unemployment rate for computer programmers over the age of 50 has been reliably estimated at an astonishing 17 percent, and as of 1997, 6.1 percent of new computer science Ph.D.s could not find stable, full-time employment

Now, with tightening of the H1 regime, those ppl still failed to find employment, just all those jobs moved to China and India thus depriving local US economies of money H1 immigrants were spending and of taxes.

Why can't European firms invest a little in turning unemployed European youth into qualified tech workers? Or make greater efforts to retain and upgrade the skills of older tech workers? Maybe they wouldn't be able to exploit cheap labor that way?

Because it's not cost-effective? And while it takes some money, organisational efforts and skills to move those jobs to China and India, it's still cheaper to move jobs there instead of re-training local personel?

In the long run, I'd put my money on a labor surplus, not a labor shortage, even in high tech

Exactly. So the policy question for H1-like programs is if you are willing to take mid-term money from fresh relatively high-payed immigrants for the benefit of local economies or are you going to forgo it outright for the benefit of China and India.

Oh, and for

If and when Russia gets its economic act together, labor market pressures on educated and skilled workers will grow.

I'm not sure what that exactly means. Russia exported millions of specialists in the last years and at the same time it's not suitable for the outsourcing due to the fact that a) good English is not common for the specialists and b) Russian specialists demand higher wages and lower work hours compared to China and India. "Getting its act together" will take a major economic meltdown.