I don't know exactly what "web tech" is, but I do know that the software has advanced to the point where a lot of people can do for themselves what it used to take experts to do before. This is happening in a LOT of computer areas. One of my own jobs is rendering graphics, and I am having a hell of a time trying to stay ahead of the curve. The software isn't $5,000 a seat any more, and even amateurs can turn out professional-looking work.

It's not that the job isn't getting done, it's that they don't have to hire it out any more.

Libertarian or not, only an idiot would say that government R & D has not been a positive influence on the economy. I might have something to say about its efficiency, but that's another story. Private R & D spending was about twice the government level in 1995, which was the last year I could find numbers for. Not surprisingly, of those amounts most government spending was on pure research, while most industrial spending was on applied research. The former does not have as high a return on a case-by-case basis, but it has the potential for real breakthroughs that can expand an economy enormously. The best private industries have pure research facilities, and a long-term outlook. 3M, where I used to work, was one. Their attitude was great--they considered a department successful if only one out of four research projects actually resulted in a useful product.

The paper I got my figures from can be found here.

It may sound odd, but there's nothing wrong with being a skilled plumber. Union guys around here make pretty good money, and there's no way their jobs are going to be sent overseas. ;)