Historically, there were multiple cases where spams were sent with faked
From: email address of existing list members. Any solution will need to
not end up blocking innocent members.
Regards,
KTC
On 10/05/2021 17:51, Adithya K wrote:
I have a suggestion to this. I think list admins can be provided with
two options: one, to block a person only from the specific list and two,
to block the user from all the mailing lists. Not sure whether it can be
implemented.
On Mon, May 10, 2021 at 10:10 PM effe iets anders
<[email protected] <mailto:[email protected]>> wrote:
I was indeed thinking of multiple lists confirming the same
behavior. But that will never fully exclude Asaf's concerns.
The use case I'm thinking about is not the individual annoying user
(no rush with that) but rather the spammer that sends the same spam
to all kinds of mailing lists. For that, some automated process
would speed up things considerably, and avoid cluttering the system.
There may be better solutions to address this though.
Lodewijk
On Mon, May 10, 2021 at 1:04 AM Dimitar Dimitrov
<[email protected]
<mailto:[email protected]>> wrote:
Yes, some global & human review seems necessary.
Maybe a higher threshold would make sense. E.g. if a user gets
blocked on two or three lists, then a global block is triggered.
На сб, 8.05.2021 г. в 14:20 ч. Asaf Bartov
<[email protected] <mailto:[email protected]>> написа:
(This could also be an attack vector, though.
Imagine a local conflict resulting in list admin A marking
person B's posts as spam on that local list, effectively
silencing person B across all Wikimedia lists.
Without mitigation against that [e.g. a global review queue
of marked spam before it becomes a global block], I would be
wary of such a feature.)
A.
On Sat, 8 May 2021, 02:05 Kunal Mehta <[email protected]
<mailto:[email protected]>> wrote:
On 5/7/21 3:04 PM, effe iets anders wrote:
> With the new mailman, I'm really excited to see
global bans. Is there
> already some workflow we're supposed to follow to
mark spam? It would be
> great if listadmins could mark stuff as spam and that
would then trigger
> global bans somehow if it happened multiple times. I
know I can file a
> ticket, but want to make sure we first arrive on a
desirable solution
> before we throw it into the dev's lap(s) :)
Yeah, that would be really nice. In the past this has
been proposed as
"Allow list admins to train spam filters"[1] after GNOME
patched their
Mailman to do so.
Upstream mailman3 has a bug for this as well[2], where
they suggest that
a plugin could be written for this. If someone writes
said plugin and it
works, we can definitely deploy/enable it.
In the meantime I would suggest if you notice a new spam
problem, post
here, see if other list admins also say they're having
similar problems,
and then we sysadmins can take action.
[1] https://phabricator.wikimedia.org/T244241
<https://phabricator.wikimedia.org/T244241>
[2] https://gitlab.com/mailman/mailman/-/issues/88
<https://gitlab.com/mailman/mailman/-/issues/88>
-- Kunal
--
Katie Chan
Any views or opinions presented in this e-mail are solely those of the
author and do not necessarily represent the view of any organisation the
author is associated with or employed by.
Experience is a good school but the fees are high.
- Heinrich Heine
--
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
_______________________________________________
Listadmins mailing list -- [email protected]
To unsubscribe send an email to [email protected]
By sending a message to this list, you email all admins of all lists. To
request technical changes for a specific list, instead create a task in
Phabricator. See https://meta.wikimedia.org/wiki/Mailing_lists
