Question that popped into my head that lets me also test out HK is what happens if someone gets a bunch of downvotes? For instance, if someone is consistently getting their emails downvoted and they hit some ratio of good to bad, will they be automatically muted/blocked/kicked off the list? Or is there a way for an admin to know someone has hit a rather negative ratio? And is there some way HK tries to prevent vote spam? Basically I'm wondering if there's an easy way to deal with bad actors beyond manual reporting that someone is viewed as such and admins needing to make a call without some objective data to back up their decision.
Mailman2 archives to 2016/07: https://mail.python.org/pipermail/overload-sig/." <overload-sig.python.org> List-Help: <mailto:overload-sig-request@python.org?subject=help> List-Post: <mailto:overload-sig@python.org> List-Subscribe: <mailto:overload-sig-join@python.org> List-Unsubscribe: <mailto:overload-sig-leave@python.org> Brett Cannon writes:
Basically I'm wondering if there's an easy way to deal with bad actors beyond manual reporting that someone is viewed as such and admins needing to make a call without some objective data to back up their decision.
Not at present. At least the last time I looked those stats are kept on messages, and there's a facility to get all the votes on a thread, but not on a poster. The data is there, though I don't know how costly it would be to retrieve it. There is primitive control on vote-stuffing (one vote per user -- not per email -- per message). I don't understand how this is in scope of this SIG, though. Although the moderators may have a need that the software we adopt can serve, I don't think the problems that have prompted formation of this SIG are related to identifiable "bad actors". We have loose cannons like me, with too much time to post and too little time to do real work. I bet if you look back at the threads that caused Antoine and Christian to blow their tops in public, you'll find them full of good citizens like Terry Reedy, Steven d'Aprano, and Paul Moore. We see the occasional over-the-top phrasing as from Sturla Molden earlier this week. None of the cases mentioned above result in "admins needing to make a call" based on a history of misbehavior -- in fact, with the names named it's quite the opposite. So mostly it's a one-time thing, and just a word to the wise from any of the old hands will do.[1] But put many such incidents together, with the current volume of interesting and relevant posts, and core devs have a problem keeping up. But it's not a problem you can put your thumb on and pin down. On the other hand, as Guido mentioned in the past, a lot of +1s might be a good signal that somebody can be promoted to "moderator" status. Footnotes: [1] Probably you'd get pushback applying that to the security debate, especially in view of Christian's intemperate "children should be seen and not heard" comment, but even there I think most of us would admit we don't know enough about security to post be confident that our contribution is net-positive.
Mailman2 archives to 2016/07: https://mail.python.org/pipermail/overload-sig/." <overload-sig.python.org> List-Help: <mailto:overload-sig-request@python.org?subject=help> List-Post: <mailto:overload-sig@python.org> List-Subscribe: <mailto:overload-sig-join@python.org> List-Unsubscribe: <mailto:overload-sig-leave@python.org>
Mailman2 archives to 2016/07: https://mail.python.org/pipermail/overload-sig/." <overload-sig.python.org> List-Help: <mailto:overload-sig-request@python.org?subject=help> List-Post: <mailto:overload-sig@python.org> List-Subscribe: <mailto:overload-sig-join@python.org> List-Unsubscribe: <mailto:overload-sig-leave@python.org>
Brett Cannon writes:
Not at present. At least the last time I looked those stats are kept on messages, and there's a facility to get all the votes on a thread, but not on a poster. The data is there, though I don't know how costly it would be to retrieve it. There is primitive control on vote-stuffing (one vote per user -- not per email -- per message).
I don't understand how this is in scope of this SIG, though. Although the moderators may have a need that the software we adopt can serve, I don't think the problems that have prompted formation of this SIG are related to identifiable "bad actors".
We're talking about replacing MM2 w/ some solution, and so that means changing the situation for admins as well. And as an admin, if I had a way to have bad actors brought to my attention or make it easier for newcomers to notice that they should generally ignore certain posters because they rarely post useful stuff based on the ML participants who have been around a while, I view that as important for lowering overload by removing those people from the equation. While you're probably right that in the specific instance of the security discussion that caused this ML to form didn't have bad actors play into it too much, that doesn't mean it isn't a general issue in other threads that get rather long (i.e. threads that have stopped when the people with useful things to say have left and those that just ramble on about anything continue to keep the thread alive, which is when Guido usually mutes a thread and people don't realize it).
We have loose cannons like me, with too much time to post and too little time to do real work. I bet if you look back at the threads that caused Antoine and Christian to blow their tops in public, you'll find them full of good citizens like Terry Reedy, Steven d'Aprano, and Paul Moore. We see the occasional over-the-top phrasing as from Sturla Molden earlier this week.
I'm talking more about people like the one who just always posts a bullet list of links that are almost always about JSON-LD. In almost all cases that person doesn't contribute something useful and so letting new users they don't need to feel the need to reply to them is useful.
None of the cases mentioned above result in "admins needing to make a call" based on a history of misbehavior -- in fact, with the names named it's quite the opposite. So mostly it's a one-time thing, and just a word to the wise from any of the old hands will do.[1]
As an admin I can tell you that is not always enough. And also as an admin I can tell you it's hard to tell when you personally find someone lacking quality contributions regularly vs the ML in general. And if ML and I happen to agree, then I have data backing up my call to go to someone and say "you're not constructively contributing, please try to change". IOW don't necessarily view this as a way to justify banning, but as a way to help people be told they aren't helping out or to let new participants know that this person can be ignored if you want.
participants (2)
-
Brett Cannon -
Stephen J. Turnbull