I always read complaints about the "disposable tech worker" but never the "disposable tech company". There's almost no company loyalty these days. Which is fine, since obviously there's not a lot of loyalty to employees either. That's the world we live in. But it cuts both ways. My company might lay me off rather than retrain me. Okay. But I might leave my company for another job if it happens to involve some cool new technology I want to learn. Or if they have beer in the break room. Or if they pay me a couple thousand more a year. Or if my manager looks at me funny one day. And, in doing so, I could totally leave my employer in the lurch in a way they, to be honest, can't do to me. If a tech worker has marketable skills (which is not true of every tech worker) then he's really in the driver's seat. Laid off? No problem; he can get another job inside two weeks. If he's an integral part of his current employer's team, though, then the potential for him to damage their bottom line by leaving suddenly is much bigger.

Expense is a reason.

When I said "predict" I didn't mean there's actually someone crunching numbers somewhere and coming up with the line. I know how it's set. Nevertheless, Vegas odds can be used as a predictor. They "predict". Ignoring the fact that bracketology is concerned only with wins and losses, nor margins, if this guy were able to predict margins significantly better than "the crowds" (i.e. Vegas odds) then he'd have a license to print money and would likely want to keep it secret.

That's the hard part: defining precisely what is meant by "shortage". If there are more candidates calling themselves engineers than there are jobs does that mean there's not a shortage? If so then there's probably not a shortage. If every company could immediately fill all its positions by offering exorbitant salaries does that mean there's not a shortage? If so then there's probably not a shortage. In my limited experience interviewing candidates, though, we seem to get a lot of people who aren't that impressive relative to what they expect to be paid. So maybe there's a shortage of "good" engineers?

His predictive ability must not exceed Vegas's by any significant degree or we probably wouldn't be reading about it on slashdot.

Here is data for C.S. and C.E. bachelors degree recipients in the U.S. See page 5. About 8.7% of degrees were awarded to blacks and Hispanics, which is about one out of 11. So Silicon Valley isn't far off what one would expect based purely on # of degrees awarded. A significant portion of bay area tech workers are likely immigrants to the United States and got their degrees elsewhere. This group likely contains very few blacks and Hispanics. So, if the discussion were limited to Silicon Valley workers born in the United States the the percentage of blacks and Hispanics may well line up with expectations.

There's no screening in the U.S., but I'm not sure we do so terrible a job of serving gifted children depending on where one lives. It's just hit or miss. The city and state where I grew up don't have a reputation for being "good" in terms of education, but there were selective magnet programs at the junior high and high school levels that were pretty decent. My elementary school split its classes by ability, so even at that level I was in a classroom with kids in the top ~quartile. That's more rare these days, but my son's public elementary does the same thing starting in 2nd grade.

...find ways for us to develop our own native talent without exacerbating inequality.

Good luck with that.

What's interesting is that if you talk about running or weight lifting...

Kind of apples and oranges. Physical activities are much more easily measured, and there the sex differences are apparent even among average individuals as opposed to only at the margins. Arguing that men have (on average) more upper body strength than women because of socialization is like arguing men are taller (on average) because of socialization. It's not plausible. It's entirely plausible, though, that women's math performance is impacted by socialization. My personal opinion is that socialization doesn't explain the entirety of the gap, but it's certainly not as obvious as the physiological basis for strength and speed.

This researcher found statistically significant differences between boys and girls at the high-end of mathematical ability as early as Kindergarten. So we're talking about 5-6 year-olds. Of course this is just one paper and (for all I know) it's been refuted elsewhere in the literature.

Does she hate math because she was indoctrinated or because she finds it frustrating and difficult? Or a combination of the two? Its pretty reasonable for a kid to "hate" a subject he or she finds particularly difficult.

It's a tough thing to analyze since the physiological explanation could coexist (and possibly give rise to) the socialization explanation. Here's another paper that specifically looks at "high-end" SAT/ACT scorers among students who took the test in 7th grade. While the male/female ratio among those who scores 700+ on the math section declined (significantly) over time, it has settled at an approximately 4:1 level. Similar results for the ACT. If socialization is the primary cause then its fairly extreme and its happening fairly early.

Yes, but more women take the SAT than men, and yet the ratio of perfect math scores is 2:1 in favor of the men or 2.5:1 after adjusting for the fact that more women taken the test. Something's going on there. Maybe it's entirely socialization; I'm not discounting that possibility. But it's not that women are opting out of taking the SAT. It's also worth noting that the math on the SAT is not particularly advanced. So it's not that women's scores are suffering because they opt out of taking more advanced math classes in high school. That might be the case if we were talking about AP Calculus exam results, but IIRC the SAT only goes up to basic algebra and geometry, classes every student (usually) has to take in order to graduate.

There's isn't ample evidence to the contrary with respect to I.D. So, yeah. If you consider I.D. to be unscientific, e.g. untestable and/or unfalsifiable, then what sort of "evidence to the contrary" could possibly be given? Neither evidence that supports an old earth nor evidence that supports of the evolution of species necessarily contradicts I.D.

When it comes to women in math, most people have probably come into contact with a woman who was more mathematically gifted than they are. Or, at least, who was competent. In high school, at university, on the job, etc. So they have personal, experiential knowledge of the fact that some women, at least, can "do math". Even the most chauvinistic folks I've met don't argue women can't do math. They may argue men are, on average, better at math, or that women shouldn't do math (even when they have the ability to do so), but they rarely go so far as to make a blanket claim that no woman, ever, anywhere, can "do math". Probably because that claim is so trivially disproved.

Think women can't do math?

Hardly anyone thinks this because there is ample evidence to the contrary. Moreover, the average woman is probably about as good at math as the average man. But when you're hiring in a "mathematical profession" you're not looking at the entire population; you're looking at the set of men and women with relatively high mathematical ability. Within that set, at least in the United States, men outnumber women. This could very well be the result of socialization; I'm not necessarily arguing from physiology. But it's hard to argue with numbers. The ratio of men to women among the set of SAT takers with a perfect math score, after adjusting for the fact that more women than men take the SAT, is 2.5 to 1. So, all else being equal we should expect about 28% of engineers and mathematicians to be women. Interestingly, if you look at the percentage of Math Ph.D.s granted to U.S. citizens (in 2010) women earned exactly 28%. With respect to engineering and computer science, approximately 20% of bachelors degrees (in 2008) were granted to women, so there may be work to be done there. My guess is that this is due to the stereotypical reputations of CS/Engineering (bearded hackers with poor hygiene and huge egos) being less appealing to women than to men.