I have found that in the Real World TM I never get a choice of the DB I get to use. The three main reasons being:
1.) Monetary constraints. The basic thought on this one being "A database is a database Why use database X when we already own Y ?"
2.) I jump into the project and it already has a DB.
3.) A project leader asserts their authori-tae and makes the decision so that they could claim more responsibility for success.

I do think this could be an interesting discussion. I am particularly interested in people who have worked with thunderstone. I haven't had a chance to use it but I have heard mixed reviews.