I'm making a new site, where there will be a lot of public comments hopefully, and this thought crossed my mind. Is it right to ban usernames with swears in them? Do people even create usernames like this? Trolls do presumably...

  • 13
    Offensive to who? (Serious question.) Is this a site that will primarily be used by a single linguistic and cultural community?
    – msanford
    Mar 26, 2013 at 15:59
  • 1
    The clientele are going to be generally middle-class British Citizens.
    – Lars
    Mar 26, 2013 at 16:03
  • 4
    Clbuttic question!
    – Izkata
    Mar 26, 2013 at 21:01
  • 3
    @Kaz, yes, but it has "swears" in it twice!
    – zzzzBov
    Mar 26, 2013 at 21:33
  • 3
    Automatic detection of offensive words is an extremely complex problem: en.wikipedia.org/wiki/Scunthorpe_problem Mar 27, 2013 at 1:53

6 Answers 6


This is not as simple as it may seem, and although your instinct may be to ban someone, you should first consider the following:

  1. How do you determine what is offensive or not? There are many surnames (last names or family names) that would be considered swear words in English, let alone some other cultures. So you run the very real risk of banning someone for using their legal name. "Bastard" is both a common English surname and a swearword. What do you do with that?

  2. Your site may be one for which freedom of speech falls within the goals of what you do. If Reddit were to ban people for offensive usernames, they would have to ban a huge amount of people. In their case (and some others) banning people for offensive usernames would be plain silly.

  3. You will create a huge potential overhead of administration simply to handle people that believe they have a right not to be offended. Sure, if there are children involved, you may have to do this, but don't underestimate the work that you will add to yourself.

  4. Some names may be ambiguous, and depend on how you read them. #susanalbumparty was a famous twitter hashtag fail which should be read as "susan album party", but many people saw a different reading in it. Or "newsex" was the short name of a company whose full name was "News Extra".

This is not a simple or easy choice. You need to weigh up carefully the impact that this will have on your community and choose accordingly. However, you may find that the best option is simply to contact the person involved and ask them to change their username to something else.

As a side note here. You will never stop everyone from being offended, and I would argue that within reason you shouldn't have to try to prevent that.

I have a Mexican friend whose name is Jesus, and in Latin America that is perfectly normal. But many people in the USA for example would find someone calling themselves Jesus offensive.

  • 2
    +1 overall complete answer; however, primarily on the question of what is offensive or not? and, by whose standards? You can, however, describe, as part of the terms of service, what character combinations will be disallowed. Facebook, for example, requires the use of your real first and last name - of course, how do you prove or disprove the real first and last names of an individual?
    – Josh Bruce
    Mar 26, 2013 at 16:56
  • 1
    @JoshBruce I would rather apply common sense and contact the person involved and give them a chance to defend / explain their choice. From there it would be a judgement call. I really don't want to have to compile a definitive list of what is vulgar.
    – JohnGB
    Mar 26, 2013 at 18:29
  • @JoshBruce in the developed world you could demand a copy of govt ID to verify names. Doing so can become problematic in the developing world. Mar 26, 2013 at 18:33
  • 5
    @JohnGB - That's fair. A combination approach might be to place in the TOS (pretty normal), the right to deny/remove accounts based on username selection. Then, as people create usernames which the site owner deems vulgar or offensives, add them to the list - either in the TOS or part of the sign up form processing itself with a message of something like: The username you have selected has been deemed inappropriate in the past, please select another (or, less honest, just say it is already in use).
    – Josh Bruce
    Mar 26, 2013 at 18:41
  • 3
    @DanNeely This is problematic in the USA too, as a surprisingly large portion of the population don't have any real government ID.
    – JohnGB
    Mar 26, 2013 at 21:19

Short answer: No

There are users who like obscenity and vulgarism, there are those who don't. Generally those kinds of people don't like to mix, and will always try to segregate into separate groups. If you ban offensive names (which is not that easy from technical standpoint), you remove one of the flags those people use to distinguish themselves. However you're not changing the people. You just artificially force them into the same group, which ultimately diminishes user experience for everybody.

For example, suppose you join a game server and see that the majority of users in the lobby have names like M***f**r, or B@D***, or similar. If your own nickname M3G@****, then stop here , end of example. But if you are Lars, you will probably want to go to another game server. Now imagine offensive names were banned. You would stick with the first server, and end up playing with those people, and get subjected to rudeness and obscenity in the in-game chat. If the chat is also filtered for slang – well, there are lots of other, non-verbal ways to be rude. In the end, you'll have a much worse experience than if you knew in advance that the people you're going to play with are impolite.

If you are building a community of respectful people, your users will be such through self-selection. And they'll have normal, non-offensive names (except for an occasional dude with a name like user958624 :). Even if you do get occasional trolls, they can be weeded out manually, and there are much more effective ways to do that than banning / forcing to change one's name. Because you can't force them to change their nature...


It depends on the service and your expected user base. Deciding between blocking and allowing potentially offensive usernames is fairly easy. The more difficult decision is where exactly you draw the line.

You are making a tradeoff. You are disallowing some users from doing something they'd like to do (use a vulgar/offensive name) in order to create a more pleasant, less hostile/offensive environment for other users. Both of these groups obviously prefer one side of the decision over the other, so consider these two points:

  • Which group do you except to be a larger part of your user base?
  • Which group will be more bothered by a decision they disagree with? Is one group more likely than the other to discontinue using your service because of this?

Depending on your target audience, you may want to allow anything, and not worry about people getting offended. If you want a more general (and more polite) user community, you should disallow offensive names.

Two points in favor of blocking such names:

  • If someone is offended/bothered by the names, they may stop using your service. If someone is put off that they can't use insert obscenity here as a username, they may not use your service. Chances are that someone who won't use your service simply because they can't have an obscene/offensive username probably only wanted to troll. Which user would you rather alienate?
  • Some firewalls/content controls filter/block sites based on content. Certain obscene words showing up your page could cause corporate firewalls or parental controls to deny access to the page (or part of it). This could be minor, but it really depends on the names used and the firewalls/content filters.

This leaves the issue of what exactly needs to be blocked. I suggest having a blacklist prepopulated with some known undesirable names (or regular expressions describing such names) that can be expanded as new trolls get creative with ways around it.

  • 1
    I'd also suggest not tying username to login info (use an email for login, or only display a "nickname" that is different from the login user id). This way, if an offensive/obscene name shows up that you want/need to blacklist, you can blacklist it and change the username/nickname to something generic (like "user1234") without impacting that user's ability to log in.
    – yoozer8
    Mar 26, 2013 at 16:27
  • +1 for note that obscene logins could ban f.g. StackExchange on company firewall, in worst case Mar 27, 2013 at 7:36

The best solution is for a site to have "safety" levels. By default, content created by users who are identified as offensive is completely hidden from view. You don't see their questions, answers, comments or accounts at all, unless you switch to unsafe mode.

When you switch to unsafe mode for the first time, you have to acknowledge a dialog that you're considered an adult wherever you live, and that you're about to see content that may be offensive.

If you simply block people, thereby denying them access to an audience, they will keep coming back under different names, and then you have to resort to IP level bans which have various downsides, like collateral damage.

Refinement: When a user is identified as offensive, they can silently be marked as such, and at the same time, their filter can be silently flipped to unsafe mode. (Posting that kind of stuff essentially constitutes consent to read it, and unsafe mode is necessary so that the users can see their own content and believe that everything is working fine, as before.)

The effect of this is that offensive users are not aware that they have been marked offensive. Not only do they continue to see their own content, but other people can see it and can respond. (But of course, all those other people are fellow offensives.)


Short answer: Yes, ban it.

Always a good idea to prevent the user from creating such names. There will be some who bypass this and use numbers for letters and stuff, but you should not make it easy for them use it blatantly.

Also, depending on the context, it is highly desirable, for example, here on stackexchanges, it shouldn't be that big a deal, but in games/public forums it becomes a big deal. The foul names create a hostile environment, say my username was "I F'd your relative", right off the bat, I create a negative impression for myself. And whatever I comment, will be influenced by that first negative impression.

Edit 1: With the interesting discussion on the multi cultural implications of certain words in the same language, I suggest that rather than banning "all" vulgar words, you just select a top % of the list and ban those. It should give you quite a good control over the issue and still leave the more ambiguous terms available.

Edit 2: For all those riding the 'It is against the Freedom of Speech' wave http://en.wikipedia.org/wiki/Freedom_of_speech

"[e]veryone shall have the right to hold opinions without interference"

Now, there is a huge asterisk mark on this right.

Article 19 goes on to say that the exercise of these rights carries "special duties and responsibilities" and may "therefore be subject to certain restrictions" when necessary "[f]or respect of the rights or reputation of others" or "[f]or the protection of national security or of public order (order public), or of public health or morals".[1][2]

Look at the limitation section for more information:

You CANNOT use your right to speech in a manner which is offensive, even though it is a minor crime (in most of the cases).

So yes, it is a good idea to ban generic offensive words. Example: if you have played any decent online game, you might have noticed that the character names are not vulgar for the most part (unless done so using 1337 5P34K or something), the reason is, the developers restrict the use of any such usernames and usually have a handy blacklist. Granted it is not a fool proof system, but it is a first step in the direction. http://www.wowhead.com/forums&topic=151067/banned-name here is an example scenario where it backfires.

Here is wikipedia's own list of rules for usernames http://en.wikipedia.org/wiki/Wikipedia_talk:Username_policy/Archive_12. If you scroll down to offensive, there is an argument that the "lesser" offensive words may be allowed but surely there is a need for the often offensive words to be banned.

  • 11
    How do you determine what is vulgar? You are answering as if it's objective and consistent across cultures.
    – JohnGB
    Mar 26, 2013 at 18:16
  • I agree, vulgarity is subjective. But I would rather be proactive in these situations than going around banning the vulgar ones. Regarding the issue of cultures, I can safely assume that when you are sticking to one language, you have a pretty good idea of what is vulgar in that language irrespective of your own culture. Can you give an example of australian vs US vulgarity difference please, I am not quite familiar with it.
    – rk.
    Mar 26, 2013 at 18:24
  • 2
    What is considered vulgar in Australia differs from what is considered vulgar in the USA. Even between English speaking countries with similar cultures there are huge differences.
    – JohnGB
    Mar 26, 2013 at 18:26
  • 2
    I would be happy to in chat, but I suspect I would violate the site's TOS if I were to post them here.
    – JohnGB
    Mar 26, 2013 at 19:21
  • 4
    You will often find people are MORE offended by not being allowed to have free speech than there are people who will be offended by a particular word.
    – JonW
    Mar 27, 2013 at 11:49

Given that you stated that you're going to cater for mostly British folks, you should consider following their laws. I am not a lawyer, but reading Wikipedia's article on the topic you'll need to account for users who use hate speech in their nicknames.

This might be a hard requirement for you, something you don't really decide “if”, but “how”.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.