establishing a code of conduct for the Mozilla community
Myk Melez
the Mozilla community.
First, a written code would make it easier for outsiders to learn how we
work and integrate themselves into the community.
Second, when someone is breaching the code, it would be easier to get
that person to stop if we could point to an authoritative document that
proscribes the behavior. Sure, some people will flame, troll, etc.
anyway, but I think more will stop when asked to do so if the request is
accompanied by a reference to a doc.
I took a look at Ubuntu's code of conduct
<http://www.ubuntu.com/community/conduct> recently, and it seems like a
decent starting point for drafting one of our own.
I do find it too wordy and too cautious in some places. Also, it's
strange that the avoidance of "flamewars, trolling, personal attacks,
and repetitive arguments" is separately enjoined as a mere "guideline"
and for mailing lists and web forums only.
But overall I think it's a pretty good statement of principles for
conduct in the Ubuntu community, and I think it would behoove us to
develop a similar statement for our own community.
Thoughts?
-myk
Mike Shaver
> Thoughts?
I think that having such guidelines would be great too, though I'm
sure we'll have to spend a month wordsmithing it. :) The guidelines
should go beyond "what is polite" and include "what is productive", to
capture some of the operating principles that we try to employ.
Perfect is the enemy, though, etc.
The "poisonous people" presentation from the Google folks has good
work to start from here too.
Mike
Myk Melez
> I wonder if the effort creating it will be an example of why it may be
> needed.
Heh, I hope not. :-/
> Yea, I'm for it, I think it's a good idea. Though I hope it doesn't get
> in the way. I think a key advantage of the Mozilla community is the
> ability to freely express opinions and not have people point to a set of
> rules.
Agreed. I'd rather not have a code at all than have one that makes us
hesitant to say what we think or looking over our shoulders afterwards.
> For example the style switcher (still not back as of 2.0 btw!),
> Mozilla Suite aka SeaMonkey being dropped, to even the formation of
> MoCo. All things that caused a fair amount of turbulence from rampant
> speculation, to those who felt "this is when the project jumps the
> digital shark/sells out/looses it's mojo/looses it's passion".
Turbulence seems fine, as does speculation and criticism (but not
personal attacks).
> But in each case, the dialog and seemingly endless babble (how much
> did Mozilla make in 200_?) seems to generally be productive and beneficial.
I can think of a number of counterexamples (mainly decisions about
Firefox changes that get rehashed endlessly without new information).
But in any case, this shouldn't be an exercise in imposing a policy
against our will. If we can't agree (or agree enough, however we can
define that) on some principles of conduct, then we shouldn't adopt them.
And we can also distinguish between stuff we don't allow and stuff we
merely discourage, if we think there are good reasons to engage in the
latter at times.
> Just my $0.02. Hope this doesn't violate the code of conduct ;-).
To the gaol with ye! ;-)
-myk
Myk Melez
> I took a look at Ubuntu's code of conduct
> <http://www.ubuntu.com/community/conduct> recently, and it seems like a
> decent starting point for drafting one of our own.
Another example: wikipedia distinguishes between guidelines and
policies. Both are considered "standard[s] that all users should
follow", but users are "expected to abide" by policies, while guidelines
"not set in stone."
They have a bunch of each:
http://en.wikipedia.org/wiki/Wikipedia:List_of_guidelines#Behavior
http://en.wikipedia.org/wiki/Wikipedia:List_of_policies#Behavioral
-myk
Gervase Markham
> I think it would be useful to establish a written code of conduct for
> the Mozilla community.
>
> First, a written code would make it easier for outsiders to learn how we
> work and integrate themselves into the community.
Have people historically had difficulty in doing this, relating to their
lack of ability to pick up community norms (as opposed to other problems
with getting integrated)?
> Second, when someone is breaching the code, it would be easier to get
> that person to stop if we could point to an authoritative document that
> proscribes the behavior. Sure, some people will flame, troll, etc.
> anyway, but I think more will stop when asked to do so if the request is
> accompanied by a reference to a doc.
Are there some recent examples of bad behaviour where you feel such a
document would have helped?
Gerv
fantasai
> Myk Melez wrote:
>> Robert Accettura wrote:
>>
>>> Yea, I'm for it, I think it's a good idea. Though I hope it doesn't
>>> get in the way. I think a key advantage of the Mozilla community is
>>> the ability to freely express opinions and not have people point to a
>>> set of rules.
>>
>> Agreed. I'd rather not have a code at all than have one that makes us
>> hesitant to say what we think or looking over our shoulders afterwards.
>
> independent thinker. Not just by MoFo employees and core contributors,
> but in the community at large. Any set of rules imposed would be
> subject to vigilante justice. I just hope they don't become misguided
> and used to justify attacks.
Gerv has written a code of conduct for Bugzilla, and I don't recall any
such problems as a result. :)
https://bugzilla.mozilla.org/page.cgi?id=etiquette.html
~fantasai
Myk Melez
> Have people historically had difficulty in doing this, relating to their
> lack of ability to pick up community norms (as opposed to other problems
> with getting integrated)?
Many haven't, but some have.
> Are there some recent examples of bad behaviour where you feel such a
> document would have helped?
I've seen a couple cases over the last year where a document might have
helped, but I don't have references at the moment. I'll try to track
some down.
-myk
Myk Melez
> Gervase Markham wrote:
>> Are there some recent examples of bad behaviour where you feel such a
>> document would have helped?
>
> I've seen a couple cases over the last year where a document might have
> helped, but I don't have references at the moment. I'll try to track
> some down.
One case is BryanS...@gmail.com's posts to mda.firefox [1]. In
response to a very long one addressing many issues, including some which
had already been debated to death, but without raising new facts [2], I
asked the author to start separate threads on each issue, be succinct,
and only reopen decided issues with significant new facts [3].
Nothing the author did should be forbidden, of course, but it would seem
useful to discourage those behaviors. Nevertheless, my attempt to do so
didn't seem to have much effect, although the author subsequently
stopped posting.
Would a written code of conduct have caused this person to contribute
more productively and perhaps stay in the community? It's hard to say,
but I think it's worth a shot.
-myk
[1]
http://groups-beta.google.com/groups/search?q=group%3Amozilla.dev.apps.firefox+author%3Abryansmiley78%40gmail.com
[2]
http://groups-beta.google.com/group/mozilla.dev.apps.firefox/browse_frm/thread/6a53e16ed307f8f2/151140d91acc4f4f?lnk=st&q=group%3Amozilla.dev.apps.firefox+author%3Abryansmiley78%40gmail.com
[3]
http://groups-beta.google.com/group/mozilla.dev.apps.firefox/browse_frm/thread/6a53e16ed307f8f2/151140d91acc4f4f?lnk=st&q=group%3Amozilla.dev.apps.firefox+author%3Abryansmiley78%40gmail.com
Zak Greant
On Dec 19, 2006, at 14:32PST (CA), Myk Melez wrote:
> Myk Melez wrote:
>> Gervase Markham wrote:
>>> Are there some recent examples of bad behaviour where you feel
>>> such a document would have helped?
>> I've seen a couple cases over the last year where a document might
>> have helped, but I don't have references at the moment. I'll try
>> to track some down.
>
> One case is BryanS...@gmail.com's posts to mda.firefox [1]. In
> response to a very long one addressing many issues, including some
> which had already been debated to death, but without raising new
> facts [2], I asked the author to start separate threads on each
> issue, be succinct, and only reopen decided issues with significant
> new facts [3].
>
> Nothing the author did should be forbidden, of course, but it would
> seem useful to discourage those behaviors. Nevertheless, my
> attempt to do so didn't seem to have much effect, although the
> author subsequently stopped posting.
>
> Would a written code of conduct have caused this person to
> contribute more productively and perhaps stay in the community?
> It's hard to say, but I think it's worth a shot.
One comment here is that the code is as much for "us", even more that
it is for "them". Defining what we expect from ourselves can help us
communicate more effectively as a group.
Cheers!
--zak
Jonas Sicking
One example I still remember vividly is bug 340318. There's plenty of
bad mojo in there, read comment 33 and on. Also comment 87 shows a lack
of understanding of our development process.
/ Jonas
Peter Weilbacher
> Myk Melez wrote:
>> I've seen a couple cases over the last year where a document might have
>> helped, but I don't have references at the moment. I'll try to track
>> some down.
>
> One example I still remember vividly is bug 340318. There's plenty of
> bad mojo in there, read comment 33 and on. Also comment 87 shows a lack
> of understanding of our development process.
Reading that bug without having been involved, I guess that a document
such as Myk is suggesting would not have helped the people who
misbehaved in that bug. Or do you think he would have stopped screaming
to read a document?
I think that's a general problem in many communities: documentation
helps, but only for people who want to read it...
Peter.
Mike Beltzner
> Reading that bug without having been involved, I guess that a document
> such as Myk is suggesting would not have helped the people who
> misbehaved in that bug. Or do you think he would have stopped
> screaming
> to read a document?
True, codes of conduct will not stop people who want to behave badly.
What they do is make it easier to recognize bad behaviour and avoid
wasting time addressing it. In the example Myk cited, a lot of
discussion surrounding the bad behaviour could have been avoided
(saving time and energy) and instead someone could merely have
pointed out: "Look, you're not following our code of conduct here,
and until you do, we can't really begin to discuss your issue."
Basically, it helps to define what is a "troll" for a "do not feed
the troll" policy.
cheers,
mike
Mike Shaver
> Nothing the author did should be forbidden, of course, but it would seem
> useful to discourage those behaviors.
I think it's probably more useful to talk about the behaviours we want
to _encourage_; enumerating badness is a bad idea in most fields, and
makes us more likely to rathole on edge cases in which something is
permitted.
(And even though there will continue to be cases in which people don't
read or care about the standards, hard cases make bad law, and we
should strive to be telling people what we want them to do, rather
than playing whack-a-mole with bad behaviours as people bounce from
one to another.)
Mike
Jonas Sicking
I think it could at least have helped to have a document to point the
screaming guy at. As things stood there was nothing we could do except
to single him out telling him he was a jerk, which clearly didn't help.
I think that if we had had a document describing unacceptable behavior
then hopefully the discussion could have been done in a calmer and more
constructive manner.
Though it's quite possible that another type of document would have been
more helpful in this situation. Something describing how we develop and
what help is needed from contributors, which is something different from
the code of conduct.
/ Jonas