There's a lot left to do. Huge upcoming projects:

  * Pods
  * Clojure-in-Clojure compiler
  * High performance higher order operations 
    on persistent data structures filled with primitives

Projects that have been talked about but no-one's helmed:

  * Predicate dispatch
  * Datalog (fast enough for the above)

Community projects in development:

  * design for asynchronous code (think Microsoft's Rx Extension)
  * network REPL
  * unification library
  * exception/conditions design

I could go on, but you get the idea.

I still want nREPL support for emacs, but I haven't had a chance to look at actually doing it...

FWIW, there's lots of stuff to be done in nREPL proper as well. Anyone interested should definitely ping me. I'll hopefully get the outstanding issues into the new JIRA in the next couple of weeks, along with a migration of the various design notes that are scattered about.

Also: * integrating fork-join!

Also:

  * Scopes
  * Better type inferencing
  * letmacro
  * The things in Rich's mind that are baking

- CLR, android

- common parse/highlight backend for emacs, vim, eclipse etc

http://dev.clojure.org/display/design/Home

Especially for a new language where it takes time to build a base - time that would be well spent bringing the toolset, libraries, and frameworks up to level.