Received: by nummer-3.proteosys id <01C19443.A93588B4@nummer-3.proteosys>; Thu, 3 Jan 2002 11:44:46 +0100 Return-Path: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C19443.A93588B4" x-vm-v5-data: ([nil nil nil nil nil nil nil nil nil][nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil]) X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message Subject: arguments pro/con patching the pool Date: Mon, 27 Apr 1992 22:24:40 +0100 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: From: "Frank Mittelbach" Sender: "Mailing list for the LaTeX3 project" To: "Rainer M. Schoepf" Reply-To: "Mailing list for the LaTeX3 project" Status: R X-Status: X-Keywords: X-UID: 695 This is a multi-part message in MIME format. ------_=_NextPart_001_01C19443.A93588B4 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Michael said: > Come to think of it, this probably could be achieved by changing the > string pool file also, since the opening and closing parentheses of > file name messages are also in the string pool. But not without > changing the checksum. For more extensive changes to the string pool, > it would be best to make a WEB change file rather than editing the > string pool file directly. This would however allow some other nice > customizations such as changing error and help messages, e.g. change > It is in fact not necessary to touch the checksum; TeX will happily use a patched pool as long as the checksum stays the same and the two digits in front of every line correctly reflect the number of characters in the line. That way I provided german error messages under PC-TeX four or five years ago. The question of whether or not patching the pool file is allowed under Don's understanding of his copyright to TeX is interesting and if there should be finally good reasons to addopt such a strategy I will explicity talk to him about it the next time I do meet him. Nelson said that another strategy would be to make a general agreed addition to every change file for every TeX implementation. This would be another way but this would mean that the additional benefit like making the latex error messages better wouldn't be possible. In any way, if such a change would come it would be easy to incorporate it into ltx3 under whatever scheme we use. The main argument against patching in my eyes is that it makes the installation of latex more difficult. This may be okay for unix etc. since installation can be done by providing a makefile, but for the many single users out there with some PC or whatever it would mean that they need to understand much more about the system. I'm not sure if it would be possible to provide some sort of installation script for every major platform. But perhaps, people can prove me wrong there? Frank ------_=_NextPart_001_01C19443.A93588B4 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable arguments pro/con patching the pool

Michael said:

> Come to think of it, this probably could be = achieved by changing the
> string pool file also, since the opening and = closing parentheses of
> file name messages are also in the string pool. = But not without
> changing the checksum. For more extensive = changes to the string pool,
> it would be best to make a WEB change file = rather than editing the
> string pool file directly. This would however = allow some other nice
> customizations such as changing error and help = messages, e.g. change
>

 It is in fact not necessary to touch the = checksum; TeX will happily
use a patched pool as long as the checksum stays the = same and the two
digits in front of every line correctly reflect the = number of
characters in the line. That way I provided german = error messages
under PC-TeX four or five years ago.

The question of whether or not patching the pool file = is allowed under
Don's understanding of his copyright to TeX is = interesting and if
there should be finally good reasons to addopt such a = strategy I will
explicity talk to him about it the next time I do = meet him.

Nelson said that another strategy would be to make a = general agreed
addition to every change file for every TeX = implementation. This would
be another way but this would mean that the = additional benefit like
making the latex error messages better wouldn't be = possible. In any
way, if such a change would come it would be easy to = incorporate it
into ltx3 under whatever scheme we use.

The main argument against patching in my eyes is that = it makes the
installation of latex more difficult. This may be = okay for unix etc.
since installation can be done by providing a = makefile, but for the
many single users out there with some PC or whatever = it would mean
that they need to understand much more about the = system. I'm not sure
if it would be possible to provide some sort of = installation script
for every major platform. But perhaps, people can = prove me wrong
there?

Frank


------_=_NextPart_001_01C19443.A93588B4--