ap

[Byron Rakitzis]

I predicted I would get a lot of flak over this. No surprise to see it coming
from one of the perl wizards.

In article <1991Jun04.182921.1685 at convex.com> tchrist at convex.COM (Tom Christiansen) writes:
>nauseated?  you must have a weak stomach.

I'm not the only one. I have found that people either love perl or hate perl. I think
I am not in a small minority when I say that I belong to the latter category.

>I've never met a *good* C programmer who's had any real problems with it.

What is a *good* programmer? One who has no trouble learning perl? This is
a very circular definition. Otherwise, I don't think you're in any position
to judge my ability as a programmer. This is just ad-hominem.

>There may be a momentary bit of discomfort at seeing dollar signs and
>thinking of BASIC, but this quickly passes.  Certainly no *good* programmer
>finds it hard.

Ok, I haven't seen any *good* programmer have any trouble with JCL either.
After the inital discomfort of having to type all those //, the feeling
soon passes. (!!)

>You mean like these:
>
>    ypcat hosts | perl -ane "print $F[0]\n"
>    ypcat hosts | perl -pe "s/\s.*//'
>
>or skipping ypcat:
>
>    perl -e 'while (@F=gethostent) { print "$F[0]\n"; }'
>    perl -e '$\ = "\n"; while (($_)=gethostent) { print; }'

Yes, I mean like those. I think my hypothetical example was more concise
than any of the suggested replies above.

>But certainly you have a point -- down with expressiveness!  Let's also
>redesign English so there are no synonyms, either lexical or phonetic.

I think you miss the point. An expressive language does not imply a perl-
like language. A language with the powerful features of perl that's been
designed carefully without trying to incorporate syntactic features of
every programming language known can be just as expressive.

I don't want my language to have a BASIC way of doing things, an APL way
of doing things, a C way of doing things...

>And let's please fix C while we're at it.  The way you declare arrays of
>pointers to functions returning pointers to functions returning pointers
>to stat structs really nauseates me.  Plus did you know that C has several
>ways of expressing the same thing?  It's really disgusting.  Check out the
>flow control: all you need is a for loop; we should abolish the while and
>the do loops.  Or how about saying char *foo versus char foo[] in formal
>parameters? What stupid redundancy!  Let's all flame Dennis Ritchie.

Do you know what part of C Dennis Ritchie is most unhappy with? The declaration
syntax! Yes! It's stupid and ugly and hard to understand. Sure, a *good* C
programmer has no trouble parsing

	double (**foo)(void (*)(int));

but I'm sorry, I forgot I was not one of those.

>If your sensibilities are so very offended by its lack of apparent beauty,
>then fine, go off and write your own language.  But don't go asking the
>net to tell you how to do it.  Great things come from single visions, not
>from designs by committee.

I don't see what your problem is. Clearly perl was not written in a single
vision. If *anything* is an agglomeration of random features, it has *got*
to be perl.

Also, *EXCUSE ME* for asking the net about their views for an alternative to
perl. Perhaps we should just lock an undergraduate in a dark-room with a
decstation and wait for her to come up with the successor to perl. After all,
programs are not designed by committees, right?

"A system without perl is like a hockey game without a fight"
--
Byron Rakitzis
byron at archone.tamu.edu