4 ms·

> If there are disparities, what, if anything, do you think should be done to lessen the disparity in demographics? Target inequality of opportunity, and do no

> If there are disparities, what, if anything, do you think should be done to lessen the disparity in demographics? Target inequality of opportunity, and do not assume that differences in outcome are prima facie evidence of differences in opportunity. In other words, the pipeline is the place to solve this.

grzm 10y ago
What does that look like in implementation? How do you ensure and measure its effectiveness? What rates of improvement should be acceptable?
- edblarney 10y ago
  I'd suggest that a starting point could be to make sure that those hired roughly reflect the diversity of the talent pool being hired from. If they 'match' - then it's a roughly good indicator that the hiring process 'is fair'. As far as I know - this is generally true: the ratio of female engs. in the valley is roughly commensurate with the ratio of females graduating from Comp Sci. Which is actually positive news. Now - that's nowhere near the whole story, but at least it's a crude basic metric.
  - grzm 10y ago
    I tend to agree with you on the starting point and that it's a rough indicator. that's nowhere near the whole story I'm glad you acknowledge this. It's a story that tends to get abandoned after looking at the types of metrics you describe. I don't think that's what you're doing here, by the way. Just the fact you bring it up makes it clear you don't.
    - edblarney 10y ago
      Thanks. But I often feel that that core metric is way too often overlooked by those lamenting injustice. The fact that 'the first rough metric' is pretty decent, I think speaks to the general fairness and goodwill in the tech industry. I've worked in tech my whole life - and without a doubt the people are pretty good. It can be a little guy-ish sometimes, and there are some bad-actors/companies, but I reject the idea that Tech is bad for women. Google and Facebook are probably 95% Obama voters, and progressive on almost every issue. To scream bigotry among this group is just too rich. So yes - the story does not end there, and there are clearly more obvious problems as you move up in management - but I also don't see an entire industry that's in some kind of trauma. Interesting tidbit: a survey of tech workers shows that men point primarily to the 'funnel problem' whereas women point to the 'lack of role models'. Which I thought was telling, and hints at an opportunity for improvement.
      - grzm 10y ago
        I'm a bit taken aback by your comment here. I was encouraged by the lack of partisanship that began our exchange, as that so often does nothing but further polarize discussions. There are indications that you're looking for a more nuanced discussion, but phrases like "those lamenting injustice" and "to scream bigotry among this group is just too rich" makes it harder, rather than easier, for those that don't already agree with you to be willing to enter a discussion: it takes extra effort and good faith to attempt to engage in substantive discussion on contentious issues (rather than fall into the echo chamber everyone decries). a survey of tech workers shows that men point primarily to the 'funnel problem' whereas women point to the 'lack of role models'. Interesting. Do you have a reference to this survey at hand? Which I thought was telling, and hints at an opportunity for improvement. I'm not sure if I'm particularly dense (though I don't dispute this might be the case), but I think people generally assume that their intent in statements like this are much clearer than what is actually the case. I have a hard time knowing how to interpret what you mean, or what opportunity you see. Is it that you think there should be more role models for women? That the lack of role models for women indicates that there just aren't that many women in tech and that's because women aren't generally interested in tech so there really isn't a problem? That there's a mismatch between how men and women perceive the problem that needs to be addressed? If that's the case, do you think one perception is more accurate than the other? My comment to which you initially responded asked for specifics on implementation and measurement on addressing disparities in tech. I'm still interested in hearing for details rather than generalities.
        edblarney 10y ago
        "I was encouraged by the lack of partisanship that began our exchange" There is no 'partisanship' in my comments. I'm just making a comment. That's all. "I have a hard time knowing how to interpret what you mean" I mean exactly what I say. There's nothing to interpret. There is a lot of noise in the press about 'injustice' in terms of gender and race in the Silicon Valley - when, as a baseline - the Valley employs almost exactly the people that are fed into it. When you read articles about such issues - they generally do not refer to the fact that the % make up of the Valley is almost identical to the % makeup of those with relevant qualifications. So - as a 'starting point' for any concern, it should be noted that by the easiest and most obvious measure - the Valley seems to be doing a pretty good job. Maybe even a great job. I think this is unfair: a lot of voices 'screaming bigotry' in the press lack this perspective, and instantly distort the scenario by taking the assumptive position that 'there is inherent unfairness in the valley because only x% of their Engineers are of minority group ABC' - when in fact, it's actually pretty fair, given the talent pool they draw from. I can't say that I've ever read an article critiquing the Valley on gender representation - that ever made mention of this fact. So even though there is an issue to be addressed - it's also being misrepresented. When I say 'there's more to the story' - I'm not offering a consolation or token to 'the other side' of the debate. I don't view these things as having sides. I'm simply stating that I think there's more to it. That's all. I'm not debating anyone or taking any sides. As for the 'male/female' perspective - I wish I had that reference for you - I read it relatively recently, and the data point really popped out at me, which is why I remembered it. I'm a guy - so 'lack of role models' is something I would never consider as a possibility for 'why I wouldn't want to work in tech'. So, knowing that so many girls feel that way, make me think about the issue differently. How can this be helpful? Well, for one, obviously by making sure that the 'stories' of female engineers are told, that there is visibility etc. - which is kind of a no-brainer, one that really should not be contentious. But there are other issues. The Canadian Navy did a study (long ago) where they found that when ships deployed with very few women on board, that it was stressful for them. But when there was a 'critical mass' of them (I think like 8%), then they seemed to get along a lot better. So, this 'critical mass' concept might be useful in planning: if a 'guy founded' startup grows, they may consider reaching that 'critical mass' as an absolute minimum for female participation. Most companies don't think about gender issues until headcount gets larger, so maybe once they start growing, they can try hard to get 'a good group' of girls in, beyond that threshold, so that they don't feel isolated. The same could very well apply to minority groups - i.e. maybe there is a 'critical mass' below which many feel isolated. Black and Latino Engineers are pretty rare. This could be an issue there. The 'role model' issue may be also strongly related to the lower participation of females in upper management. But that's a whole other can of worms, and I've already written too much so ...

grzm 10y ago

What does that look like in implementation? How do you ensure and measure its effectiveness? What rates of improvement should be acceptable?

edblarney 10y ago
I'd suggest that a starting point could be to make sure that those hired roughly reflect the diversity of the talent pool being hired from. If they 'match' - then it's a roughly good indicator that the hiring process 'is fair'. As far as I know - this is generally true: the ratio of female engs. in the valley is roughly commensurate with the ratio of females graduating from Comp Sci. Which is actually positive news. Now - that's nowhere near the whole story, but at least it's a crude basic metric.
- grzm 10y ago
  I tend to agree with you on the starting point and that it's a rough indicator. that's nowhere near the whole story I'm glad you acknowledge this. It's a story that tends to get abandoned after looking at the types of metrics you describe. I don't think that's what you're doing here, by the way. Just the fact you bring it up makes it clear you don't.
  - edblarney 10y ago
    Thanks. But I often feel that that core metric is way too often overlooked by those lamenting injustice. The fact that 'the first rough metric' is pretty decent, I think speaks to the general fairness and goodwill in the tech industry. I've worked in tech my whole life - and without a doubt the people are pretty good. It can be a little guy-ish sometimes, and there are some bad-actors/companies, but I reject the idea that Tech is bad for women. Google and Facebook are probably 95% Obama voters, and progressive on almost every issue. To scream bigotry among this group is just too rich. So yes - the story does not end there, and there are clearly more obvious problems as you move up in management - but I also don't see an entire industry that's in some kind of trauma. Interesting tidbit: a survey of tech workers shows that men point primarily to the 'funnel problem' whereas women point to the 'lack of role models'. Which I thought was telling, and hints at an opportunity for improvement.
    - grzm 10y ago
      I'm a bit taken aback by your comment here. I was encouraged by the lack of partisanship that began our exchange, as that so often does nothing but further polarize discussions. There are indications that you're looking for a more nuanced discussion, but phrases like "those lamenting injustice" and "to scream bigotry among this group is just too rich" makes it harder, rather than easier, for those that don't already agree with you to be willing to enter a discussion: it takes extra effort and good faith to attempt to engage in substantive discussion on contentious issues (rather than fall into the echo chamber everyone decries). a survey of tech workers shows that men point primarily to the 'funnel problem' whereas women point to the 'lack of role models'. Interesting. Do you have a reference to this survey at hand? Which I thought was telling, and hints at an opportunity for improvement. I'm not sure if I'm particularly dense (though I don't dispute this might be the case), but I think people generally assume that their intent in statements like this are much clearer than what is actually the case. I have a hard time knowing how to interpret what you mean, or what opportunity you see. Is it that you think there should be more role models for women? That the lack of role models for women indicates that there just aren't that many women in tech and that's because women aren't generally interested in tech so there really isn't a problem? That there's a mismatch between how men and women perceive the problem that needs to be addressed? If that's the case, do you think one perception is more accurate than the other? My comment to which you initially responded asked for specifics on implementation and measurement on addressing disparities in tech. I'm still interested in hearing for details rather than generalities.
      - edblarney 10y ago
        "I was encouraged by the lack of partisanship that began our exchange" There is no 'partisanship' in my comments. I'm just making a comment. That's all. "I have a hard time knowing how to interpret what you mean" I mean exactly what I say. There's nothing to interpret. There is a lot of noise in the press about 'injustice' in terms of gender and race in the Silicon Valley - when, as a baseline - the Valley employs almost exactly the people that are fed into it. When you read articles about such issues - they generally do not refer to the fact that the % make up of the Valley is almost identical to the % makeup of those with relevant qualifications. So - as a 'starting point' for any concern, it should be noted that by the easiest and most obvious measure - the Valley seems to be doing a pretty good job. Maybe even a great job. I think this is unfair: a lot of voices 'screaming bigotry' in the press lack this perspective, and instantly distort the scenario by taking the assumptive position that 'there is inherent unfairness in the valley because only x% of their Engineers are of minority group ABC' - when in fact, it's actually pretty fair, given the talent pool they draw from. I can't say that I've ever read an article critiquing the Valley on gender representation - that ever made mention of this fact. So even though there is an issue to be addressed - it's also being misrepresented. When I say 'there's more to the story' - I'm not offering a consolation or token to 'the other side' of the debate. I don't view these things as having sides. I'm simply stating that I think there's more to it. That's all. I'm not debating anyone or taking any sides. As for the 'male/female' perspective - I wish I had that reference for you - I read it relatively recently, and the data point really popped out at me, which is why I remembered it. I'm a guy - so 'lack of role models' is something I would never consider as a possibility for 'why I wouldn't want to work in tech'. So, knowing that so many girls feel that way, make me think about the issue differently. How can this be helpful? Well, for one, obviously by making sure that the 'stories' of female engineers are told, that there is visibility etc. - which is kind of a no-brainer, one that really should not be contentious. But there are other issues. The Canadian Navy did a study (long ago) where they found that when ships deployed with very few women on board, that it was stressful for them. But when there was a 'critical mass' of them (I think like 8%), then they seemed to get along a lot better. So, this 'critical mass' concept might be useful in planning: if a 'guy founded' startup grows, they may consider reaching that 'critical mass' as an absolute minimum for female participation. Most companies don't think about gender issues until headcount gets larger, so maybe once they start growing, they can try hard to get 'a good group' of girls in, beyond that threshold, so that they don't feel isolated. The same could very well apply to minority groups - i.e. maybe there is a 'critical mass' below which many feel isolated. Black and Latino Engineers are pretty rare. This could be an issue there. The 'role model' issue may be also strongly related to the lower participation of females in upper management. But that's a whole other can of worms, and I've already written too much so ...