One of the things I really like about Concur.next, and Tim Bray in general, is the humility with which he approaches things. He's totally open to the idea that maybe he's going down the wrong road, and attempts to back up ideas with quantifiable results.
Too often, programming language discussions devolve into pointless argument about aesthetics and edge cases. Concur.next seems to cut away a lot of the wankery to try and see the true strengths and weaknesses of these tools.
For an excellent example, check out the "Wide Finder" series. It's eye-opening to see how these languages perform under real workhorse conditions.
I just spent the weekend messing around with Clojure. The docs aren't great (mainly due to lack of examples, IMHO) and it's been 10 years since I seriously programmed in any Lisp so it's been a bit slow going.
But I am in love with the concurrency model. Clojure's software transactional memory is amazing. It's like garbage collection for concurrency. Instead of having to remember to synchronize everywhere concurrent access could occur, you have to start a transaction any time you want to manipulate shared state.
Too often, programming language discussions devolve into pointless argument about aesthetics and edge cases. Concur.next seems to cut away a lot of the wankery to try and see the true strengths and weaknesses of these tools.
For an excellent example, check out the "Wide Finder" series. It's eye-opening to see how these languages perform under real workhorse conditions.