Re: Bug#227159: ocaml: Worse, the QPL is not DFSG-free
On Mon, Jul 19, 2004 at 11:58:11PM +0100, MJ Ray wrote:
> [Sven wants cc]
>
> On 2004-07-19 17:34:12 +0100 Sven Luther <sven.luther@wanadoo.fr>
> wrote:
>
> >On Mon, Jul 19, 2004 at 04:50:26PM +0100, MJ Ray wrote:
> >>You have mentioned this to me before, but I did not find it in the
> >>list
> >>archive. Do you have a more specific reference, please?
> >No sorry, searching for me will cost the same as searching for you.
>
> You remember the conversation, which I do not. It should be easier for
> you to locate it, if it is there. I don't think it's in the list that
> you think it is, but I don't know where it is.
Sorry, but it was three years ago or more, and i did do a search on rms in my
own ocaml mail archive, and didn't find any. It was about the LGPL, and static
linking of the runtime lib, ...
Mmm, i believe the thread was titled :
Re: [Caml-list] License Conditions for OCaml
And dated from :
Date: Tue, 04 Dec 2001 13:55:54 -0500
Thread starting here probably :
http://caml.inria.fr/archives/200112/msg00000.html
Not directly about the QPL though, altough i mention the exact licencing stand
of ocaml, including the QPLed compiler suite, in the response to the above
message.
> >>they are not a corporation does not prove that they are not evil. (I
> >>am not
> >>saying that they are, but I find your argument offensive.)
> >Well, what about [...]
>
> Fine to me, but that wasn't what you wrote.
>
> >Well, i have the feeling than many on debian-legal overindulge in
> >sterile
> >discussion, maybe not out of pleasure, but maybe some morbid
> >fascination. This
> >is an external view, but still a feeling i get when reading this many
> >hundred
> >long thread, which does't really broaches, or rarely, the main points.
>
> I think the extreme length of this thread is partly due to an Offended
> Maintainer sending messages three at a time restating the same point
> over and oevr that people haven't had time to answer yet.
Well, i was told that i should use repetition in order to get attention, was i
not ? And the thread hardly needed me to become lengthy, given the 2+ hours i
needed to go through it before i even started posting.
> >>reason that I lobby politicians and work on other not-for-profit
> >>projects:
> >>ultimately, if I don't, I will not be able to continue doing what I
> >>enjoy
> >>doing.
> >Well, does that mean that i should start lobbying debian-legal on the
> >fear
> >that i will no more be able to do my packaging work ?
>
> No.
He :).
> >This is also the reason
> >i lost more than two hours reading this often ridicoulous thread.
> >Imagine how
> >many bugs i could have fixed in the meantime ?
>
> I don't know. I could have done lots more directly useful stuff too,
> but this has a role too.
Well, maybe, but i have some doubts. Others have claimed some doubts about the
real representiveness of debian-legal, and its real decision power.
But sure, which is why i followed up on it, that and i received request for
status report from at least two packagers of ocaml related stuff, expressing
fear that we will have to remove ocaml without prior notice from main, so
close near the sarge release, thus voiding all our work of these past two
years.
> >>>Also, one of the clauses you have problems with, the "court of
> >>>venue",
> >>>if waived, might limit their possibilities to defend against people
> >>>>not
> >>>respecting the licence [...]
> >>This language is confused. They do not need to defend against such
> >Suppose someone is using some GPLed code in violation of the GPL, you
> >as
> >author of it, have to defend yourslef against this violation [...]
>
> Surely that is a prosecution, not a defence?
Well, maybe, but in this case, the crime is the violation of the licence, and
you defend against said crime. It feels a bit sad that debian-legal has more
the interest of those would-be licence violators in mind that the code
contributing upstream though. Which doesn't speak a great deal about the
respect we have for upstream, without whom debian would be nothing, and seems
to indicate that we take the gift of their upstream source code ot the
community for granted.
> >Well, yes, but in that case, a lone developer without extensive
> >monetary
> >ressource, has no way to sue a potential violator of the licence,
> >thus making
> >said licence totally void. You have to strike a balance here, and not
> >see only
> >the potential licence breaker.
>
> "Potential licence breaker"? You mean the licensee? That's not a good
> attitude towards users, "let's assume they're all potential
> lawbreakers".
Well, tell me what reason there would be for a law suit if there was no
licence violation ? and let me tell you that i seriously doubt that a french
court, under french law, would be amene to filling bogus law suite intented to
harass innocent third parties, but then i hear this is not the case in the US.
And what about our attitude toward our upstream authors ? Do you think it is
not important ? And that it is our due that they give away their code ?
And again, if i were to release a GPLed work, and some US corporation decide
to include, in violation of the GPL, code of it in their own proprietary work,
what chance i have to get them to stop it ? Really ? Most probably i would not
know, and if i knew, i would just drop it, since i have no chance to ever win
such an action, and stand to loose more money than i own in it anyway.
> It is entirely proper that we worry about the licensee, because they
> expect us to. We are worried about the licensor using choice of venue
> as a termination clause, by starting an unprovable complaint that is
> decided against a poor licensee, simply because they cannot appear in
Sure, again, i understand this in a lawsuit-country like the US, but not all
the world is so, it is even a matter known for ironic (and maybe often
disgusted) joking in the rest of the world.
> the stated court. Now, if you are sure that France does not allow the
No i am not, i am no lawyer after all, but this seems reasonable, and our
judges have enough work as is without giving in to bullshit lawsuits.
> plaintiff's venue to hear this case, then this wouldn't cause ocaml to
> be non-free, but then why cause confusion by having this clause there?
Well, probably because it was copied verbatim from the QPL, because upstream
are academic researchers, no evil company with a legal batallion, and they
have a need to be reasured over not getting troubles they don't want to have
anything to do with, which include troubles like what we are having here. And
i strongly agree with them, and resent the fact that i can't place my trust in
the debian-legal team to provide reasonable interpretations. And as said,
Brian seems to have his own interest in this, since he claimed to be wanting
to release its own proprietary modification of ocaml without giving them back
to upstream.
> It means changes in French law could make ocaml non-free, which sucks.
Well, the same would go about US patent law and other stuff ? And was i
consulted when it was decided to report about my every debian activity to the
US secret services ?
> I am aware that the questions of termination clauses are being
> discussed in another debian-legal thread.
Err, one which i maybe not follow, please provide a pointer, as i did for you
earlier. And i don't understand what the termination clause has to do with
this.
> >>You cannot have it both ways! Informing packagers early in the
> >>discussion
> >>brings accusations that we are wasting their time;
> >Well, ok, but at least they should be allowed a time to voice their
> >opinion,
> >and in time for it to be considered also.
>
> I think that is a good idea. Would a d-d-a post when a summary reaches
> consensus on the -legal list help?
Well, this would be too late wouldn't it, since consensus would be reached,
and the ftp-masters/RM or whoever else would be in charge would be strongly
counseled by debian-legal to remove the packages from main without prior
notice. Same as your politician which inform you that the law has been voted,
nice to know, but too late to do anything about it.
> >I have some loss of faith in
> >debian-legal to reach reasonable conclusions though, and as thus i
> >would like
> >to be informed if there is some likeliness of disaster happening.
>
> Subscribe.
Sure, would you pay me for the time spent reading it ? Already i follow too
many mailing lists, and as i am not lawyer inclined, the real info/noise ratio
is too low to be worth considering.
> >>informing them after the issues have been discussed a bit brings
> >>accusations that we "didn't even bother to inform" them. As I'm sure
> >>you've
> >>noticed, there's nothing like consensus on this yet. It's only
> >Well, yes, but there is Brian, filling bug reports claiming there is
> >consensus.
>
> Brian has apologised to my satisfaction, and I have apologised to him
> for my erroneous allegation. I am sorry you seem to have have rejected
> his apology and are still basing your view on my mistaken comment.
Well, i have not read it maybe ? There were how many mails in this threads? I
thought i had read them all, but i may have missed the one or the other.
> >>recently that there is a concise summary of the interesting
> >>questions,
> >>although I think its title overstated things a bit.
> >Yeah, but as a debian-legal outsider, it is scary even to think of it.
>
> Don't fear the wardrobe monsters.
Well, as i stand to lose, i have reason to fear. And as said, i have gotten
bad treatment over my -legal participation in the past, so ...
> >>If you have a strong opinion on how it *should* be done, then say
> >>that,
> >>instead of just destructive criticism that you think it shouldn't be
> >>this
> >>way.
> >Well, first you mussle Brian and don't let him fill upsetting bug
> >reports
> >while falsifying the state of consensus of debian-legal :).
>
> Do you know constructive criticism? Hello? Anyone in?
And, did i not offer constructive critism of the summary and nobody replied ?
Also, i didn't find any of the mails Brian sent to me constuctive, not a one
of them, except maybe one of the latest i read today, and i responded
positively to it. I can count on the fingers of one hand the positive people
posting about this. You and Michael Poole figuring prominently among them, and
maybe some others i don't remember right now.
> >This has not only
> >brought fear into me, but also into many of the small team of ocaml
> >related
> >packages, which would also be affected by this decision.
>
> Again, NOT HELPFUL.
Well, sure, how can we help ? There are two issues involved, and i disagreed
with them both, mentioned reasons why i did, and even mentioned points i find
reasonable in the objection to the QPL (namely clarification of who is paying
the data transfer fee in 6c and the lack of time limit in them), but most of
the responses i got where full of bullshit. So who is not being helpfull.
If you want to start a real discussion, there are two points that deserve
interest :
1) Is it reasonable or not for debian to accept a licence which forces
modifications to be sent upstream. Some claim not, but i am doubtfull if
this is the right way for debian to go. After all, it only allows people to
keep the modifications in their own corner, and free software in general
looses (same as it looses if you release stuff under the BSD and not the
GPL).
2) does a choice of venue really is non-free ? Those who have felt not so
are mostly used to US courts, and their trigger happy lawsuits, and fear
some evil corporation suing them for no reason. They have no respect
whatsoever for upstream, and their right to have the licence respected,
which is something i would be ashamed of if it where trully debian's
position, and would not strengthen our position with our upstream.
And finally are these points really worth fighting about ? Does it really
advance our cause to figth over them, i don't think so, and maybe we should
reserve our forces, and legitimacy with upstream for real problems, instead of
engaging in petty fights over less important stuff, where our legitimacy is
dubious to start with.
No, you see, if i am speaken reasonably to, i know how to respond in kind.
Friendly,
Sven Luther
Reply to: