Message-ID:  <9604031050.AA26966@r8h.cs.man.ac.uk>
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@relay.urz.uni-heidelberg.de>
In-Reply-To:  <199604030745.JAA04157@murnau.idris.fr> (message from Bernard
              GAULLE on Wed, 3 Apr 1996 09:45:16 +0200)
Date:         Wed, 3 Apr 1996 11:50:12 BST
From: David Carlisle <carlisle@CS.MAN.AC.UK>
Sender: Mailing list for the LaTeX3 project
              <LATEX-L@relay.urz.uni-heidelberg.de>
To: Multiple recipients of list LATEX-L
              <LATEX-L@relay.urz.uni-heidelberg.de>
Subject:      Re: Changing \hyphenchar
Status: R

DC> (The fd files distributed with the dc fonts do set it, but that is
DC> another issue)

BG> hum... this is the root of our discussion.

Sorry I should have made myself more clear.

Currently there are three sets of fd files distributed for dc fonts.

* The set that LaTeX uses by default if `unpacked' in the standard way
  use the new release 1.2 names but do *not* set \hyphenchar to 127

* The alternative set in olddc.ins use the old release 1.1 names, and
  do *not* set \hyphenchar.

* The set distributed by with the dc fonts use the new release names
  and *do* set hyphenchar to 127.

By `not the issue' I just meant that in that particular message I was
only discussing the versions in the LaTeX distribution, ie the first
two sets above.

Clearly it is not ideal that there should be two `competing' sets of
fd files for the new fonts, but this is, I think, a reasonable
compromise position at present for the following reason. As I
mentioned last time, LaTeX currently does not have any real interface
for users to make a choice about this hyphenation behaviour, Hopefully
this round of discussion may lead to such an interface....
So in the present circumstances having the two sets of fd files is
quite useful as it makes a choice reasonably easy to document as
follows:

If you have the new dc fonts you can

*either* use the traditional TeX - hypenchar, and traditional
         hyphenation patterns, in which case use the Standard LaTeX fd
         files,
*or*     alternatively use the hanging punctuation character from
         position 127 and extended hyphenation patterns describing
         hyphenation near -, in which case you should use the fd files
         coming with the dc fonts.

I am not saying this is an `ideal' interface, but is more or less the
work-around solution that is in place until an interface can be
developed.

> again, as long as fd files distributed with the DC force
> \hyphenchar to `177
> there is no issue to \defaulthyphenchar.
True, but I was only describing how to use the 127 slot with the fd
files in the LaTeX distribution. Your example comes from the `third
set' of fd files in the dc font distribution.

> obviously ^^7f needs to have catcode of 11, otherwise it will not do
> its job. Why not? but we can't imagine the user typing ^^7f each time
> it needs a compound word mark.

Well I agree that anyway typing ^^7f is not a possibility, but unless
I misunderstand what you mean, even catcode 11 would not help. The
^^7f glyph in the new dc fonts is not suitable as a compound word
mark, with any catcode. It has to come at the end of a line it may not
be used mid-line as it sticks out on the right.

However all this discussion is clarifying what the *current* position
is. Probably a more useful discusion is to discuss what functionality
people need for various languages, and how that might be declared...

David