On Fri, Mar 03, 2006 at 10:20:59PM +0200, Lars Wirzenius wrote:
> A code of conduct has often been mentioned as a possible solution to
> various communication problems we have. The code would have to specify,
> either explicitly or implicitly, some rules for acceptable behavior.
>
> What do you think of a code of conduct?
So, obviously Steve and I tried this last year, by first making up a
code of conduct, creating a new IRC channel that that would apply to,
finding a variety of moderators for that channel, and then trying to
have discussions on it.
(Err, Steve Langasek and I created it, Steve McIntyre's also an op)
There are a couple of things that are worth drawing from that:
* a code of conduct isn't enough; anyone can draw one up, you need
to have confidence that everyone else in the room is going to
abide buy it too before you're going to be interested
* there needn't be one code of conduct for the whole project --
-tech's code of conduct doesn't apply to anyone not on the channel,
nor for the most part does it apply to what anyone says outside
the channel.
There _are_ some areas of the project that are "special" and so that
doesn't apply, such as -devel-announce, which every developer is
expected to read, and -vote, which is the list where discussions
about GRs are expected to take place -- but for pretty much
everything else it's possible to have separate lists/channels/etc
with different rules that different people might like, or the same
people might like at different times.
> What in your opinion would be a lower limit on acceptable behavior?
Focussing too much on lower limits probably tends to encourage people
to sink to them; I'd rather say what I'd like to see, and that's really
intelligent people proposing good ideas for the betterment of Debian
and free software and analysing them using the highest scientific,
engineering or mathematical standards. I love the fact that we've got
a world class system of deciding votes eg; I think that should be the
level we set for every discussion of any importance.
> Do you think that strict rules would
> be better than general guidelines? Who should be the judge if a
> particular case follows the code of conduct or not?
I have no idea; I think the best idea is to try different approaches
and see what works best. So far we've tried "Have maintainers/delegates
be communicative and polite at all times, and the flaming will stop on
its own", during Martin's term as DPL, which I personally think was an
abject failure, and "Setup a new channel with a charter that gets actively
enforced, albeit mostly only by aj", which I don't think's worked either.
> Would the code be a
> good thing, or would it necessarily be a threat to freedom of speech,
> and stifle innovation?
There's a saying that "the net routes around censorship", which I think's
pretty true, and in particular, I think means that no matter what happens
freedom of speech is perfectly safe -- if people find debian-devel
too obnoxious to use, they can always setup somewhere else to talk,
eg. Likewise for pretty much everything except the -announce lists and
-vote -- but even then, they're both already augmented by people posting
to Planet.
> Should any kind of behavior be allowed on Debian mailing lists?
Some behaviours are mutually exclusive; so I don't think it's possible
in practice to allow any kind of behaviour on Debian mailing lists --
it's just a matter of choosing whether we're going to decide what we
want ourselves, or put up with whatever we end up getting.
On Fri, Mar 03, 2006 at 10:08:15PM +0100, Frank K?ster wrote:
> And, do you think this code of conduct should be enforced? How?
We already have a code of conduct for the lists that's not enforced;
it says you can't swear, and that you shouldn't flame. And it's widely
ignored, both because people don't think those expectations are either
realistic or important, and because people who violate it not only don't
get any rebuke over it, but get support for the violation.
Others [0] have already said that the mere existance of a code of conduct
is excessively unpleasant, and, if I understand correctly, that they
find /any/ sort of active enforcement as vile as I find having people
be told they have "personality problems" just for participating in the
project. I don't see a resolution to that -- you either have enforcement
without any up front rules which doesn't seem terribly fair, or you have
a code of conduct that offends people, or you let people be as vicious
as they please, and discourage people who don't want to put up with that
from being actively involved.
Personally, I don't think code of conducts are a big deal -- if you
disagree with them, you suggest changes; if you don't think the people
trying to enforce them are doing a good job, you help them do a better
one; and if you find you need somewhere to scream and yell at the stupid
things someone does, you find somewhere else to do it -- which usually
means not doing it to their face. So I tend to think people shouldn't
be scared of rules in the first place; but obviously some people are,
and that has to be dealt with.
I've tried what I could think of though, I think it's other people's
turn to come up with ideas and see if they work. I really hope that that
will happen by building on previous ideas, not just repeating them or
ignoring them though.
Cheers,
aj
[0] http://raw-output.org/20060228/selfbanning
http://www.grep.be/blog/2006/02/27/#debian-tech_2nd_retort
http://www.grep.be/blog/2005/09/22/#playing_police_will_not_work
Attachment:
signature.asc
Description: Digital signature