[betterunix]

Eh...

In theory, the PhD is this fantastic credential that opens all kinds of doors. In practice it might make a small difference in your starting salary in industry. The only place where having a PhD is truly important is in a research position, which basically means academia. Even then it is not the last word -- tenure track positions might be out, but there are a number of schools who are willing to hire a good researcher whose highest degree is an MS.

Even with a PhD your prospects may not be all that much better than they are with an MS. I have seen people get a PhD with only one or two published papers, in a field where you need at least ten to even have a serious shot at a tenure track position. If you are the kind of person whose attitude is, "Screw it, I am in school to learn as much as I can and not just to publish a bunch of incremental improvements," you might not even get a shot at an academic job. Sure, someone can pad their CV as a postdoc -- but that is not all that much better than being a grad student, just moderately better pay.

[NhanH]

I wasn't talking about PhD as a credential, I was talking about how for the normal good-but-not-Feynman aspiring scientists, going for a PhD is one of the few ways you can properly learn how to do research.