LATEX-L Archives

Mailing list for the LaTeX3 project


Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Frank Mittelbach <[log in to unmask]>
Reply To:
Mailing list for the LaTeX3 project <[log in to unmask]>
Sun, 16 Feb 1997 21:07:29 +0100
text/plain (95 lines)
Narendra Ravi writes:
 > > but before going into interface questions, what are the items related
 > > to language? what are the answers to the above questions?
 > >
 > > good night
 > > frank
 > I have been following this discussion and agree with Frank. I have one
 > request/suggestion. There must already exist a list of language dependent
 > items under consideration. It will help a lot if this list is posted to
 > this mailing list. This will give us an idea of what Frank is talking
 > about.

i don't know if there is a compiled list for babel or whether one has
to get this out of the babel language files. perhaps Johannes can tell
us if he has such a list (otherwise i might compile that list myself
but that would take some time)


 > Judging from the discussion so far, the only issue of concern across
 > different languages seems to be hyphenation. I'm sure several people will
 > correct me :-).

well, there is certainly more, although perhaps not discussed so
openly. For example, the quotation characters differ in different
languages traditionally. Unfortunately LaTeX in its current usage
doesn't offer explicit markup for this, eg we neither have

 Within an English sentence one could say \quotedphrase{Dies ist
 ein deutscher Satz} or even ...

nor \begin{...} ... \end{...} or even something like

 Within an English sentence one could say \leftquote Dies ist
 ein deutscher Satz\rightquote{} or even ...

Instead we only have `` .... '' as explicit markup (suitable for
english) so that language support usually added explict support as
well, eg the EC fonts do have << >> as ligatures for french quotes
(also used by other countries) or the german package or the german
option for babel offers "`..."' (and also explicit command names) and
so on

there are many other examples and those mentioned already, eg the
usual way of displaying chapter numbers and and and, all can be viewed
as having to do with a language interface.

however, the boundaries are vague, ie while it is kind of necessary to
select the right hyphenation patterns for the current language (and
even if there is not a single right choice but several there are
clearly wrong choices, eg hyphenating english text using german
hyphenation patterns is not that good even though you might want to
use different patterns than those provided by knuth :-) the
presentation of numbers is certainly something that allows are large
variety applicable to many languages even though there might be rules
or suggestions for certain languages.

 > Dates are not really an issue, because, I consider dates to have a meaning,
 > and authors are responsible for all content related issues. In my opinion
 > both  "Dec 10, 1996" and  "9 Jan 1997" are OK. If authors choose to write
 > the date in a different form, then they are risking misinterpretation.

dates in itself might not be an issue although even this could be
challenged in certain situations, that for certain type of documents
(eg to avoid misinterpretation and to produce informtiy if for example
some dates are generated automatically and others are manually entered
by the users).

but LaTeX offers a command called \today and in fact it probably
should provide a command like \formatdate{}{}{} as well (which is not
the case). Now this command \today is an abstract thingie, eg you
might want to write something like

  Printed \today

in a document. thus in a line like that \today should not reply with a
date spelled out in German, should it?

In other words any textual string that it produced by LaTeX as part of
the formatting of abstract contents, eg the word "Bibliography"
produced by \begin{thebibliography} should match the language of the
document. Of course there might be more than one possible solution for
this and that is why one might need a flexible interface but clearly
something needs to be there for such entities.

 > LaTeX can and must only handle aspects of presentation and not interpret
 > the content of documents.

right, but that is an aspect of presentation of logical data at least
for the \today case, for the others, as i said one cane argue
depending on the application.