You apparently didn't read what I wrote very carefully.
On Aug 13, 2005, at 7:17 AM, Jonathan Ellis wrote:
Wow. That's totally the opposite of what I've seen. Lisp guys
*loved*
their Lisp Machines. It's like the Amiga community. "Back in the day
we totally kicked everyone's butt, man."
Yeah, what I said was that Lisp Machines were EXPENSIVE, so they
weren't available to hackers. The cheap unix workstations took
over. ITS at the MIT AI Lab got replaced with a commercial OS that
wasn't hackable. That killed the Lisp hacker culture there.
If you re-read what I said, I also noted the grumbling that took
place when Lisp hackers were forced to use Unix. They most certainly
loved their Lisp Machines, and having seen a screen dump video of the
development process on them, I can see clearly why. Development
environments are starting to catch up with (and in some ways have
surpassed) what was on the Lisp Machine, but back then, Unix wasn't
even close.
And how is Lisp Machines more commercial than your "cheap unix
workstation" vendors like Sun?
They weren't necessarily more commercial, but they were definitely
more proprietary, far more expensive, and available from far fewer
vendors. Here's a little bit of perspective on the timeframe we're
talking about, and the way in which Lisp Machines and Unix spread.
MIT ITS ran on the PDP-10. The PDP-10 was created in 1967. The MIT
Lisp Machines were created in 1977. The first commercial Lisp
Machine was created in 1982. The last Lisp Machines were created in
1989. Unix was created in 1970. In 1976-77, Unix was given to
Berkeley, and BSD was born. At this point, people still had to buy
licenses from Bell Labs to use it, but they could get the code from
Berkeley. Certainly the explosion of Unix varieties that happened
around this time speaks to the open nature of Unix. The Lisp
Machines had almost full source available on them, but since their
operating systems only ran on Lisp machines and couldn't be licensed
to run elsewhere anyway, there was no similar explosion of Lisp
Machine varieties.
I think it's wishful thinking to blame Lisp's decline on
proprietary-ness rather than the market saying, "yeah, we just
don't get
lisp. No thanks."
I think you're very naive to think that programmers 'getting' lisp
had a great deal to do with its decline, and I think you've grossly
misrepresented my argument. Certainly the market did cause Lisp's
decline, but it was more a matter of regular market things like
price. If Lisp had run well and could have been ported widely on low-
cost general-purpose hardware, it would have continued to have
success after the rise of Unix. But had Java existed at the time, it
would have required special-purpose hardware to perform well too, and
it would have suffered the same fate as Lisp.
There are probably other reasons for the decline of Lisp as well,
such as being painted as an 'AI' language when AI fell out of style,
simply because AI is a hard problem and Lisp is good for working on
hard problems. But regardless of that, Open Source was not a factor
in the success of languages when Lisp declined (because it did not
exist yet), so it can be pinned primarily on business reasons rather
than programmer preference reasons.
Anyway, this is largely irrelevant, because there are several
different commercial Common Lisp vendors today, a few commercial
Scheme vendors, many different free varieties of both, and active
communities surrounding both. Scheme is taught in many universities
(including prestigious ones like MIT), and Common Lisp is taught in a
few as well. It just so happens that the Lisp and Scheme communities
have better things to do than write stupid apps and post them to
Sourceforge. As long as I can continue to use the languages, I'm
happy, regardless of how much FUD people who dislike them spread.
--Levi
.-----------------------------------.
| This has been a P.L.U.G. mailing. |
| Don't Fear the Penguin. |
| IRC: #utah at irc.freenode.net |
`-----------------------------------'
