18

I read this, on the official answer to What can I do when getting “We are no longer accepting questions/answers from this account”?

My account is in good standing. Why am I still blocked?

The ban also takes in account the IP address. If you browse from a shared computer, or from a location that gives the same IP address to many end users, it's enough that another user who was banned or triggers the ban, will affect everyone else accessing from the same location.

Does this really block all accounts - even clearly trustworthy accounts with a strong history of good questions and answers?

I understand the reasoning behind banning based on IP - to stop bad users side-stepping bans by creating new accounts. But a bad user can't magically create a new account that is, say, 6 months old, or has 99%+ quality posts, or has 1,000+ rep. Are such accounts really included in the block, even though they clearly couldn't be the same person?

For example, what would happen in one of these scenarios:

  • A colleague, housemate, family member or guest of yours - yes, you - is, right now, posting lots of terrible questions, and they're on an IP you'll also need to use (e.g. most WIFI networks, many company networks or any shared workstation or hotdesk). Will you be blocked too when your colleague causes the IP to be blocked, or is your very good track record taken into account? This could cause good users real problems.
  • I log on to SE from a hotel, cafe, guesthouse, library, university residence, rent-a-desk, community room, venue or other public place that gives me an IP used by bad users. Am I caught up in a collective punishment, or is the fact I've proven I'm a good user taken into account?
  • Jon Skeet takes some orphans into his home, who are instantly inspired to become programmers - but before Jon can show them how to ask good questions, he is urgently called away to stabilise a failing nuclear reactor. When he gets home an hour later (the traffic was really bad), one of the orphans has asked a heap of bad questions, causing his IP to be blocked. Is Jon Skeet also blocked, in his own home?

Are all accounts on the IP blocked, or only accounts that could actually be the bad user trying to evade a ban? And if it is all accounts, why, and what is the response to the rare-but-plausible scenarios above?

  • 4
    Jon Skeet does not get stuck in traffic. However, apart from that, good question. – Tim Mar 14 '15 at 12:13
  • 3
    I actually drafted a paragraph addressing that before posting, but deleted it because the question was getting too long. It read, "If the Jon Skeet scenario seems implausible - while he'd normally always be able to avoid bad traffic, there's always the possibility an accident would take place in front of him and he'd need to stop and rescue survivors from the burning vehicles" – user568458 Mar 14 '15 at 12:16
  • 4
    I love the example with Jon Skeet :) – nicael Mar 14 '15 at 12:17
  • I think change it to the slightly more plausible accident one. Although I'd have though accidents wouldn't happen near him. Everything just works. Not to worry though. – Tim Mar 14 '15 at 12:22
  • I don't know if it is true for the question ban, but other IP-based restrictions take user rep into account and established users won't get blocked by them just because they are on a suspicious IP. – Mad Scientist Mar 14 '15 at 15:52
14

...as a long-time user with some amount of reputation on the network, you're exempt from many of the checks we would normally impose.

The system is pretty smart, and it's not going to start blocking a user because other users are doing malicious things on their IP address. Once you reach a certain level of activity and reputation on the site, you can consider yourself safe from these system checks that start blocking IP addresses. They are only meant for new users who don't have experience on the site and are more likely creating accounts for the purposes of doing things they shouldn't be doing.

  • Are you able to elaborate on what the certain level of activity and reputation is? When there's a good clear answer it'd be good to edit it into that other answer (e.g. "...will effect everyone below X rep" instead of "...will effect everyone" – user568458 Mar 14 '15 at 20:41
  • 1
    I doubt there are any hard-coded criteria; it likely still uses secret algorithms to determine if a user is blocked based on the IP. It's just that at much higher reputations, it becomes near impossible to trigger an automated block on an account using those algorithms. – animuson Mar 14 '15 at 20:47
  • Also, in that linked answer, it sounds like the user - who was a long-standing user, 3+ years - was being blocked at the request level before their userID was even seen by the system, due to their IP (?) being mistaken for a malicious bot. It'd be good to be sure that this check isn't like that check. – user568458 Mar 14 '15 at 20:48
  • Getting a "too many requests" error is an automated defense of the server that will happen to anyone sending way too many requests to the server in short periods of time, regardless of who it is. Jon Skeet could hit that block. In fact, moderators used to hit that block frequently when we tried to run cross-references on IPs. But that block also only lasts for a short period of time (usually less than a minute). -- Pretty much everything else you can hit is based on algorithms. You, for example, being a 15k user on Graphic Design would never have to worry about hitting a quality block there. – animuson Mar 14 '15 at 20:55
  • Would this apply to VPN providers? Say a troll is creating new accounts using a VPN, would an IP block stop all users who use that VPN or would trusted users be able to pass? – apaul Dec 16 '17 at 2:24
-1

It does ban all accounts on the IP address. It would certainly be a good feature to add, provided people don't abuse it. I know that that is an implausible request, but it's the only fool-proof way. Who's to say that Someone has 3k rep, becomes annoyed by a surge of bad users, and in an attempt of revenge makes a new account and posts bad answers?

Certainly, this wasn't included when the IP bans were made, but it could be a reason to keep them in place how they are. I'd love to hear an argument from the other side.

  • 3
    "...Someone has 3k rep, becomes annoyed by a surge of bad users, and in an attempt of revenge makes a new account and posts bad answers?" - seriously? Who would that even be "revenge" at? Even if that was a serious concern, they could just do it on their phone (different IP). If we're imagining malicious scenarios, here's a more believable one: developer falls out with company, and before leaving, deliberately gets the entire office banned from asking questions on SO by posting bad questions. – user568458 Mar 14 '15 at 14:53
  • @user568458 your argument makes waay more sense, but argues in the opposite direction. – Travis Mar 14 '15 at 15:18

You must log in to answer this question.

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