It almost looks like they tried to present the data in the most-negative-to-fb possible way, just in terms of ordering if nothing else. I'd also be interested in citizen vs. immigrant (of various status); someone from Russia moving to Menlo Park to work for Facebook really is not directly interchangeable on a diversity metric vs. a 5th generation American of any race, IMO. Native-Bay-Area vs. moved-to-Bay-Area before vs. moved for FB might be an interesting metric, too.<p>It really only matters that they keep measuring it the same way and then watch for trends as they try to change things, though. Good job for caring and being quantitative IMO.<p>What amazes me is how low the "two or more races" category is. I would have expected a much larger percentage there.
I think it's the government's job to make sure that the candidate flow is balanced across those facets. As a private corporation, when there are N candidate for a position, the best one should get the job. Doing something else makes charts look better, but hurts corporations, the economy as general, and is totally unnatural. Enforcing diversification also assumes wrongdoing without a proof.