RE: [Axiom-developer] CCL maintenance.

[Bill Page]

On May 30, 2007 11:38 AM Stephen Wilson
> 
> Waldek Hebisch <address@hidden> writes:
> > Stephen Wilson wrote:
> > > Are any of the current maintainers working with CCL at all?
> > > Is maintaining compatability with this Lisp of importance to
> > > anyone?
> > > 
> > 
> > I checked that CCL from Nag Cdrom builds OK.  ATM I do not intend
> > to do any serious work on CCL, but I intend to check if current
> > Axiom is still compatible with CCL.

As part of the open source release NAG (Mike Dewar) claimed that
they had delivered Axiom to Tim Daly in a form that would compile
using CCL on Linux. I have never tested this and I think it would
be good to know since it at least gives us a standard benchmark
for comparison.

CCL was the Lisp that was used to implement the commercial version
of Axiom on Windows. As such it might be a good source to mine for
Windows compatibility issues. For example CCL was extended with a
lisp-callable GREP command specifically for Windows compatibility
since grep is not available on native Windows. Other things that
might be for some use in CCL relate to the "saturn" browser interface
for Windows which unfortunately was not part of the open source
release but might be important for future attempts to rivive this
kind of interface on Windows and Linux.

> > IMHO we should choose one of possible directions:
> > 
> > 1) Make sure that CCL really works for building Axiom
> > 2) Drop CCL support completly
> 
> It was in thinking about these two options exactly which promoted
> my email.

If resources permitted, I would opt for 1) but I cannot offer more
than just some support for testing such a configuration.

> 
> > Current state, that is having a lot of code to support CCL, but
> > no testing of this code gives us the combined disadvantages of
> > the possibilities above: we are spending time on CCL support code,
> > but CCL does not work out of the box.
> > 
> > From my point of view main advantage of CCL is that it is quite
> > portable.  Also, it looks that CCL can be cross-compiled with
> > basically the same effort as for native build.  There is also
> > licencing issue: IIUC CCL licence is pretty liberal when somebody
> > wants to deliver closed source program on top of CCL, but is
> > GPL incompatible.  
> 
> I am not a fan of the license myself.  I doubt Codemist Ltd would
> let it go under three-clause BSD.  Im not interested in improving
> software under a license like this.

As far as I know, Arthur Norman, the author of CCL is still a
member of this email list. *If* we were to choose to continue some
support for Axiom in CCL I think he might be interested and may
be quite flexible about license issues.

> ...

Regards,
Bill Page.