Aucbvax.4908
fa.editor-p
utzoo!decvax!ucbvax!editor-people
Mon Nov 2 10:42:01 1981
one last Lisp and Editors statement
>From RYLAND@SRI-KL Mon Nov 2 10:31:05 1981
I don't want to add to the general melee here, but I think the source of
the confusion about Lisp as an implementation language is the lack of
information about what Lisp has become, e.g., as incarnated at MIT.
There is no standard LISP, other than perhaps the lambda calculus, which
is a mathematical toy (very interesting toy). What MIT has done with Lisp
is hard for people who haven't been at MIT (or other places such as CMU
or Stanford) to understand: it has become a full-fledged language with
every major language feature you could hope for (and a LOT more), not just
a cons-based interpreter.
To use your structure example (as Earl Killian gently pointed out), you'd
define a structure which is every bit as clean and powerful as a Pascal
RECORD. And, because Lisp is truly polymorphic, it's much more powerful.
(Polymorphism is, of course, a long rope to hang yourself with.) But on
the Lisp Machine (and Maclisp if you enable bounds checking), you're
giving up no safety, convenience, or efficiency by using structures.
This may sound haughty, but for the life of me, I can't understand why
someone would choose to use a language which fundamentally
distinguished between read time, evaluation time, and run time. Note
that I didn't say "compiled vs. interpreted"; there are some good
examples of languages which are compiled, but in such a way that there
are no read vs evaluation vs run times (e.g. Smalltalk). And, some
environments go so far to give you source-level debugging that it's
hard to call them compiled (e.g., the latest Mesa systems at Xerox).
For that reason, Pascal, Ada, etc., are fatally flawed. But now I'm
getting into language theology, so I'll quit.
-------
-----------------------------------------------------------------
gopher://quux.org/ conversion by John Goerzen <
[email protected]>
of
http://communication.ucsd.edu/A-News/
This Usenet Oldnews Archive
article may be copied and distributed freely, provided:
1. There is no money collected for the text(s) of the articles.
2. The following notice remains appended to each copy:
The Usenet Oldnews Archive: Compilation Copyright (C) 1981, 1996
Bruce Jones, Henry Spencer, David Wiseman.
