Re: [Matlisp-users] Matlisp subset in pure CL

[Raymond T. <to...@rt...>]

>>>>> "simsek" == simsek  <si...@ee...> writes:

    simsek> But ... there is one thing you may not be aware of.  IIRC, someone
    simsek> correct me if I'm wrong, one of the main reasons why we did matlisp
    simsek> the way it is was to avoid writing such routines in lisp altogether.
    simsek> When I first contacted Ray, he had already worked out the generic
    simsek> foreign wrapper suitable for fortran code.  I then wrote a script
    simsek> to generate the wrapper code automagically from the lapack files.
    simsek> So the idea was not to write basic matrix operatinos at all.


Yes, that was certainly my idea too.  Having said that, let me also
say that I experimented a bit with a full Lisp implementation of some
of the basic routines, and at least with CMUCL, Lisp was no slower
than LAPACK for small matrices.   But why redo the work when someone
had already done it, and better than I would do?

    simsek> Having said that, matlisp has evolved along the way, with contributions
    simsek> from select individuals like yourself.  I think it may be time to
    simsek> consider more drastic functionality of the kind you're suggesting.

I also think this is a good idea.  It would be nice if this approach
could be made seamless with Matlisp, so the user wouldn't have to know
unless he wanted to.  If it also means changing some Matlisp
functions, that would also be ok.  We might want to ask other users
about that, though. :-)

We probably also want to do an official 2.0 release before adding such
features for the "3.0" release.

    simsek> I, unfortunately, personally would not be able to participate in the development
    simsek> but I think there are people out there who might be.

I can help some, but I don't use matlisp that much anymore.

Ray