[linux-elitists] Opera beta for linux available
Sun Oct 8 10:58:13 PDT 2000
On Sun, 8 Oct 2000, Eugene Leitl wrote:
> > Artificial languages do not have the ability to express *anything*.
> Hogwash. Proof: you have a job.
I meant any arbitrary thing. For example, I can't express the difference
between thought and reality accurately in programming. Except perhaps with
> > In part, but in many cases it's simply the pragmatism of the approach. For
> > example, Smalltalk is not a commercial success as far as distributing
> > software goes, but it has a niche in in-house development, mostly in
> > academia and sometimes in corporations. The reason for its lack of
> > commercial succes IS almost exclusively because of the language concept.
> Smalltalk is not a bad language, I've read some reports of a company
> exclusively using it. (It may have warts, I've never used it).
Smalltalk's concept was to have an entire operating environment rolled in
as a part of the language and, when saving code, you're actually saving
the state of the environment.
Needless to say, this has mondo overhead. Some of the more recent
> > Rick and I had a talk about this over lunch. About the kinds of things
> > Prolog was good for as opposed to Lisp as opposed to your typical
> > imperative language.
> Prolog was just a language front end to an inference engine (which you
> can write in a few screenfulls of Lisp).
Prolog is a language that's really good for fuzzy logic. Lisp however is
much better at binary tree AI.
> Lisp isn't a purely functional language as you well know, you can mix
> imperative and functional stuff ad lib. The major minus of Lisp that
> it has never been enforcibly standartized, had a relatively fat memory
> footprint and initially lousy compilers and arithmetics, which hasn't
> been true for a while.
I know it's not purely functional, which is one reason why it's not used
these days to demonstrate what a functional language IS.
> The reason Lisp is not being used now widely is the same reason Forth
> is not: vogues and subcriticality.
Actually, I suspect is there is a more simple reason: most people don't
think in functional programming. Some people do and they're drawn to
languages like Lisp.
For most people, imperative languages make more sense.
As for prolog, I described it yesterday as "programming via wff 'n proof"
which made my mom shudder. (http://www.wff-n-proof.com/ -- naturally)
_Deirdre * http://www.sfknit.org * http://www.deirdre.net
"More damage has been caused by innocent program crashes than by
malicious viruses, but they don't make great stories."
-- Jean-Louis Gassee, Be Newsletter, Issue 69
More information about the linux-elitists