Re: Dealing with ITS abuse
Chris Knadle <Chris.Knadle@coredump.us> writes:
> On Friday, April 12, 2013 13:52:42, Russ Allbery wrote:
>> Chris Knadle <Chris.Knadle@coredump.us> writes:
>>> Emailing anyone privately leads down the path of "privatization".
>>> [I've already been down this road.] As such I think it might be
>>> better to publicly CC leadership, to invite public comment rather than
>>> private conversation, because private conversation cannot address the
>>> public problem.
>> I think both of you have a very strange understanding of how human
>> psychology works if you think public callouts are the best first step
>> in dealing with inappropriate behavior. I also wonder what places
>> you've worked in and what sorts of management interactions you've had
>> if you don't believe private conversation can ever address public
>> problems.
> Are you saying that if someone communicates abusively in the BTS publicly,
> they _shouldn't_ be publicly confronted about that at all?
No, I'm saying that's often not a productive place to *start*, and hence I
completely disagree with this critique of "privatization." If private
communication doesn't work, some sort of public confrontation may be a
next step (in fact, it's possibly inevitable), but it's probably not a
great place to start.
The goal isn't to punish people; the goal is to convince people that their
behavior could be improved and that it would be better for everyone
involved (including them!) if it did. Punishment is more of a last resort
when we can't manage to find any other social solution. Obviously,
getting the abusive behavior to stop is also a goal, but taking a bit of
time to try to convince people to do that voluntarily is something that I
think is worth it.
People generally get much more defensive and therefore less to change
anything when called out in public, which is why pretty much every
grievance procedure that I'm aware of encourages people to start
privately. (Please note: that does NOT mean that public complaints are
invalid, or that people should be attacked for not choosing to do that.
The onus is still on the person being abusive to change their behavior.
But as long as the situation isn't dire, the chances of a positive outcome
for everyone are higher, usually, if one starts privately.)
> Two particular bug reports I was invovled in recently had repeated
> abusive communication in them with no consequences that I could see for
> the one communicating abusively. Private communication was used to try
> to deal with that, and did not stop the abusive communication.
That's clearly a problem, and hopefully further action was then taken, but
I think it's a rather sweeping conclusion to draw that therefore private
communication is useless because in two anecdotal cases for you it didn't
help.
> At the moment I think the above is more relevant than my prior or
> current places of employment, but I'm willing discuss that if that's
> more relevant than what happens within Debian.
No, no, my point there is just that we're not doing something novel and
different here. Humanity has been dealing with social conflicts for quite
a while, and there is a lot of established understanding of what tends to
work and what tends not to work. If one is advocating an approach in
Debian that one would never follow at a place of employment, I think that
should at least call into question what we might be missing.
> Okay. Forgive my ignorance -- I'm not able to find definitive
> information about how this is dealt with in Debian. [Is there an
> "Employee Handbook" for Debian?] Up to now the only penalty discussed
> was expulsion AFAIK.
Most of the sanctions Debian can take are various forms of temporary or
permanent revocation of privileges. When it comes to abusive discussion
in public places, suspending someone's ability to use that place of
discussion is an obvious possibility. If the problem is related to
maintenance of a specific package, the Technical Committee can change the
maintainership of the package (although I don't recall if that's been
done).
Most of the first rounds of intervention involve varying amounts of social
pressure, with people like the DPL or other team members of affected teams
taking the person aside privately and saying "look, no, this isn't okay."
--
Russ Allbery (rra@debian.org) <http://www.eyrie.org/~eagle/>
Reply to: