(str "if " "Xbox 360 has all you need to run Clojure"), I was just trying to emphasize that even if you could emit code you still had other problems.
On Sep 23, 5:45 am, Timothy Baldridge <tbaldri...@gmail.com> wrote: > > Is this still a problem if you compile everything ahead of time? I > > noticed that if I make a jar with leiningen my clj files are included > > in the jar but I assumed that it was just part of some dogma that I > > should share my source code. > > Actually I think it's a bit worse than that. LISP treats all code as > data and all data as code. Basically, in LISP you can take any list > and attempt to execute it. So even if you completely AOT compile the > code you're going to have to rip out any JIT functions to get it to > properly transfer to the Xbox. > > > > > Also, I asked a similar question where I was concerned about the > > things Timothy mentioned. The one answer I got was pretty much that > > if the Xbox version of .Net had everything Clojure needed the Xbox > > just wouldn't run clojure very well. Just like with Dalvik, the > > Compact .Net handles GC poorly. > > The biggest issue is that Clojure, IronPython, etc. are based on the > Microsoft DLR. This DLR makes heavy, heavy use of on-the-fly code > generation. As far as I know the 360 does not implement a full-blown > JIT (the bytecodes are translated to machine code on the development > PC then the resulting binary is shipped to the Xbox). Therefore the > DLR will not run on the Xbox and languages that require it will > either. That link you supplied doesn't state that "Xbox 360 has all > you need to run Clojure" but simply states that in addition to any > other issues you may have, Clojure would kill the GC. > > Now what could happen is this: Someone could take the AST classes in > Clojure-CLR and convert them from code generation routines to > interpreted routines. Interpreted Clojure would run on the Xbox, there > just isn't an interpreter written yet. > > Timothy > > -- > “One of the main causes of the fall of the Roman Empire was > that–lacking zero–they had no way to indicate successful termination > of their C programs.” > (Robert Firth) -- 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
