[Off] Interchange Language was: Re: [plt-scheme] conventional syntax (was: Scheme contradictions)

From: Erich Rast (erich at snafu.de)
Date: Tue May 9 06:33:28 EDT 2006

> I assume that you mean for the interchange language/library format to 
> actually be human readable (otherwise how is would your system differ 
> from JVM/CLR?).  So working under that assumption. . .
> Jeffrey Mark Siskind gave a presentation describing a similar system 
> back at NEPLS in October
> http://www.nepls.org/Events/16/abstracts.html#siskind-october

Thanks for the info, Siskind's proposal looks interesting, but isn't 
really what I was looking for.

Yes, it would preferably be human-readable, using XML as format with a 
presumably long lifespan. But there shouldn't be any mechanisms for 
distributed computing or source code distribution or mappings to other 
languages. Designers of other languages could choose to write a 
compiler from an interchange language layer to their own language and 
possibly back from a subset of their language to the interchange 
language, so people can contribute algorithms in their favorite 

As opposed to JVM/CLR, there would be no virtual machine or other kind 
of virtualization (like a JIT compiler). It would be a pure algorithm 
and datastructure specification language. The idea is that a language A 
that implements a layer B of the interchange language must be able to 
compile, interpret, or in any other way run any code written in B 
exactly according to the specification of B. That's it. No more, no 

  Anyway, I'm neither computer nor rocket scientist and busy with my own 
work, so it was just meant as a suggestion for someone else to pick up.

Best regards,


Posted on the users mailing list.