TDS on MS-DOS (was: Helmut Kopka's interpretation of the TDS)

Paul A Vojta twg-tds@mail.tug.org
Fri, 22 Nov 1996 23:20:31 -0500


Under MS-DOS, with appropriate changes made to TeX and its support programs,
you can get away with:

	set FOOTEXCNF=c:/footex

in your environment, and use a config file c:/footex/footex.cnf
to contain everything else; e.g.,

	TEXINPUTS.amstex        = .;c:/footex/texmf/tex/{amstex,plain,generic}//

etc.  

Borland C, for example, uses a similar approach.
In fact, BCC looks at argv[0] to guess where the config file is located,
so you don't even need an environment variable.

This is not much environment space.

-----

At Berkeley, we have a full teTeX setup, with some additional macro files,
compiled for two architectures.  Total space:  90M.  Look in PC Magazine,
or anywhere else, for that matter.  It's hard to find a hard disk smaller
than 500M these days.

So I don't see a big demand for splitting up the TeX tree, and if so,
it can be handled in the config file.  Is there any reason for limiting
the size of the config file, other than programming convenience?

-----

Recursive search does not have to be slow.  An appropriately modified TeX
can use an ls-R file; this is much quicker.  Numerous versions of "ls" are
available for DOS.

-----

Now you may be thinking, without ever saying so, that with the current TeX,
you can't do the above; you have to do stuff like

set TEXINPUTS=.;c:/footex/texmf/tex/amstex/base;c:/footex/texmf/tex/eplain; \
  c:/footex/texmf/tex/generic;c:/footex/texmf/tex/generic/babel;c:/footex/te \
  xmf/tex/generic/config;c:/footex/texmf/tex/generic/hyphen; etc.

instead of the above.

That is, of course, not feasible, but (AFAIK) all the advocates of TDS
freely admit that TDS is not feasible without appropriate changes to TeX
and associated programs.  And those changes, I admit, are not easy to
implement.

If you just said, "I don't want to make those changes," then that's perfectly
fine.  In that case it's perfectly reasonable not to use TDS.

-----

> The problem is not that *I*, as an individual, uses or not config
> files or recursive search. The problem is that *users* can use
> long environment variables, this is allowed. So your argument is
> more or less the same as "to avoid car accidents, forbid people
> to use their cars".

To avoid car accidents, obey the rules of the road and use common sense
(in this case: edit the config file).

-----

--Paul Vojta, vojta@math.berkeley.edu (temporarily at math.ias.edu)