All turing complete languages make anything possible, right? The only question there is whether or not you detour through Greenspun's Tenth Rule to get where you're going.

I think Steve's point about Clojure having a limited time to gain acceptance before it's considered "over" is valid, even if he doesn't necessarily show us a way out.

> Steve's point about Clojure having a limited time to gain acceptance before it's considered "over" is valid

Is there an hourglass somewhere?

In any case, nothing's even close to "over" or "dead at the starting gate". If that were the case, you wouldn't have messages like this: http://groups.google.com/group/clojure/msg/661747c952310b41

Programming language adoption happens one developer at a time. If the silly TIOBE index is one's preferred benchmark, it's worth noting that every single language in the top 50 has been around for decades (with the exception of Go and F#, which have generous and powerful corporate sponsors).

    All turing complete languages make 
    anything possible, right?

But there must exist a point of diminishing returns. There is a difference between Turing Complete and Gosling Complete.

"Gosling Complete" is such an awesome phrase.

Agreed.

If your language is, ipso facto, saying Yes, then the core language can be left alone and features can be built by those who need them.

Sort of like assembly. :-)