[gclist] Get Real! Require GC (was re: quality of impl., etc.)

[Nick Barnes]

> >A legal Lisp compiler could simply not have a garbage collector.
> 
> I think this is getting silly.
> 
> Maybe the language standard doesn't require garbage collection, _but_it_
> _should_.  The fact that acceptable GC behavior is hard to define in the
> hard cases does not mean that it shouldn't be required in the paradigmatic
> and easy cases.  It doesn't mean that we shouldn't state our intent:
> "This here language requires GC."
> 
> A Lisp without a GC is simply broken, and the language spec should say
> so explicitly.  Language lawyers should take a tip from real lawyers, and
> write specs that have a clear intent, even if they have fuzzy boundary cases.

Rumour has it that some early Lisp Machines had broken GCs. Unlike
[typical programs in] some other languages, [some typical programs in]
Lisp allocates slowly enough that the absence of a GC used to be
acceptable.

Nick Barnes, speaking for himself