What about Lisp? I don't believe it's gaining new programmers at a fast enough rate to be a sustainable community in the long term.
All the kids who might've gone into Lisp are going into Python instead and asking the legitimate question, what can Lisp do that this can't? There's a lot of hand-waving from the Lisp community about this and that, but a lack of concrete example. On the academic high ground, there's Haskell and OCaml - both communities are actively working on "batteries included" distributions to bring their languages to the masses. The Lisp community took too much for granted for too long, they have become overtaken by events on both sides. It's not that Lisp will ever "die"... It'll just become irrelevant by the next generation.
A couple of years ago we started this project called Squeak,
which is simply not an attempt to give the world a free Smalltalk,
but an attempt to give the world a bootstrapping mechanism
for something much better than Smalltalk,
and when you fool around with Squeak,
please, please, think of it from that standpoint.
Think of how you can obsolete the damn thing
by using its own mechanisms for getting the next version of itself.
Ruby, if memory serves, started out as a tiny Scheme that got a whole lot of agglomerations from Perl (regexps &c) and Smalltalk (objects &c).I see Py3k as proof positive that Guido et. al. intend to create something enduring, and I'm also reasonably excited about F# and the ML family in general. Clojure seems to have reinvigorated interest in Lisps also, but with the benefit of one of the larger available libraries. Should be an interesting decade in language development.