interesting, I like that idea.
We still need command line interpretation though...
Martti Halminen
|>siarnne wrote:
|>
|>> On the other hand, I can see why Autodesk isn't falling over themselves
|>
|>> to make visual DCL, Visual DIESEL.
|>
|>> AutoLISP is a proprietary subset of CommonLISP which is almost non-existent
|>
|>> in desktop apps.
|>
|>
|>Strictly speaking, this is incorrect: AutoLISP is far more like old
|>pre-Common Lisp implementations (originated as XLISP), the closest
|>relative would probably be MACLISP; it has several things which are not
|>compatible with CL, so it can't be regarded as a subset. It lacks most
|>of CL's powerful parts; for comparision, when AutoLisp still had a
|>printed manual, it was about 60 pages. The index for Steele's "Common
|>Lisp: the language", 2nd edition, is 60 pages, the whole book is about
|>1000 pages. The official Common Lisp standard would be about 1500 pages.
|>
|>Historically this is understandable, as way back in early 1980's when
|>AutoLisp was introduced, running a full-size Lisp at bearable speed took
|>a Symbolics Lisp Machine which cost about 100 000 $. The PC I am writing
|>this on is about 500 times faster than a Symbolics 3650, so these days
|>there is no problem using a full-size Common Lisp on commodity machines.
|>
|>The most intelligent thing AutoDesk could do regarding
|>AutoLisp/VisualLisp would be to throw it away and replace it with a
|>full-size CL from one of the commercial vendors, with a compatibility
|>package to allow old AutoLisp code to run.
|>
|>- Whether this is politically possible is another question.
James Maeding
Civil Engineer/Programmer