On the other hand, the fact that the bytecodes are wel-documented (both text docs, as well as a Redex model) has been a huge boon to apps that can rely on all these front-ends to do their job, and then port just the back-end to other platforms.
For the reasons Jay explains, I'd be hard-pressed to imagine someone wanting to compile a reasonably standard language directly to bytecodes. Put differently, this is a "test case" of a language's design, another one like its ability to express a meta-circular interpreter. Shriram On Tue, Sep 21, 2010 at 3:11 PM, Neil Van Dyke <n...@neilvandyke.org> wrote: > Jay McCarthy wrote at 09/21/2010 03:02 PM: >> >> We already have hosted Javascript, Algol 60, Datalog, R5RS, and R6RS. In >> the past, we've had version of Java in the core. We have Python and SML out >> there from people at Utah. >> > > Do these all translate to Racket syntax, and then the normal Racket > interpreter/compiler takes over? Or does anyone generate VM bytecode > directly? > > -- > http://www.neilvandyke.org/ > _________________________________________________ > For list-related administrative tasks: > http://lists.racket-lang.org/listinfo/users > _________________________________________________ For list-related administrative tasks: http://lists.racket-lang.org/listinfo/users
