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
Chris Rowley <[log in to unmask]>
Reply To:
Mailing list for the LaTeX3 project <[log in to unmask]>
Tue, 10 Nov 1998 23:57:23 +0100
text/plain (62 lines)
>   (Beware: latex support takes more
>   time than you think: I once thought it would only take up one weekend
>   and a free trip to Hamburg).

I can't think who gave you that idea;-) there's no such thing as
a free trip---what's new?

Latex/Linux: compare and contrast!

I have not yet read the article and only skimmed the discussion here
so far but the following bit of history may be relevant.

A few years ago we were challenged (I think that is the right word) to
make the LaTeX distribution and modification conditions more
GNU-like;.  Following e-mail discussions that make the current
discussion look like a single lost packet, this led to a meeting with
the Archbishop of Free Software himself, with whom Frank and I had a
very challenging (definitely the right word here) and profitable

The result was that we (well, at least the three of us) understood
several fundamental differences between maintaining (and developing) a
certain two large bits of software: one a compiler system and the
other a document processing system.

The results of these discussions for LaTeX, clarifying and explaining
our policy, are in modguide.tex.

I am not at all keen on the concept of "lessons from history" but it
seems likely that there are also fundamental differences between an OS
system and a document processing system.

One of these differences, which is very important and I think also
distinguishes LaTeX from Linux, is that LaTeX is both defining a
language for describing documents as well as implementing a processor
for that language; whereas a lot of parts of both Linux and gcc are
implementing processes whose interfaces are defined (and often
standardised) elsewhere.  Very few people send C code to others
who are not C programmers and certainly not to publishers,
secretaries, their bosses etc etc etc.  (OK, it is possible to
distribute some code packages quite widely to people who have access
to a reasonable compiler and expect it to get configured, compiled,
installed etc reasonably easily).  There is nothing in my experience
of the Linux world which is anything like any of this at all; what am
I missing?

Another is that, because the functionality of a compiler or an OS is
itself usable only via compiled executables, there is typically a
layer of people between the users of the systems and the writers of
the code that defines the systems; this multi-level structure does not
typically exist with LaTeX.  If `using Linux' for most people involved
writing bits of code in (a small subset of) C, and that code acted
directly on the kernel data structures, then I would not be `using it'
and Linus would not be a sane and contented person!

None of the above is intended to argue in favour of any particular
set-up for handling LaTeX but just to indicate that it needs its own
solution, not that which works for a different system.