On Mon, Dec 06, 2010 at 09:12:26PM -0700, Robyn Bergeron wrote:
Most of the cited "need" for a code of conduct revolves
list conduct, and IRC conduct (mostly the former). We cannot prevent
anyone from joining a mailing list, or IRC, over and over and over, using
different addresses, coming from different IP addresses, etc. If someone
is bound and determined to be a thorn in our sides, well, they're going to
The code isn't expected to deal with people who are deliberately
trolling, and it's obviously impossible to deal with someone who's
determined to make trouble for us. But that's true regardless of whether
or not we have a code of conduct.
If we are setting up community behavioral expectations and
people need to sign their names to, they expect them to be executed upon.
If it's something that is, in the end, unenforceable, it reflects poorly
upon everyone - particularly the Board, who would likely wind up losing
credibility in anything else they say in the near term. ("Well, you told
us you could get rid of poisonous people, why should we believe you now?")
At the point where someone is repeatedly signing up under false names or
using other technical means to subvert our policy, it's clear that
they're not part of our community. There are limits to what we can do -
such behaviour won't be tolerated (in the sense that the moment we
observe it appropriate action will be taken to remove the individual via
whatever technical means necessary) and demonstrating that intolerance
sends a signal even if the target then reappears from another address.
But do you really think this is a likely outcome? I haven't noticed any
significant attacks of this nature on our community (other than the
occasional crapflood on IRC), and I don't see why introducing a code of
conduct would make it any more prevelant.
My other main gripe with Codes of Conduct is that they, in my
tend to stifle discussion, because people tend to want to not be
confrontational, because they don't want to get the ban-hammer. I think
this tends to just dull rapid innovation, and that's not something we
should be doing (again, IMO).
When I see confrontational threads on devel-list, I see threads that
aren't getting anything done. There's no problem in engaging in
I *really* think that before we go crafting things, or talking to
in various groups, we should do the following:
1) Determine if any policy we create around a "worst-case scenario" could
actually be enforced. What types of punishment are we dealing with? If it
can't be enforced, 100% of the time, we are just digging ourselves a
deeper hole of problems - are people being favored? Are people not doing
their jobs? Etc.
I think it can be enforced in 100% of the cases it's designed to cover.
Deliberate trolling clearly falls outside its remit.
2) If we determine that a "worst-case scenario" could
actually be dealt
with and enforced - who is going to do it, and ARE THEY ACTUALLY GOING
TO. We've had a long-standing policy of "be excellent" - with the
more or less responsible for "dealing" with situations - and I am not sure
that any of the situations that could have been dealt with *actually* had
If we agree on the need for a code then we need to discuss enforcement.
I don't see the benefit of worrying about our enforcement strategy
unless we decide that a code is beneficial.
Frankly, I'm not sure that we've really *had* any
situations that were so
dire that they required any sort of serious intervention. And that makes
me wonder if we really need one at all.
I regularly see situations in our community that are outside the realm
of reasonable behaviour, and I think this reduces our ability to market
ourselves as a welcoming community. We're certainly not as bad as
Debian, but we're nowhere near as well-behaved as the core of Ubuntu.
Matthew Garrett | mjg59(a)srcf.ucam.org