[twg-tds] scripts, enc/lig/map

[Paul Vojta]

> From: Reinhard Kotucha <reinhard.kotucha at web.de>
> Date: Fri, 20 Feb 2004 00:49:07 +0100
> To: Hans Hagen <pragma at wxs.nl>
> Subject: Re: [twg-tds] scripts, enc/lig/map
> Cc: twg-tds at tug.org
> 
> Why do we need texmf/scripts at all?
> 
> There are several reasons:
> 
>  1. Maintenance:  if someone sends a patch, it is desirable to be
>     able to apply it once rather than walk through the directories and
>     overlook most of them.

I've been maintaining TeX for the UCB Math Dept's system for many years now,
and also for my home machines, but I don't think I've *ever* applied a patch.

>  2. A modified script can be put into texmf-local or
>     texmf-use-with-care.

Or /usr/local/bin, or ~/bin.  Probably most users would be more comfortable
with ~/bin than setting up their own texmf trees.

>  3. The way how a script is launched is system dependent even if the
>     script itself is not.  UNIX systems understand the comment
>     "#!/usr/bin/perl", others require to call perl with the name of
>     the script as an argument.

Seriously, how many people worldwide are using the same texmf tree for
such different systems?  To do that, you'd pretty much have to be running
two different TeX distributions with the same texmf tree (e.g., tetex
runs only on systems that respect #!, afaik).  Has this been discussed
before?

In my department, we run TeX on Solaris and Linux, but the way our servers
are set up, we have two separate texmf trees even for these very similar
systems (both of which are modified tetex).  Lots of people have dual-boot
machines at home (including myself), but since Linux doesn't mount NTFS
with write access unless you're willing to take risks (last time I checked),
sharing texmf trees would also be dubious IMHO.

Sharing texmf trees between distributions sounds like poor sysadmin practice.
For example, pdftex.cfg has different options depending on the version.
You'd have to have some coordination between distributions, which I haven't
heard of to date.

[snip]

> And if Fabrice compiles texexec into a binary, I don't see any reason
> _not_ to put this at bin/Win32/texexec.exe.  IMHO this is the correct
> place. 

Exactly.  Scripts should be no different.

> As I said before, beeing able to switch between texmf trees from
> different distributions is very interesting.  The more I think about
> it the more I like it.

I don't want to interfere with your computing enjoyment, but right now,
doing that seems like it would lead to problems far more difficult than
lack of a scripts directory.  And, again, how many people worldwide would
be sharing texmf trees between distributions?

The vast majority of TeX users write documents, they don't create macro
packages.  They don't develop TeX software, and they don't modify the
scripts that come with TeX.  Wrapping all the scripts exacts a small
but measurable performance hit *every time they use those scripts*.
And what is the benefit?  It goes to a small minority of system admins
in a very few special cases, and only makes a difference when they are
updating the system, which is also rare.  This goes to the heart of why
I've never liked kpathsea.

>     > I'd also like to have this directory optional, at the discretion
>     > of the distribution maintainer.
> 
> If the result of the discussion will be that texmf/scripts is useful
> then I don't think that it should be optional.
> 
> If people understand what it is good for, they will use it.  If not,
> empty directories can be omitted anyway.

This last paragraph sounds like you're already assuming that texmf/scripts
is optional.

As for the middle paragraph, there are trade-offs.

--Paul Vojta, vojta at math.berkeley.edu