mwm@eris.UUCP (04/22/87)
I'm sorry I ever started this mess ("so are we," most of the net
replies). Not because of what I said, but because of what followed.
I'm going to defend myself a bit, clear up a minor misrepresentation,
and then shut up about it ("Good!" says everybody). If you want to
talk about it, I'll answer email.
From 7OHN:
> Hey Mike, earlier rebuttal (about you)was not totally serious (but to the person
> that rubutted my rebuttal to your rebuttal to my rebuttal). Which
> brings us back to the *original* issue:
Nah, you just took the chance to say I was a blithering idiot. Whether
it's true or not, I resent it. You think what I said was vague &
abstract. I'll conceede that on one of the six counts. The rest of the
points used terminology that anyone who keeps up with language design
issues should be familiar with. Anyone who knows both the terminology
and the languages in question can probably point out the features (or
lack thereof) I meant. The first point (broken type system) is vague
because nobody seems to agree on what a "non-broken" type system is.
Further, the problems aren't "abstract," or "theoretical" problems.
I've had to code around each of these problem areas at least once.
Mitsu posted a nice summary, plus a language taxonomy. I don't quite
agree with his taxonomy, but it's perfectly valid. Two major problems:
1) I think Modula 2 should be classed with the 4gl languages
(Ada/CLU/Simula-67/Smalltalk). After all, it purports to have data
abstraction.
2) The post-3gl languages aren't "AI" languages. They just used
concepts from that field. They can be, and are, used for many types of
programming tasks.
And a minor nit:
I don't think there _is_ a 5gl language yet. there are lots of things
floating around pushing new concepts: FP, KRC, CSP, prolog, Russell,
and others. I don't expect any of those to be one of the major
languages of the next generation; I expect some set of features from
those to be in all the 5gl languages, though.
And now, more from 7OHN:
> Programming on the Amiga: To C or not to C, Modula-2 or C.
Now that we've covered the specifics, I can make a general statement.
That situation is almost true, and it _sucks_.
> I am weary of theory and circumlocution of issues.
So quit flaming and circumlocating.
> Not I will to any more of this line anyone respond...
Good.
Finally, I'd like to point out that 1) programing languages are like
toothbrushes. If you're perfectly happy with what you've got, there's
no need to change, and 2) there's nothing wrong with not keeping up
with language design issues, or asking for explanations of terminology
you don't understand. But flaming it as being "to abstract" when you
don't understand it is another matter.
<mike
P.S. - just to prevent flames; I've been using Ngl in the context of
Mitsu's posting. This only roughly corresponds to standard useage of
the term "fifth generation."
--
Here's a song about absolutely nothing. Mike Meyer
It's not about me, not about anyone else, ucbvax!mwm
Not about love, not about being young. mwm@berkeley.edu
Not about anything else, either. mwm@ucbjade.BITNET