LATEX-L Archives

Mailing list for the LaTeX3 project


Options: Use Classic View

Use Monospaced Font
Show Text Part by Default
Condense Mail Headers

Topic: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Content-Transfer-Encoding: 7bit
Sender: Mailing list for the LaTeX3 project <[log in to unmask]>
From: Frank Mittelbach <[log in to unmask]>
Date: Fri, 19 Jul 2002 12:15:49 +0200
Content-Type: text/plain; charset=us-ascii
MIME-Version: 1.0
Reply-To: Mailing list for the LaTeX3 project <[log in to unmask]>
Parts/Attachments: text/plain (329 lines)
Jeff wrote in one of his mails he is waiting for me to return with comments
and I intend to do so, but first like to have things a bit more focused (for my
own benefit at least :-)

so what i present here is essentially a set of concerns and comments that I
gathered from the various mails that people put up as a problem with LPPL or
rather as a problem behind the ideas behind LPPL (rereading and compiling took
me roughly 7 hours last night and 4 this morning --- which makes me feel that
i'm too slow a worker these days).

I have structured it into

 a) a number of concerns together with sub-arguments if i thought that helps
    (wordings are mine)

 b) a set of mail quotes which i thought being relevant to the question of
    determining whether or not LPPL (in some form, eg after further
    cleanup/changes) would be DSFG-complient

 c) a set of mail quotes which i consider important for me decide whether to
    carry on with this or stop (and rather get some sleep again)

 d) some other mail quotes that i thought might be worth thinking about

I apologize beforehand if I have something misrepresented, or taken out of
context in a way that its meaning changed. I clearly hope not, but if so
please tell me (that's one of the purposes of this message). Similarily I
might have missed something important, again please say so in case.

I deliberately left out (finer) details of the current license text that were
discussed by Jeff and some others but which I think are more "technical" (eg
what is a filename) and resolvable easily; either by further discussion or
simply by clarifying or changing the license text (of course you may think
differently and point them out as something more fundamental which should be
discussed up front).  I have, however, put the most relevant mails (i hope) of
that type at the end of d).

What I now would like to ask you about these four blocks is:

  on a):  have i missed any important concerns or any important sub-argument
          within a concern?

  on b): have I missed any important aspects here?
         do you (dis)agree with the statements made in these mail excerpts?

  on c): do you (dis)agree with the statements made in these mail excerpts?

  on d): nothing really, though you of course are invited to suggests further
        remarks for which it would be helpful if we (everybody on this list,
        but mostly the LaTeX Project Team) think about and perhaps comments

If we could get some answers/comments with respect to this mail and the
questions posed above then i hope we can resolve these issues/concerns in one
way or another after another round of focused discussions.

compiling this stuff has helped me a lot I hope you find it useful too.

frank (taking a break)

ps I'm bcc'ing this particular message to LATEX-L (Bcc so that you don't get
rejects when you reply on debian-legal) in case people there wish to join or
listen again in at this point. As LATEX-L is a closed list (unfortunately for
this particular case) I have given up forwarding replies to it or cc'ing it
(as it was too complicated), and this will be the only message send there, so
folks if you like to see what is happening please listen at debian-legal.

| Block a)
| Concerns (not really ordered)

Concern 1: requiring a change of filename in case of modification
           in case of distribution

this seems to me the major stumbling point  for most people and (unfortunately
the most important point for us)

Argument 1.1: the need to fix a file because of a security problem

Argument 1.2: the need to fix a file because of a bug

Argument 1.3: the wish to change a file because of improvements made

Argument 1.4: when keeping the name somebody else has to approve

Argument 1.5: Changes to the LaTeX kernel to support new code (e.g klingon
              example) are made impossible this way

Concern 2: the ability to make modification without filename changes
           in case of private or "closed" use

Argument: 2.1: change article.cls to run klingon and to share it with friends
               on SuperH architecture

Concern 3: the uselessness of the approach taken by LPPL

Argument 3.1: write a new (unrelated/related?) package from scratch
              and give it the same name as the original one.

Argument 3.2: Trademarks and certification marks are tools better suited to
              controlling endorsements of conformance with a standard or set
              of usage practices.

Argument 3.3: If the core can be changed in any way without changing it
              directly, then you can break output exactly as well by this
              mechanism as you could by editing it directly.

Argument 3.4: why have a complex license if you probably never
              legally enforce it?

| Block b)
| Some more comments from mails, you tell me please if they are relevant (for
| determining LPPL to be DFSG-complient or not) and whether or not you agree
| with them:

From: Mark Rafn <[log in to unmask]>

 In order to be free, it must allow exactly what LPPL seems designed to
 prevent.  A Debian user can take LaTeX, make it behave differently than
 the original, (including producing different output), and distribute the

 If I can't call it latex, fine.  I'll call it latex-improved and set up
 the scripts that invoke it to use that by default.  If I can't do that,

 If the latex folk are looking for sections of DFSG to comply with,
 DFSG 3 is the one.  It must allow modifications.  Not "bugfixes", not
 "modifications that the author approves", not even "modifications that
 pass a test suite".

From: Mark Rafn <[log in to unmask]>
From: Jeff Licquia <[log in to unmask]>

 > It's gross, and I lose respect for the software author who put the stupid
 > requirement in the license, but it doesn't stop me fixing it nor
 > distributing the result, so it's free enough for Debian.

 I see your point, gross as it is. :-)

From: Steve Langasek <[log in to unmask]>

 So, to be DFSG free, the license must permit us to distribute software
 built from modified source code, and this freedom must even apply to
 TeX macro files; but the license may require us to use different names
 for derived works.  A modified version of a TeX macro is a derived
 work, so your license can impose this restriction and still be

 Your goals (which seem to differ markedly from the opinions of at least
 some members of the LaTeX community, judging by this thread) seem
 consistent with the DFSG.  The devil, of course, is in the details of
 the license text.  One would conclude that if your goals are consistent
 with the DFSG, and the license text is not, that the license doesn't
 truly reflect your actual goals and would need to be revised for our
 mutual benefit.

From: Jeff Licquia <[log in to unmask]>

 > On Wed, Jul 17, 2002 at 02:44:19PM +0100, David Carlisle wrote:
 > > I agree that anyone should be free to modify latex in any way and
 > > distribute that modification. I just don't agree that they should leave
 > > my email address at the top as the place where people should report
 > > bugs, and I don't agree that they should call the software by the same
 > > name, so as to deliberately trick users who believe they are using one
 > > piece of software into using a different piece of software.

 That's totally fine.

 I believe it was Frank who pointed out that there is a process for
 forking LaTeX that involves pulling the name of the LaTeX Project and
 LaTeX out of various places in the code.  If that process were described
 in the license, or explicitly named as an acceptable procedure to get
 around the file naming problem, I think Debian's complaints about the
 license would evaporate.

From: Jeff Licquia <[log in to unmask]>

 > If the command the latex uses isn't called latex then basically all bets
 > are off and it can produce anything at all.

 This is, I think, the most crucial part of this whole discussion.

 I did not see any statement to this effect in the LPPL draft that was
 posted here:

 I would love to hear that I had completely missed it, or that you've
 changed the draft to include such a statement.

 It's my belief that if a statement to this effect were added to the
 license, then there would be no question as to the license's compliance
 with the DFSG, either as it is or with DFSG 4 removed.

From: Sam Hartman <[log in to unmask]>

    Jeff> I'm curious.  From my perspective, the Debian people seem
    Jeff> not to have been contradicting themselves so far; I

 But we do seem to be contradicting our actions and unclear on the
 implications of DFSG 4.

 I think DFSG 4 means that you can require renaming or patch files of the
 sources.  It also seems that you can require renaming of the distributions.

 What Debian finds unacceptable is the assertion that we must break the
 TeX or LaTeX API (hey you said it was a language) in order to make
 some changes.

| Block c)
| Some more comments from mails that I found important for my further
| involvment. Please tell me if you agree with them.

From: Branden Robinson <[log in to unmask]>

 TeX is currently under discussion, so I will await the outcome of this
 discussion to opine on it.  (In Debian, LaTeX is provided as part of
 the TeTeX distribution, so any license problems with LaTeX will have a
 transitive effect on TeTeX for us.)

From: Steve Langasek <[log in to unmask]>

 And I definitely think moving LaTeX to contrib would be a loss for all
 Debian users.

From: Jeff Licquia <[log in to unmask]>

 Our Social Contract states that "Our Priorities are Our Users and Free
 Software".  User needs are as important to us as freedom, and we
 wouldn't be serving our users if we broke LaTeX document compatibility.

| Block d)
| Finally some snippets from mails or pointers to mails
| that i thought worth thinking about or
| commenting on (is isn't the full list probably) ...

From: Nick Phillips

 If your intention is to ensure that anything called LaTeX produces identical
 output to the then-current "official" implementation, then apply Occam's
 Razor and say that.

From: Glenn Maynard <[log in to unmask]>

 > >> but the documents created using that distribution. If I get a
 > >> document by "John Smith" (somehow), how can I see if _his_
 > >> system had a modified latex?

 Others (eg Boris) seem to be saying that the Latex developers don't
 mind if you rename Latex to something else and modify it such that
 \usepackage{article} in a document does something different than it
 does in the real latex.  You're contradicting this, so I'll discontinue
 this thread unless Frank or David speaks up to clarify this.

From: David Carlisle <[log in to unmask]>

 LPPL does not (unlike some other licences:-) insist that any derived code
 is licenced in the same way, but it does state that any derived works,
 apart from having different names, should be licenced with a licence
 that forbids renaming back to the original.

From: Chris Lawrence <[log in to unmask]>

 How does the LPPL actually prevent a distributor from FUBARing their
 distribution of LaTeX?  The fact is people regularly ignore licenses,
 copyrights, patents and trademarks (if this weren't the case, there'd
 be almost no GIFs or MP3s in the world).

From: Jeff Licquia <[log in to unmask]> (if i remember correctly)