Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ames!oliveb!apple!claris!hearn
From: hearn@claris.com (Bob Hearn)
Newsgroups: comp.lang.misc
Subject: Re: Language Design
Message-ID: <9140@claris.com>
Date: 22 Mar 89 20:14:56 GMT
References: <9122@claris.com> <2970@goofy.megatest.UUCP>
Reply-To: hearn@claris.com (Bob Hearn)
Organization: Claris Corporation, Mountain View CA
Lines: 43

In article <2970@goofy.megatest.UUCP> djones@megatest.UUCP (Dave Jones) writes:
>The term "context-free", as Chomsky defined it is not very useful
>to us computer types, because no one ever, ever writes grammars any
>other way.  To him, "context-free" meant that a rule could only
>have one symbol on the left-hand side. "Context-sensitive" meant that
>production rules could have strings of symbols on the left:
>
>       a X b <= a x y b
>
>In the above, the a and b form a "context" for the reduction
>X <= xy.

Precisely!  But, gee... I've considered myself a computer type for, oh, 12
years or so, and it's useful to me.  How you write grammars depends on
what you're going to use the grammars for.  I don't think anyone could
write a compiler, for example, without being at least familiar with 
terms of formal language theory like "context-free" and "context-sensitive."

>
>I haven't been following the discussion, but I suspect that the
>correspondents were not alluding to the unuseful Chomsky definitions,
>but instead were using the terms quite literally: "context-free" meaning
>"context-free".  I think that is a perfectly reasonable thing to
>do, especially if one is unfamiliar with the confusing Chomsky
>definition.  (I never did like the practice of "defining" something
>that already had a meaning in normal English.)

If you *had* been following the discussion, you'd have seen that my original
posting was a followup to someone complaining about how screwed-up C's
grammar was, that it wasn't context-free.  When someone uses the words
"context-free" and "grammar" in the same sentence, I assume he knows what
he's talking about, and what he said was incorrect... K&R's grammar for C
*is* CF, and not particularly screwed-up.  I've received no end of mail
and flames for that simple correction!  As for "context-free" already
having a meaning in normal English, I disagree.  One could say that
you need context to parse C, but if you use the phrase "context-free" in
the realm of computer science, it's because you've heard it before, and
you should find out what it means before using it.

Bob Hearn
hearn@claris.com

P.S. I'll chill when I stop getting flamed.