In-reply-to: <950627121109.202022c7@vms.rhbnc.ac.uk> from "Philip Taylor" at
 Jun 27, 95 12:11:09 pm
Reply-to: Mailing list for the LaTeX3 project <LATEX-L@VM.URZ.UNI-HEIDELBERG.DE>
Message-id: <9506271159.AA08030@inuo30>
MIME-version: 1.0
Content-type: text/PLAIN; CHARSET=US-ASCII
Content-transfer-encoding: 7BIT
Date: Tue, 27 Jun 1995 13:59:19 +0200
From: Johannes Waldmann <joe@INUO30.MATHEMATIK.UNI-JENA.DE>
Sender: Mailing list for the LaTeX3 project <LATEX-L@VM.URZ.UNI-HEIDELBERG.DE>
To: Multiple recipients of list LATEX-L <LATEX-L@VM.URZ.UNI-HEIDELBERG.DE>
Subject: Re: The proposed policy on distributing and [not] modifying files
Status: R

Some thoughts on LaTeX modifications. I'm not at all
a LaTeX programmer. I'm just a user, but I had to act as a site
maintainer (at work, and at home - linux) from time to time.
 
A restiction `you are not allowed to change file xxx' sounds
quite useless to me. This is like telling your system administrator:
you should not make `gcc' an alias for `rm -rf *'. If he really wants
to do that, then he will do; but normally, he could have no reason
to want such a thing.
 
Of course, matters are easier with `gcc' because there is only one such
thing and all agree on what it is, and should be doing. Things are
different with `my array.sty' and `your array.sty' because different
people might think different on what an array style should do.
 
I see two possible solutions here (at least theoretically).
 
Most important is, could the `you shall not change file xxx' be applied
to the CTAN/incoming directories. A program could easily check if
someone tries to submit files that duplicate already existing names,
and reject such submissions. So spreading of duplicate names is
essentially prohibited because all of the larger archives (should) copy
their stuff from some CTAN host and not from somewhere else.
 
Another point here. Even if `official/array.sty' exists, perhaps
`fsf/array.sty' should be allowed. But then the LaTeX executable needs
a different method of file searching. What I have in mind is
what (g)cc does: _never_ search recursively! If you need fsf/array.sty,
then you should write \input{fsf/array.sty}. Or, if you want everything
formatted `fsf-like' (of course I'm referring to Stallman's hypothetical
example), and you don't want to change your \input{array.sty} in the
document, then you should call LaTeX with -I./fsf/ or something.
 
Another question then is whether to allow such `-I' options inside
a LaTeX document, or _only_ inside a document (and not on the command line)
- this seems like `use' clauses in Ada, and is controversial over there
as well.)
 
--
Johannes Waldmann,  Fakult\"at f\"ur  Mathematik und Informatik
UHH,  Jena,  D-07740,  Germany,  phone (03641) 6 30793,    joe@
inuo30.mathematik.uni-jena.de ... Themenwechsel.  Themenwechsel
ohne  Tuwoerter  als schlichte  Geraeuschbeispiele, nicht mehr.