Oh give me a stub any day. It performs three useful functions: firstly it provides information to readers, secondly it provides a basis for further research and expansion (or merging) and thirdly it notifies editors of a piece of verifiable information that (probably) belongs somewhere on Wikipedia. If we left the job of thinking up new articles to me, we'd end up with nothing but articles about odd programming languages.
A lot of stubs fail to be verifiable and don't cite any references. But the main problem I have with stubs is the location they're put at. For example (totally random), someone could've written a stub about Alice at the Mad Hatter's tea party and name the article "Tea Party", causing immediate work for others.
1) It's miscapitalized. 2) It's already covered elsewhere. They fail to see the big picture, search if we've already got the info at the most likely place ([[Alice in Wonderland]]) 3) Anyone looking for info on the mad hatter's tea party can end up at the stub, when in fact we have loads more info elsewhere hidden by the fact the reader found the stub. 4) Wikipedia shouldn't be about article creation, it should be about structuring the incoming stuff in a logical way and place it were readers should be looking at it.
You could write a stub about DNA structure. But if someone can't find an article there, shouldn't it be reasonable to assume they search in DNA next?
--Mgm