Branden Robinson <[EMAIL PROTECTED]> wrote: > On Sun, Jul 11, 2004 at 02:38:20PM +0200, Hilmar Preusse wrote: >> On 11.07.04 Branden Robinson ([EMAIL PROTECTED]) wrote: >> > Hmmm. I don't suppose it's a *huge* deal, but do you think we >> > could ask upstream to apply the new LPPL to the existing codebase? >> > This doesn't require anything more than an email on their part, >> > which we could then stick in debian/copyright. >> > >> Well, our upstream is TE. Most of the code is not written by him, so >> he doesn't really have control over these things. Take e.g. >> KOMA-Script: the package is explicitely linked with LPPL 1.0. If you >> ask Markus Kohm, he'll refuse to upgrade to the next version (at >> least I read some postings about this by him in dctt). Well we could >> put 1.3 into teTeX 2.0.2 and hope, that most of the problems will be >> resolved then... > > Er, well...it doesn't really help the package's DFSG problems if the > DFSG-free version of the license isn't actually used. > > Also, is it the new version of the LPPL Markus Kohm doesn't want to upgrade > to, or the new version of some software? > > I think it was established during the long, long discussions on -legal that > previous versions of the LPPL were not DFSG-free.
Markus does not want to upgrade the license. However, this is a kind of special problem, and I am not sure which license versions he dislikes - he is using LPPL 1.0, while the current change is from 1.2 to 1.3. Might be that we can convince him to use LPPL, or maybe LPPL with some additional restrictions for the distribution of unmodified copies (namely, only with the original documentation included). On the other hand, while I personally think KOMA-Script is an important package, quantitatively Markus' contribution is only one small part. There are probably many packages with a "version X or later" clause, but also many without. There are 716 directories in our texmf tree. I estimate that at most one third of them are higher-level directories, that makes nearly 500 lowest-level directories, or lets say 400 individual packages. The average number of authors per package will probably be below 2, but you can see that even if only 20% of the packages do not have a "or later" clause, and some use GPL or Artistic license, we would need to contact dozens of individuals, many of which have new E-mail addresses now. Sorting this out and contacting all upstream authors would take a hell lot of time. Asking them to upload a new version to CTAN instead of sending us a mail, because such a mail would be problematic wrt to DFSG clause 8, would take even more time. So much time that we will not be able to accomplish this within this year. > I'm open to suggestions for how we should cope with this. I currently see three possibilities: a) release sarge without TeX packages b) find a group of about 10 people who are willing to make those upstream contacts, create a separate mailing list (debian-tetex-maint would become unusable), somehow make sure that those people are "gentle" to upstream authors. bb) finding the problematic cases and splitting them out as tetex-nonfree, as we had previously, would probably be faster, but still need additional people to work on it. c) Make an exception for sarge (and perhaps an other exception for a licenses-clarification-only upload into stable) Some comments on this: Option a) would not only be bitter for us, the maintainers, but also affect many other packages. apt-rdepends -r tetex-base | \ sed -e 's/.*Reverse Depends: //g' | \ sed -e 's/[[:space:]]*(.*)//g' | \ sort | uniq |wc -l gives 136. Some of this would have to be left out, too, some are meta-packages (e.g. education-*), some would need work and tweaking to take out the dependencies (e.g. font packages that provide their files both for TeX and X). Which wouldn't speed up sarge's release, either. Option b) is completely unrealistic. If anybody wants to convince me of the opposite, please start working on #218105 (but please wait until I've submitted my preliminary work I did this weekend). Option c): I think this could be regarded as a case for an exception, since it is mainly because of the timing of sarge's release in comparison to the license update and teTeX release that we currently have this situation, and that most authors chose the LPPL because they wanted a free license. I also think that if we left teTeX out of sarge we wouldn't do ourselves a favor, but rather offend sensibilities of the people you have so productively worked with to make the LPPL better, and free. I would be willing to raise this issue on -release, but not without some further input from the -legal people. Regards, Frank -- Frank Küster, Biozentrum der Univ. Basel Abt. Biophysikalische Chemie