References: <199811030953.JAA04308@nag.co.uk> <98102213123551@man.ac.uk>
            <199811031653.KAA12001@dcdrjh.fnal.gov>
            <v03110701b264e9f56400@[130.237.37.58]>
            <19981104145716.B29267@maths.tcd.ie>
Message-ID:  <199811041953.UAA29329@frank.zdv.uni-mainz.de>
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@URZ.UNI-HEIDELBERG.DE>
In-Reply-To:  <19981104145716.B29267@maths.tcd.ie>
Date:         Wed, 4 Nov 1998 20:53:35 +0100
From: Frank Mittelbach <Frank.Mittelbach@UNI-MAINZ.DE>
Sender: Mailing list for the LaTeX3 project
              <LATEX-L@URZ.UNI-HEIDELBERG.DE>
To: Multiple recipients of list LATEX-L
              <LATEX-L@URZ.UNI-HEIDELBERG.DE>
Subject:      Re: Quotes and punctuation
Status: R

Timothy Murphy writes:
 > Apologies for asking a very elementary question,
 > but is the problem in this case that one gets into difficulty
 > if one says \quote{...\quote{...}...} ?
 > Would there still be difficulty if one said
 > \quote{...\inquote{...}...} ?
 > Ie is it the nesting of \quote that causes the problem ?

essentially what we are trying to explain is that TeX's mechanism of
\futurelet and its expansion mechanism are not adequate to model
character stream parsing if those character streams might be hidden
within commands (expansion) or arbitrary nesting of brace groups can
intervene.

 in the later case expansion can also screw up simple \futurelet types
of parsing, eg in

  \textbf{...\quote{...}},

by the time the end of the quote could look forward to see if a ,
follows it will encounter all kind of expanded internal code coming
from the fact that \textbf has expanded and placed it there and it has
no way of dealing with this.

the two mechanisms just don't work together except in very tightly
controlled cases of which general text is not really part of.

so what is needed for typesetting support on such a level are more
powerful concepts that better distangle character stream parsing and
analysing and macro (or other) processing via expansion etc.

Omega's OTPs might be a partial answer to this kind of question
(though in my understanding they will not be the final answer at least
not in their present form)

for this reason Chris "homework" suggestion is meant really serious
and would help a lot more than discussing Godel theorem or what can be
represented in a computer.

frank