3 ms·

We should all heavily invest in the pipeline problem through outreach and what not, but I'm genuinely curious: what happens if the underrepresented minorities d

We should all heavily invest in the pipeline problem through outreach and what not, but I'm genuinely curious: what happens if the underrepresented minorities don't make it past the blind skill validation?

tomrod 8y ago
That's a great question, and one I don't have a great and proper solution for. My splitting suggestion above I consider better than blind-only, but it certainly is not bullet-proof.
- fizwhiz 8y ago
  I ask because Github attempted this while organizing ElectronConf a year ago, but it turned out their blind submission review process[1] resulted in an all-male panel. Someone noticed and got upset[2], after which Github canned it. There was an HN discussion about it too[3]. Now we can speculate how many submissions were from people that don't identify as male (was it 50 or less%? Github hasn't released details to my knowledge). The issue though is that once the results (of what appears to be a fair process) didn't match the desired narrative, people screamed sexism and made mostly emotional arguments. You've mentioned an "inter-generational correction schema" elsewhere in the thread. In most institutions, this essentially gets instated as quotas which embitters the majority in the system (including the people they target to enable because it's assumed they had to get past a "lower bar"). [1] https://archive.is/8SsNp#selection-381.122-494.0 https://archive.is/8SsNp#selection-381.122-494.0 [2] http://archive.is/NrLqE http://archive.is/NrLqE [3] https://news.ycombinator.com/item?id=14480868 https://news.ycombinator.com/item?id=14480868
  - tomrod 8y ago
    Indeed, and I agree. I've tried to be careful from assuming specific norms, trying to think through the positive/observational aspects instead. Applying the selection quota to the hiring pool that passes skills would get around the "quota-implies-less-skilled" issue.
Baeocystin 8y ago
At what point do we decide that a difference in culture may make some careers more or less attractive to a group on average than a different population, and that that is OK? What if it's a cultural difference that has negative effect on the population in question? Do you respect the difference, or try to change things so that those in said population who would prosper under different outlooks can do so? This is a serious question, and I genuinely don't have a good, general answer. I can see this line of thought being used to superficially validate any bias one cares to in a "that's just how things are" kind of manner, which helps no one. But we also have to acknowledge that differences will lead to, well, differences. How do we differentiate between negative and neutral causality?
- fizwhiz 8y ago
  I think it's perfectly permissible for different cultures to value different careers; I believe the argument is that there shouldn't be any other imposed barriers (ex: due to discrimination) for them to re-calibrate their value system and have the option to pursue other things.

tomrod 8y ago

That's a great question, and one I don't have a great and proper solution for. My splitting suggestion above I consider better than blind-only, but it certainly is not bullet-proof.

fizwhiz 8y ago
I ask because Github attempted this while organizing ElectronConf a year ago, but it turned out their blind submission review process[1] resulted in an all-male panel. Someone noticed and got upset[2], after which Github canned it. There was an HN discussion about it too[3]. Now we can speculate how many submissions were from people that don't identify as male (was it 50 or less%? Github hasn't released details to my knowledge). The issue though is that once the results (of what appears to be a fair process) didn't match the desired narrative, people screamed sexism and made mostly emotional arguments. You've mentioned an "inter-generational correction schema" elsewhere in the thread. In most institutions, this essentially gets instated as quotas which embitters the majority in the system (including the people they target to enable because it's assumed they had to get past a "lower bar"). [1] https://archive.is/8SsNp#selection-381.122-494.0 https://archive.is/8SsNp#selection-381.122-494.0 [2] http://archive.is/NrLqE http://archive.is/NrLqE [3] https://news.ycombinator.com/item?id=14480868 https://news.ycombinator.com/item?id=14480868
- tomrod 8y ago
  Indeed, and I agree. I've tried to be careful from assuming specific norms, trying to think through the positive/observational aspects instead. Applying the selection quota to the hiring pool that passes skills would get around the "quota-implies-less-skilled" issue.

Baeocystin 8y ago

At what point do we decide that a difference in culture may make some careers more or less attractive to a group on average than a different population, and that that is OK? What if it's a cultural difference that has negative effect on the population in question? Do you respect the difference, or try to change things so that those in said population who would prosper under different outlooks can do so? This is a serious question, and I genuinely don't have a good, general answer. I can see this line of thought being used to superficially validate any bias one cares to in a "that's just how things are" kind of manner, which helps no one. But we also have to acknowledge that differences will lead to, well, differences. How do we differentiate between negative and neutral causality?

fizwhiz 8y ago
I think it's perfectly permissible for different cultures to value different careers; I believe the argument is that there shouldn't be any other imposed barriers (ex: due to discrimination) for them to re-calibrate their value system and have the option to pursue other things.