Path: utzoo!yunexus!geac!syntron!jtsv16!uunet!ncrlnk!ncr-sd!hp-sdd!ucsdhub!ucsd!nosc!helios.ee.lbl.gov!lll-tis!ames!oliveb!intelca!mipos3!omepd!bobdi
From: bobdi@omepd (Bob Dietrich)
Newsgroups: comp.lang.pascal
Subject: Re: Pascal dying out?
Summary: WRONG !!!!!!!!!
Message-ID: <3884@omepd>
Date: 28 Oct 88 09:12:31 GMT
Article-I.D.: omepd.3884
References: <267@lafcol.UUCP> <9631@swan.ulowell.edu> <6157@claris.com>
Organization: Intel Corp., Hillsboro, Oregon
Lines: 58

[This is going to be a lot shorter than I'd like (quit cheering!), but I'm
going out of town in about 9 hours.]

I think the conclusions you've reached based on your limited sample are
wrong. I work for a company that sells Pascal, have regular contact with
many other Pascal vendors, and I think it's far from dying out in the near
term.

Certainly one factor that may have tinged the response to your question is
the forum you asked it in: usenet. The majority of machines on usenet are
running some flavor of Unix. With C being the implementation language of
Unix, and with Unix needing to be worked on constantly, there's certainly
going to be a much higher percentage of C devotees on usenet. Also, often
(note I'm not saying always!) these Unix machines are not in a department
doing product development work. Even though Unix is a popular bandwagon, it
does not come close to controlling a majority of the computers in the field.

Now that I've stirred up the hornets's nest (he he), on to other points. I
think if you simply take a look at the languages offered by vendors that are
in the business of selling languages or computer systems, you'll see that
Pascal is almost always listed. Unless it's trying to be a tax write-off, a
company simply cannot afford to support a language unless there's demand for
it. And frankly, educational institutions don't have enough clout with
industry to force a manufacturer to support a language solely for teaching.

Successful languages, like many things, go through a life-cycle. At
first, there's just a small band of missionaries spreading the word.
Then, if the word has any merit, it catches on. Articles and books are
written, further spreading the enthusiasm. The word becomes widely
discussed, and some people even go so far as to actually use the
language. The language becomes more commonplace, a better vehicle of
communication, and it enters its workhorse phase. Since the word is now
common knowledge, there's no sense writing about it so much anymore, so
editors and publishers jump on the next bandwagon that rolls along. The
workhorse continues to plod along, creating excitement only if it's not
there to do its job. It gets replaced only when there's a workhorse that's so
much better that it's economical to forget how much you invested in the older
workhorse. Pascal is in the workhorse stage, and there hasn't been a suitable
replacement come by yet. [Yeah, I know I switched metaphors in midstream; so
what.]

In the relatively short time people have been seriously using programming
languages, only a handful have made it to workhorse status: COBOL, FORTRAN,
ALGOL, BASIC, Pascal, LISP, and C. Others faded before they got there, and
ALGOL and BASIC have pretty much reached the end of their useful life. The
others that have survived have gradually changed over the years, and Pascal
is about to undergo such a change with Extended Pascal. Extended Pascal
should offer enough of what users want (modularity, strings, random I/O,
flexible arrays, standardization, etc.) that it should renew interest in
the language. All this without screwing up sets like Modula-2 did. If it
doesn't renew interest, it will be around until something better comes
along. Which will be a while.

Enough. I have to go pack.

usenet:	uunet!littlei!ihf1!bobd			Bob Dietrich
  or	tektronix!ogcvax!omepd!ihf1!bobd	Intel Corp., Hillsboro, Oregon
  or	tektronix!psu-cs!omepd!ihf1!bobd	(503) 696-2092