I thought about that a little too. The thing is that clojure is doing something almost imposible. There is almost know way around writing you json library backand in clr interop.
If we want to make clojure programm that run on the clr, the jvm and possibly in the browser we have to write pure clojure but not using native stuff is against the clojure way. So how can we solve this? I think we need a kind of tagsystem. With the tags you could indicate if a library is pure clojure or if it uses lower level stuff. For librarys that use jvm we have to define a interface with the clr version can mirror. That way we would at least not need to change code that runes on these librarys. That would need lots of communication between the clr and jvm people. But there will always be code that only runes on one host. You cant (trivially) call hadoop on the clr for example. The other possibility is to develop to diffrent ecosystems but that would make it even harder to switch from one to the other. In the long run the librarys that are writen in pure clojure will hopfully become the standard. --------------------------------------------- The same problem we have with the typesystem. No idea how to solve that. -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en
