling. One person says (essentially): "organization need to take systematic efforts to avoid bias in hiring". The other person say (essentially): "you can't prove we're sexist, this is a witch hunt".
- hacker789 13y agoNot quite. One person says, "We deliberately handle women with kid gloves. A under-qualified woman has a better chance than an under-qualified man." A feminist activist responds, "But you still hire more men! Therefore, it's almost a given that you're biased against women in other ways. Fix it." The original person says, "Wow. We just can't win with you people."
- joe_the_user 13y agoIf that happened, I could understand it. If you claim the thread above is summarized by that, you're hallucinating. I simply said that bias can be invisible and it useful to have formal methods for catching it. I don't see myself as a "feminist activist" by any means. I think the feminist activists who do sometimes uncritically toss around accusations of bias aren't really helping. But when I notice people reading my nuanced position and hallucinating a feminist rant, it increases my sympathies. It also doesn't speak well of people's abilities to see their biases (I wouldn't claim perfection for myself of course but the argument seems involve the claim that some people have no need for formal methods to correct their biases. I beg to differ, everyone needs that).
- dspillett 13y ago> The company was still 90% men in engineering, because that's what your talent pipeline looks like. Same here. I think we've hired all the women who have applied for Dev jobs as we expanded though not because of any positive discrimination (as far as I know, though hiring/firing is out of my remit). All two of them. I get the impression that women are put off a career in the sort of technical role we need (or, to look at it the other way around, are drawn to other things) fairly early in their education. Those that stick with it are determined so tend to end up making the effort and being damn good, but the rest end up thinking it isn't for them for more so than boys in a similar position. A lot of that is going to be cultural (the "boys toys" attitude mainly) but I see this falling away as personal technology has become much more ubiquitous in the last decade or so with everyone having/wanting flashy phones, tablets, and other techie toys. The proportion of my younger family members' female friends who are specifically/actively interested in a technical profession seems much higher than was the case when I was that sort of age about 18 years ago (though game designer and game tester are the main dreams there, I'm not sure how to tell them about the disparity between their expectations and the level of stress people are under in that part of the industry!). > I'm sure they were secret sexists. Valleywag would write a scathing article about them. I think we are pretty much not sexist, or anything-else-ist for that matter, but in a way that might be a problem for some. I don't know if this holds industry wide, but our sense of humour can be quite non-PC though it is applied fairly: all stereotypes (gender, age, country, size, religion, diet, hair, ...) are fair game for a joke and we don't tend to reign it in for any particular group or individual (though we aren't completely heartless in that regard). No harm is meant and no-one has thus far taken offence but I imagine there are some sensitive souls out there who would find it makes for an environment uncomfortable to work in (so one of these days it might cause problems if someone goes too far and someone does take significant offence, I can certainly imagine circumstances where that might happen).
- mathattack 13y agoIBM may have been, but women of that era made it to CTO and CEO. (IBM has had other issues so I am not defending)