Re: [Axiom-developer] RE: Boot vs. Lisp

[root]

> > http://www.staff.uni-mainz.de/weiss/axiom_cmu.tgz
> > 
> > So, couldn't we add CMUCL as a second compiler?
> 
> Using the CMU compiler is not a big deal. The tar archive
> you mentioned has, I think, a running (but old) version in it.
> It would not be difficult to fold the changes into the
> current version. As is, it does not support the external
> programs (hyperdoc, graphics) as I did not link the 
> C library into the lisp image. This could probably
> by done, or better, as CMU CL does support an almost
> complete interface to Unix system calls (and especially
> sockets) one could easily recode the glue to the
> external programs in Lisp.
> 
> The compiler in CMU CL gives a lot of good error messages
> and warnings. I found quite a few bugs in the
> AXIOM sources with it. And I think the system is a little faster
> then GCL on average (did not do too much benchmarking though).

the first common lisp port of axiom to the ibm/rt was to cmucl
(then called spice lisp under scott fahlman at cmu). it should
still work. i'll fetch the archive and diff it with the current
sources.

the key issue is the browser/graphics integration.

t