Bug#688772: gnome Depends network-manager-gnome
Sam Hartman writes ("Bug#688772: gnome Depends network-manager-gnome"):
> I'm disappointed to immediately see this discussion turn to assumptions
> of malice and reprimands.
Josselin described the TC decision as a "crusade" in his upload.
I think that is sufficient for me to infer malice.
> One is that the gnome-meta maintainer is trying to meet the letter of
> your intent while trying to work around it.
Right...
> Would you be willing to set that aside for a moment and think about
> other possibilities.
OK, fair enough ...
> Another possibility is that the gnome-meta folks have been confused and
> frustrated by this whole discussion. They don't see what the big deal
> about n-m is and they want to provide a good experience for the users.
> They received a decision they don't really like from the TC with some
> complex rationale and so they tried to follow through that rationale
> and balance their goals against the rationale the TC stated as best they
> could.
I don't think you could read the rationale that way. At least, not
without a lot of wilful blindness. It clearly explains that the
biggest part of the problem was that users who had gnome but not n-m
in squeeze would get n-m when upgrading to wheezy:
3. [...] users who have gnome or gnome-core installed but have
removed or never installed network-manager will have
network-manager installed during an upgrade from squeeze.
4. [ description of why network-manager can be problematic ]
5. The Technical Committee believes that this will cause undesireable
behavior for upgrades from squeeze, and (of somewhat lesser
importance) will make it more difficult than necessary for GNOME users
to swap network management components, something for which there
appears to be noticable demand. We therefore believe that
network-manager should be moved to Recommends in gnome-core.
> I think a reasonable person could read that section of the resolution
> and conclude that if n-m were pushed into more inclusive meta-packages,
> then the argument might be different.
As I write I don't understand why, if this idea was thought to be a
good one, the maintainers didn't suggest this approach to the TC.
It's not as if there wasn't time, and it's hardly a difficult thing to
think of.
The real answer is of course that if anyone had suggested this as a
possibility our resolution would have unambiguously ruled it out. So
the purpose of subverting the decision was only served by producing
this change afterwards.
> Now, I'll admit that there was probably some searching going on for how
> to fit some goals into what the TC proposed. I'll admit that there
> might have been some ask for forgiveness not permission going on.
> But all those things are normal with frustration.
What kind of emotional state do you think I should have after the
legitimate and unanimous authority of the TC has been undermined in
this way ? Perhaps I would be frustrated.
> Would you be willing to consider
> 1) focusing on accomplishing the specific immediate goal you
> want--perhaps points 1-6 in your proposed resolution.
To be honest I don't expect to be able to get a 4:1 majority in the TC
in favour of publicly denouncing Josselin. (Since the resolution as a
whole overrules a maintainer, that would be necessary.) But I wanted
to put my feelings, which IMO are legitimate, on the record.
> And then later having a serious discussion about how you and the TC can
> write resolutions that are more likely to achieve the long-term goals
> of the TC while avoiding frustration.
Normally we write our resolutions with the intent that people will not
subvert them, or work against their intent. That is after all
required by the Constitution.
If we need to make them watertight against malicious and lawyerish
interpretation, then we will need to anticipate every way in which the
maintainers might try to subvert our intent. Along the lines of point
6 in my proposal.
I think it would be very rude to routinely write that kind of thing in
overruling resolutions. It it amounts to assuming and anticipating
bad faith on the part of the overruled maintainer.
> I would be happy to contribute some thoughts there if desired.
Please do.
Ian.
Reply to: