Re: When should -legal contact maintainers [Was: Re: Question for candidate Robinson]

2005-03-10 Thread Sven Luther
On Thu, Mar 10, 2005 at 12:23:26AM -0800, Don Armstrong wrote:
 [This is wildly OT for -vote, MFT set to -legal and CC:'ed, please
 follow up there or privately.]
 
 On Thu, 10 Mar 2005, Wouter Verhelst wrote:
   On Thu, Mar 10, 2005 at 12:52:20AM +0100, Sven Luther wrote:
Still, debian-legal should inform the maintainers and invite them to 
take
part of the discussion when examining packages which have been in main 
for
years.
 
  I think he's right about this. For one thing, as he just explained,
  he got upset precisely because he wasn't informed; it's reasonable
  to assume that the way in which his discussion would have been
  performed would have been 'slightly' different had he been informed
  in time.  I *do* think it is good practice for d-legal contributors
  to inform a packages' maintainer if they are discussing its license;
  we do the same with other types of bugs.
 
 If -legal is specifically discussing a license of a package, the
 maintainer is generally informed[1] when the discussion is actually

Can be, and if so it is nice, but it was not in this case, since the first
mention i had was that consensus was reached and my package should move to
non-free. And it was a nominal discussion about my package. And again, the
mail saying the above was CCed to debian-legal, and nobody there bothered to
correct the misconception.

 happening. However, (almost) no one bothers to inform the maintainers
 when general discussion of a license is occuring, in the first part
 because most of the discussion isn't particularly useful to most
 maintainers, and secondly, because people have better things to do[1]
 than track down which packages are covered by a license when the
 critical issues (if any) haven't been discussed or discerned yet.
 
 In the latter stages of the discussion, if there really are issues
 with a license that packages in Debian are using, bugs are typically
 opened against the packages, ideally with a short summary of the
 specific issues that the license has, and suggestions for what the
 maintainer can do to fix the license. (And quite often offers of help
 in explaining the problems to upstream as well.)

And in this case, suggestion was ask upstream to GPL his software or dual
licence, as trolltech did for Qt. not even bothering to examine the package in
questionand noticing that none of the QPLed part of the package was indeed a
library, and thus had no GPL-interaction problems.

This shot first ask later attitude based on half informed guesses and backed
by the fanatism of the debian-legal posters was what mostly irritated me back
then, and also what makes me believe that debian-legal is not to be thrusthed
on licencing issues, which makes ti totally useless.

 As far as the analogy to normal bugs goes, the preliminary
 discussion is generally on the order of is this really a bug? as is
 typically seen on -devel. [Or, in the extreme case, figuring out
 whether mass bug filing is sane.] Surely no maintainer expects to be
 notified every time someone asks on -user, -devel (or $DEITY forbid,
 IRC[3]) whether specific behavior from a package constitutes a bug.

no, but maintainers get over-angry when people modify the seveirty of one of
their bugs they have been ignoring for age, no ? And this reaction seems to be
backed up by the powers that are, and a real analogy to the please ask
upstream to GPL his software or we will recomend ftp-masters to remove it from
main kind of request.

Friendly,

Sven Luther


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Re: When should -legal contact maintainers [Was: Re: Question for candidate Robinson]

2005-03-10 Thread Don Armstrong
On Thu, 10 Mar 2005, Sven Luther wrote:
 On Thu, Mar 10, 2005 at 12:23:26AM -0800, Don Armstrong wrote:
  If -legal is specifically discussing a license of a package, the
  maintainer is generally informed[1]
 
 it was not in this case, since the first mention i had was that
 consensus was reached and my package should move to non-free.

In this particular case, the package and license combination that
brought up the QPL was libcwd (#251983).[1] To be honest, no one seems
to have equated the libcwd discussion about QPL being non-free with
the ocaml discussion about the QPL being GPL incompatible until Brian
Sniffen brought it up,[2] and since you're in the Maintainer: field on
ocaml, you were notified. [This isn't particularly surprising as it's
almost impossible to figure out what licenses packages are under in
Debian in an automated fashion.]

  In the latter stages of the discussion, if there really are issues
  with a license that packages in Debian are using, bugs are
  typically opened against the packages, ideally with a short
  summary of the specific issues that the license has, and
  suggestions for what the maintainer can do to fix the license.
  (And quite often offers of help in explaining the problems to
  upstream as well.)
 
 And in this case, suggestion was ask upstream to GPL his software or
 dual licence, as trolltech did for Qt. not even bothering to examine
 the package in questionand noticing that none of the QPLed part of
 the package was indeed a library, and thus had no GPL-interaction
 problems.

Dual licensing under the QPL and GPL (or as actually suggested, QPL +
LGPL[3]) would have solved both the DFSG freedom issues with the QPL,
and the ocaml emacs binding issues of #227159. It may not be the
optimal solution for ocaml, but it would have solved the immediate
problems.

  Surely no maintainer expects to be notified every time someone
  asks on -user, -devel (or $DEITY forbid, IRC[3]) whether specific
  behavior from a package constitutes a bug.
 
 no, but maintainers get over-angry when people modify the seveirty
 of one of their bugs they have been ignoring for age, no ?

I'd hope that maintainers wouldn't get angry,[4] and instead be
willing to help discuss the issues (or lack thereof) that make the
changed serverity of the bug reasonable or unreasonable. After all,
it's not like we're making up these issues purely to spite
maintainers. In most cases, reasonable people have examined the
issues, discussed them, and felt there was enough of a problem to
warrant bothering a package maintainer about it.

After all, things change, and a bug that was normal severity today may
end up being RC tomorrow.

 And this reaction seems to be backed up by the powers that are, and
 a real analogy to the please ask upstream to GPL his software or we
 will recomend ftp-masters to remove it from main kind of request.

I'm afraid I cannot parse what you're trying to say here.


Don Armstrong

1: http://people.debian.org/~terpstra/message/20040709.215918.1224a82f.en.html
2: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=227159msg=65
3: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=227159msg=41
4: But then, bts ping-pong doesn't happen because maintainers are
always calm...
-- 
What I can't stand is the feeling that my brain is leaving me for 
someone more interesting.

http://www.donarmstrong.com  http://rzlab.ucr.edu


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]