Closed
Bug 681925
Opened 15 years ago
Closed 8 years ago
User banning
Categories
(developer.mozilla.org Graveyard :: User management, task)
developer.mozilla.org Graveyard
User management
Tracking
(Not tracked)
RESOLVED
FIXED
People
(Reporter: sheppy, Unassigned)
References
Details
(Keywords: meta, Whiteboard: [triaged][type:feature])
I wanted to put some notes in on the banning process and what we need.
We should be able to ban users by their username, and we should be able to ban IP addresses. Keep in mind while implementing the latter that we have a load balancer; don't ban that by accident (which is what MindTouch does when we try to ban by IP, because it sees everyone as coming from the same IP address).
When banning, we should be asked for how long to apply a ban. This should let us choose between a week, a month, and forever. MindTouch has more options, but realistically those are the only ones we really use or need.
In addition, we need to be able to explain the ban, both for our reference but also for the person that's been banned so they can see an explanation. MindTouch uses an edit box to enter a ban reason, but this is always one of a few specific things, so I'd prefer a popup menu that lets us choose among:
* Spam
* Inappropriate content
* Not a web hosting service
* Willful destruction of content
And maybe "Other" to let us enter an explanatory sentence.
Banned users should see a message based on the reason when they attempt to log in, and should see how long they have left until their ban expires.
| Reporter | ||
Comment 1•14 years ago
|
||
We have to have banning support before Kuma can go live. We get far too many troublemakers to not have it.
Blocks: 710713
Comment 2•14 years ago
|
||
As a former admin of Wikipedia, I have banned several *hundreds* of users. So I would add a few elements.
1) We really need it. Even if we don't have too many crazy people, I remember that I had to set up my alarm clock at 4am for more than a week to give a hand to Sheppy last year (a Brazilian SEO company).
2a) Length of the ban. We should have Sheppy's default values, but also be able to tailor it. From 15 minutes (I used these to calm down a regular fair contributor who just got angry/crazy. "Please cool down, get a drink come back in half an hour :-) " to long bans (several years). On IP we should not put infinite bans (though 2-3 years ban are ok): when the ISP reassign the IP, the new owner is doomed.
2b) Let us choose to communicate the length of the ban or not. Some people set agenda reminders to come back and start again (!) In such cases, it is useful not to communicate the length of the ban (especially useful for IP bans, of course).
3) Like Sheppy said, we should have default messages, but also "free text" messages. That's important. Some times, people are of good will and it is a way to engage with them. We block them because of a big error they do and invite them to talk to us.
4) We should be extremely cautious with IP bans. We need IP range bans but not with the a.b.c.d/8 or a.b.c.d/24 syntax. On Wikipedia, we banned 25% of the city of Bordeaux (~1 mio/inhabitant) once because somebody mixed the /8 and the /24 case... By default, IP bans should prevent new account to be created and only optionally old account to be used (how old should be configurable: 1 day, 1 week, 1 month, 1 year: some spammer create a lot of empty accounts before starting their destructive work!). Also, during a work week, if we ban all Mountain View IPs by error, we want to be able to connect to correct it :-)
Also we need tools (Reverse DNS, ...) only one click away from the IP ban page to check the IP. Banning schools or range of IP of large providers must be done carefully. Especially in these case, we need to craft specific messages.
I know we are not Wikipedia, but that also means that too large ban will be more difficult to detect.
Comment 3•14 years ago
|
||
FWIW, what we have in place now is the ability to make a user profile inactive. But, I don't think we enforce that in terms of rejecting login.
As for the rest of the above... I don't think what we have on hand is adequate, so this will me some fresh development vs just tweaking existing code.
That said, I have to imagine that *someone* in the Django world has thought about this stuff before and hopefully we can find some reusable code to bring in
Updated•14 years ago
|
Updated•14 years ago
|
Updated•14 years ago
|
| Assignee | ||
Updated•14 years ago
|
Version: MDN → unspecified
| Assignee | ||
Updated•14 years ago
|
Component: Website → Landing pages
Updated•14 years ago
|
Whiteboard: p=
Updated•14 years ago
|
Component: Landing pages → User management
Updated•14 years ago
|
Whiteboard: p= → u=user
Updated•14 years ago
|
Whiteboard: u=user → u=user c=User management
Updated•14 years ago
|
Whiteboard: u=user c=User management → u=user c=User_management
Updated•13 years ago
|
Updated•13 years ago
|
Whiteboard: u=user c=User_management
Updated•13 years ago
|
Summary: Kuma: Banning of users → User banning
| Reporter | ||
Comment 5•13 years ago
|
||
This is a meta bug for several other user ban related issues, so no.
Flags: needinfo?(eshepherd)
Updated•12 years ago
|
Priority: P1 → P3
Updated•12 years ago
|
Comment 6•8 years ago
|
||
At the risk of blowing things up, spam appears under control. We added new spam detection and banning features, and it has kept bad edits under control over the last two years. It can be improved, but I think that should be done in the context of new bugs.
Status: NEW → RESOLVED
Closed: 8 years ago
Resolution: --- → FIXED
Updated•6 years ago
|
Product: developer.mozilla.org → developer.mozilla.org Graveyard
You need to log in
before you can comment on or make changes to this bug.
Description
•