[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#189370: acknowledged by developer (irrelevant)

On Wed, 23 Apr 2003 16:45:11 +0900 (JST), Atsuhito Kohda <kohda@pm.tokushima-u.ac.jp> said: 

> From: Manoj Srivastava <srivasta@debian.org> Subject: Re:
> Bug#189370: acknowledged by developer (irrelevant) Date: Mon, 21 Apr
> 2003 20:45:44 -0500

>> For example, I set up a Debian machine in a lab with other, non
>> debian machines. I note that all the machines have default
>> texmf.cnf behaviour. No problem, I create a custom texmf.cnf, and
>> distribute it to all machines.

> I don't know how you create texmf.cnf but it would be enough if you
> create it in Debian and distribute it to other machines (but under
> the condition that you use only compatible TeX components which is
> your case, perhaps).

	No, since the Debian admin is not in charge. This is an
 established lab, with Debian trying to make inroads. If Debian
 is not inter-operable, it is useless.

	I am boggled at the Microsoft like insistence that sure, we
 are not inter-operable with the rest of the world, but if the rest of
 the world lets us take charge, and follow the Debian way, they shall
 be enlightened.

	What is the other TeTeX installations on these machines also
 had non standard procedures for configuration? 

	Right now, Debian is not compatible with the other TeTeX
 installs, and this is a bug. 

> But it seems you dislike it by some reason or other and they are
> your machines so it's okay how you treat them and to complain or to
> file a bug as you like.

>> Every other machine works. But the Debian box, despite having my
>> nice, fancy, /etc/texmf.cnf, does not pay any attention to it.

> Ditto.

	Yes, this is a Debian bug.

>> Hmm. Red Hat Works. Suse Works. Solaris Works. Debian fails.  Why
>> does Debian have to be incompatible? I say this is a bug.

> Ditto.

	Yes, this is a Debian bug.

> Because RedHat, for example, is a comercial distribution so RedHat
> would be designed from the biginning what TeX components it would
> include, therefore a static texmf.cnf worked.  Further, RedHat
> doesn't have something similar to our policy, perhaps, so it can
> modify texmf.cnf freely if necessary, I guess.  (Correct me if I
> misunderstand RedHat.  I've never used it.)

	It does not matter why they did not break compatibility; we
 did. And I can coime up with half a dozen mechanisms, including using
 ucf, that creates working, compatible, inter-operable,  solutions.

	I am surprised you can't seem to think of even one of these.

> From: Manoj Srivastava <srivasta@debian.org> Subject: Re:
> Bug#189370: acknowledged by developer (irrelevant) Date: Mon, 21 Apr
> 2003 20:48:14 -0500

>> In he old scheme, my changes were never lost.

> If you like the old scheme, it is possible only if policy doesn't
> forbid to modify a conffile, texmf.cnf, by packages' scripts.  (In
> fact, some packages did it before.)

> BTW, does policy force us that our configurations should be
> compatible with those of RedHat?

	No, common sense does. Not all bugs are policy violations. And
 this is not just Red Hat. It is _any_ other TeTeX installation,
 since I mentioned more than Red Hat in this DARPA lab.

If you want to see card tricks, you have to expect to take
cards. Harry Blackstone
Manoj Srivastava   <srivasta@debian.org>  <http://www.debian.org/%7Esrivasta/>
1024R/C7261095 print CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E
1024D/BF24424C print 4966 F272 D093 B493 410B  924B 21BA DABB BF24 424C

Reply to: