According to this answer, automatic bans are at a lower level than the account, i.e. bans are based on IP address.

There are two problems with this:

  1. There are many people who share IP addresses that can negatively affected as a side effect of the ban.
  2. Banning an IP doesn't really accomplish much of anything as far as preventing a person from continuing to post: it's not that hard to change your IP address, and you can always just switch computers or networks.

IP bans are ineffectual in stopping the activity of malignant people, and have collateral damage that can affect normal people. In short: don't ban IP addresses.

Instead, use account bans. They don't accomplish anything less than an IP ban, but they don't cause collateral damage.

  • 1
    And then what? Assuming you are not contesting the post ban itself, how would it work?
    – yannis
    Oct 11, 2012 at 1:38
  • @YannisRizos What does an IP ban accomplish that an account ban doesn't? Oct 11, 2012 at 1:41
  • 1
    People who generally are not smart enough to read and respond to the many warning signs they will receive before a question ban usually are not smart enough to realize "oh, I need to change my IP".
    – waiwai933
    Oct 11, 2012 at 1:42
  • 2
    @Peter It stops quite a few people from creating secondary accounts. While it's true that there might be some false positives every now and then or that some people may change their IP, from what I've seen it's working as expected more often than not.
    – yannis
    Oct 11, 2012 at 1:43
  • It depends whether people are doing it through stupidity or just a disregard toward rules. Why is let me change accounts such an easier conclusion to make than let me change computers or let me switch my network? Oct 11, 2012 at 1:44
  • @YannisRizos It doesn't prevent people from creating secondary accounts. Even with an IP ban, it's trivial to create a secondary account. How can you know people aren't creating secondary accounts? If, as far as the system can tell, it's a legitimate different account, there's no way to track it. Oct 11, 2012 at 1:47
  • @PeterOlson I don't know why, but it is what it is. Typically (anecdotal) post banned users that don't care of reversing the ban the right way, will go for a secondary account or, you know, asking their coding questions on MSO.
    – yannis
    Oct 11, 2012 at 1:47
  • @PeterOlson Creating a secondary account is not an issue, didn't phrase that correctly, what I meant was creating a secondary account and using it for the sole purpose of circumventing the ban. They can create all the accounts they want, but if the ban is at a lower level they can't use them...
    – yannis
    Oct 11, 2012 at 1:48
  • @PeterOlson Case in point <sigh>
    – yannis
    Oct 11, 2012 at 1:49
  • @YannisRizos Sure they can. It only prevents them from circumventing the ban when they have the same IP address, which is like putting a chair in front of the door and pretending the door is locked. Oct 11, 2012 at 1:51
  • @PeterOlson Yes and some do (I'm battling such a troll on Programmers right now). But most are simply not smart enough, read waiwai933's comment again.
    – yannis
    Oct 11, 2012 at 1:52
  • @YannisRizos I think there's a selection bias here. The people who switch somehow to make it so that they're not using the site with the same IP do not have any observable behavior because the system cannot recognize that they are the same person. The people who are too clueless to switch of course will come up with a different solution. Oct 11, 2012 at 1:52
  • @PeterOlson Going in circles, let's try this another way: Your question is based on the premise that IP bans may produce false positives. Any data to back that up, even anecdotal? The very next question posted on MSO after yours is by someone trying to circumvent the ban by asking a coding question here, anecdotal, but...
    – yannis
    Oct 11, 2012 at 1:58
  • @YannisRizos, there have been cases here where coworkers have managed to get each other banned through bad question asking. These are in the minority.
    – Charles
    Oct 11, 2012 at 2:01
  • @Charles Certainly, I'm not denying that there were instances of false positives, I'm denying it's a problem affecting a lot of people. There will be false positives in any post ban mechanism we can think of... Right now I think IP ban false positives are quite rare and handled appropriately. If they are still in the amounts that the team can handle them manually, I'd say the IP ban is working fairly well.
    – yannis
    Oct 11, 2012 at 2:02

1 Answer 1


See Jeff's answer to "Can we prevent some of the low-quality questions from entering our system?" where he announced the IP ban for users who have a history of extremely low quality posts. In particular:

Based on our queries and a random audit sample of affected accounts / IP addresses, it seems effective, but we'll have to see now that it's deployed and perhaps tweak further.

We used to ban individual accounts for low quality posts, but we found that people would just create new accounts to get around the ban.

  • 2
    While I understand and agree with the sentiment here, this is likely to be an unsustainable system since for the foreseeable future we are working with the IPv4 address space and many users will be working on dynamic IP connections. The influx of new accounts appears exponential - by trial and error I have found that the current newest user ID is 1782549 (by the time you read this it will likely be more), meaning that the number of registered accounts has more than doubled since I joined just over a year ago. I would propose that the IP blacklisting be temporary - say for 2 weeks. Oct 29, 2012 at 10:11
  • 2
    Agreed. On today's age, most people change IP addresses every time the open their computer. Sometimes, while walking the street with their mobile phone. I can't really see how IP blocking can be that effective. If the user opens a new account, one of two things would happen. 1. He'll ask more bad questions, and be banned again. 2. He'll learn from his experience and start asking two questions. If it's the former, we don't mind. The community moderation will take care of that. If it's the latter, the community earned another good member. Oct 29, 2012 at 10:16
  • lmao nice response Feb 21, 2013 at 7:50
  • 2
    Sounds like a crap idea to me. You shouldn't try to be to smart. Simple solutions are often best. On a simple level any relation between "user desirability" and "ip address" is a bad idea because many user share ip addresses (corporate firewalls, tor network, other proxies, ...). Another reason not to do it, it's not necessary or even very effective, and has a lot of very nasty user experience as consequence, in short don't ever go there unless you want people to nag at you incessantly.
    – user148312
    May 19, 2014 at 19:30

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .