Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!rutgers!ucla-cs!zen!ucbvax!decvax!ima!johnl
From: johnl@ima.UUCP
Newsgroups: comp.compilers
Subject: Re: Algol 68
Message-ID: <652@ima.ISC.COM>
Date: Fri, 14-Aug-87 12:00:26 EDT
Article-I.D.: ima.652
Posted: Fri Aug 14 12:00:26 1987
Date-Received: Sun, 16-Aug-87 05:45:46 EDT
References: <646@ima.ISC.COM> <648@ima.ISC.COM>
Sender: johnl@ima.ISC.COM
Reply-To: harvard!ut-sally!utah-cs!shebs (Stanley Shebs)
Organization: PASS Research Group
Lines: 31
Approved: compilers@ima.UUCP

Original-sender:


In article <648@ima.ISC.COM> steven@cwi.nl (Steven Pemberton) writes:
>[...]
>On the subject of two-level grammars, it seems to me a pity that they
>receive so little attention. They use an extremely simple mechanism,
>and are as powerful as a Turing machine, so you can define the syntax,
>context conditions, and semantics with the one formalism. Again I
>blame the rather heavy style of the Algol 68 report for giving them a
>bad name. For instance, if they'd only used ROWS-OPTION instead of
>ROWSETY to indicate an optional ROWS, it would have made it much
>easier reading.

Two-level grammars have the same problem that attribute grammars do -
they are based on the belief that the world is linear strings of tokens.
Sooner or later, simple grammars don't work, so people have introduced
some pretty bizarre schemes to patch things up.  (My first response to
attribute grammars was the word "kludge".)  The schemes inevitably go all
the way to Turing equivalence, which should be a warning signal that
you're no longer dealing with strings of tokens, except in a purely
formal sense, and that other general computational mechanisms are worth
thinking about (recursive functions, logic, term rewriting, etc).

Of course, the fact that I've been hacking Lisp for the past five years
has nothing to do with this view of grammars.  :-)

							stan shebs
							shebs@cs.utah.edu
--
Send compilers articles to ima!compilers or, in a pinch, to Levine@YALE.ARPA
Plausible paths are { ihnp4 | decvax | cbosgd | harvard | yale | cca}!ima
Please send responses to the originator of the message -- I cannot forward
mail accidentally sent back to compilers.  Meta-mail to ima!compilers-request