Boards›Player Behavior

Would a "ban your own word" system work in LOL?

y0r1ck·4/4/2019, 10:11:01 PM·3 votes·1,915 views

A ban your own word system would work like this, you enter words you find offensive into the client and then every chat message comprised of one or more of those words is filtered from your view.

One immediate issue I see is what if people enter in words like "the," but then who cares? A person can only affect himself with this system, and if someone wants to do that let him.

Does a system like this work for phrases? Theoretically yes, but in my opinion there are too many ways to phrase something to ban phrases.

Another issue is you won't be able to report people you would normally report. Of course how would you report something you don't know exists? But then again if you can't tell it's there, is it a problem for you?

One more issue I see is that people may be able to get around the word filter by replacing letters with symbols or switching letter around. I do not know how effective RIOT's algorithms are, but regardless of their efficacy the current report system represents a solution. If someone gets around a filter, report it in such a way that the filter learns to recognize a new word you dislike. I don't know exactly how RIOT would do this, but I imagine that it could be like you specify the word you find offensive, and find in a chat log the word that the filter should have recognized. Now the filter knows the word.

Another important issue is the issue of ratings and government regulations. As we know RIOT has to comply with regulations to meet a certain age rating. Does a system like this meet the requirements of rating agencies and governments?

The last issue I see is that people are lazy and will enter as few words as their conscience allows. Personally I think that's good, the less words we allow to offend us the better. I don't think this point is a real problem, but some might?

This last part may or may not be ethical. Discuss this as well. If a system like this is implemented, should RIOT collect data from the players and use it to build a list of words that when used lead to punishment? Say 61% of players ban a word, is it ethical to say that nobody can say that word?

Most importantly a system like this avoids the whack-a-word that RIOT plays, and lets users take their emotional response into their own hands.

There are some people who say that some words should never be spoken. Clearly this post is not for those people, since this is about letting you decide what language you see and use as opposed to letting others decide for you. If you're one of those people feel free to leave your thoughts, but the premise of the post goes against your beliefs so we'll just end up agreeing to disagree.

11 Comments

zPOOPz4/4/2019, 10:24:56 PM3 votes

It should not be my responsibility to enable or shield people who cannot control themselves in chat. While I don't have a problem with zero tolerance word or fuck or shit or asshole, etc. I still don't care to see it used for flaming even if not directly against myself.

I consciously turned off language filter in the client because I am not a kid. It doesn't mean I won't report you if I see you use it in offensive ways (whether directed against me or my teammate or the enemy).

I will not use your system. I don't get triggered by word. But I will mute, report, and move on with my life. They are not mutually exclusive. Just because I don't mind seeing it doesn't mean I should have to see it. Learn to control yourself or I will help you (via report).

Umbral Regent4/4/2019, 10:38:37 PM3 votes

A ban your own word system would work like this, you enter words you find offensive into the client and then every chat message comprised of one or more of those words is filtered from your view.

There are a number of issues with this idea, but the first few that I want to bring up are;

  1. Voluntary systems typically go largely unused. There are certainly some people with sensibilities that such a system would help, but by and large, I can't imagine a majority of players going out of their way to set up their own personalized chat filter, especially if they believe that they're tough enough to handle whatever comes their way. The idea ultimately may not see enough use to justify its existence.

  2. Word filters in general are only half-effective. You don't know what exactly is being said, but provided you know what the filter covers (be it mature language or a personalized list of words), you'll still know generally what the word is. Especially if we're talking a super-simple basic filter that replaces the letters in the word with asterisks or percent-signs. It wouldn't take much to guess what's being said when you know how long the word is.

Another issue is you won't be able to report people you would normally report. Of course how would you report something you don't know exists? But then again if you can't tell it's there, is it a problem for you?

This issue isn't precisely as much of one due to the effectiveness point mentioned earlier, but it could be if the filter flat-out hides the message containing the filtered word.

That said, it is still worth considering that players should be able to know that they need to report someone. Having a filter doesn't damage this ability, more so as it does muddy the line for knowing what should and shouldn't be reported - there are some things, after all, which aren't punishable offenses that may still offend some people's sensibilities.

One more issue I see is that people may be able to get around the word filter by replacing letters with symbols or switching letter around. I do not know how effective RIOT's algorithms are, but regardless of their efficacy the current report system represents a solution. If someone gets around a filter, report it in such a way that the filter learns to recognize a new word you dislike. I don't know exactly how RIOT would do this, but I imagine that it could be like you specify the word you find offensive, and find in a chat log the word that the filter should have recognized. Now the filter knows the word.

This presents two separate problems with similar costs; first being that, one could simply take an input (for example, the word "damn"), and have the system automatically register every conceivable variation of the word...But that would take an immense amount of time and effort to code up, since you have to account for lowercase, uppercase, accents, numerals, and some other symbols - all mixed together. Elsewise, someone might be able to work past the filter with something like d@Mñ.

The other problem would be letting players handle such work-arounds themselves, either having them write up however many variations of the word to block, or letting them do trial-and-error if/when a work-around appears.

Neither options are really feasible, though.

Another important issue is the issue of ratings and government regulations. As we know RIOT has to comply with regulations to meet a certain age rating. Does a system like this meet the requirements of rating agencies and governments?

Well, IIRC, age ratings don't cover online interactions, just the actual game content.

The last issue I see is that people are lazy and will enter as few words as their conscience allows. Personally I think that's good, the less words we allow to offend us the better. I don't think this point is a real problem, but some might?

As mentioned earlier, volition is a problem in and of itself. That said, I'm not sure if it'd really be necessary to have a personalized filter, since those with sensibilities can simply toggle the Mature Language Filter on, and those without can simply toggle it off. Though, this comes with the caveat that there are some words that can offend sensibilities that aren't covered by the mature language filter AFAIK, so I'd say it's a moot point.

If it's a problem for a lazy person with sensibilities, then it's a personal problem.

This last part may or may not be ethical. Discuss this as well. If a system like this is implemented, should RIOT collect data from the players and use it to build a list of words that when used lead to punishment? Say 61% of players ban a word, is it ethical to say that nobody can say that word?

I'd say, no. The Mature Language Filter is a good example of the middle-ground here. It filters out swears, and if I'm not mistaken, racial and homophobic slurs, amongst some other things, but some of those filtered words (swears in particular) aren't considered punishable in and of themselves.

Most importantly a system like this avoids the whack-a-word that RIOT plays, and lets users take their emotional response into their own hands.

I'm not sure I understand what you mean by "whack-a-word". Do you mean how Riot has to reactively punish misbehaviors/abusive words, or is there something I'm missing?

Kei1434/4/2019, 10:59:02 PM3 votes

It's not words that annoy me, it's attitudes.

  • Things not going their way? Give up and AFK farm.
  • Play that went bad? Proceed to blame others.
  • Flame war started? Got to keep the flame going and have people typing rather than playing the game.
  • Got flamed? Going to throw, cos they don't deserve to win.

The words can be changed and replaced, but the attitude remains and it is dam annoying to have to deal with the consequences ingame.

Telephone Booth4/4/2019, 11:55:59 PM1 votes

The part about using those words to create their own filter reminds me.of the current standard for toxicity that we have. I've heard that what they consider toxic, is based on how people voted in the tribunal. Kind of the same concept. Just interesting to me, i have no points to make lol.