On folks getting banned by automated censorship:

We pay money to have certain privileges on this site, and freedom of speech should be included when you're forking out $5-$14/mo (depending on the plan you choose). Of course there should be limits to vulgarity, but "kicking" somebody from live chess using an automated "cursing detector" is downright ridiculous. What if you are playing against a close friend and just messing around, talking trash for the fun of it? At least allow enough breathing room for the opposing player to report the profanity if it really bothers them to that point. There are plenty of other chess sites out there  (which I won't name for fear of being banished on grounds of treason) that don't go around policing language and what have you. It's too bad that those other sites aren't creating more competition with Chess.com in order to cause a bit more balance. Alas, this is all merely my opinion. 

Kaynight, how I chat with my friends while playing chess should be nobody's business. Behavior has nothing to do with it. I usually have very good sportsmanship, but I simply see a lot of drawbacks to letting an automated censorship system discern what "bad behavior" is. The best solution would be to just let players play and chat, and allow a human moderator decide if someone has crossed the line. It's the principle of the matter that bothers me- I'm all for good behavior. 

Yes, I agree public forums such as this should be moderated. However, chatting during a game with an opponent is not public, and I believe it should remain private. 

Yeah I think the point is that the system is too inflexible. kaynight, I would think that in most games on this site no third part is even watching, whereas OTB at chess tournaments these days, in the US at least, there are constantly kids around.

Bigger point that just using a curse word is not really that big of a deal. If I'm playing Walla, who is a friend of mine, and he says something like "I'm going to whoop your a$$" that is not offensive to me. But a stranger can certainly offend me without even cursing. So just whether or not someone curses isn't any kind of grounds for banning them imo.

Most of the time it's just people being rude 24/7

No adverse action should be taken against a members account unless it has been reviewed by a real live member of the staff.Get rid of the robot judges.

Can you imagine the pile of work waiting for the poor guy who comes into work at 8 in the morning and has to review all the rude chat that happened in the tens of thousands of live games that took place that night?

This is why there's an automated system. Either that or pay for 100 full time employees to read juvenile chat for 8 hours a day.

Nimzonick: Not only is no third party watching the vast majority of live games, but chat between the players during the game is invisible to all but the players themselves. Likewise, chat between kibbitzers is invisible to the players. Changing the automated detection code so that it only auto-boots the latter should be a very simple matter. It could still flag the player for potentially abusive language (my understanding is that no record is retained of chat, but the flag would be persistent), so that if the opponent complains, moderators would have something more than hearsay to go on.

On the other hand, it's pretty easy to either just be polite or to use alternate characters and spellings to bypass the moderator bot's detection. I do it all the time, with an extremely limited set of friends who I know will tolerate or even enjoy such banter.

As for the "keep it OTB" comment, I'm pretty sure he's talking about city park blitz games (not the best environment for children anyway), where there's a rich tradition of trash talk during the games... some of it quite witty and original. Besides, I'm sick of the whole "protect the children!!!!!11" argument being used to prohibit adults from behaving like adults. If you don't want your child to hear it, maybe you should practice some parenting and keep it away from places where it might hear something you deem damaging to its innocent ears. (Use of "it" intentional. It's not that I hate other people's children, but in most cases I hate their parents. Bring that self-righteous judgmental crap to me when I'm playing in Dupont Circle and I'll make you cry.)

Because it's election year in the U.S., I'll reply with this: "At least we're creating jobs." 

    I think you can curse in private messages too  ( I haven't tried it recentl... well nevermind)

The auto-censor normally warns you first and then it escelates from there.

You might post a suggestion to staff to allow turning off the censor in explicit private chats with those in your friends list. Maybe they would entertain such an option.

I generally don't curse, but I agree that mindless robot detection of curse words can catch people in unusual circumstances that you might not ordinarily think of, and an automatic penalty can be harsh. I got caught by this once, not in this forum but in another.

I had been chatting with a friend: "I haven't had enough sleep."

Then I got warned for profanity.

It took my friend and I a moment to remember that we had been chatting in French, and apparently the auto-censor didn't like the French word for "enough".  When I pointed this out to the administrator, I was told with holy righteousness "the French word for 'enough' is very offensive, you know!"

Now, at the time I got a warning, not an auto-boot.  Still, it was ery uncomfortable watching what we said from then on because we weren't allowed to use a very common word.

Imagine if you were told some common English word was off-limits, and if you accidentally used it a number of times then you would be booted.  (Also imagine, for those of you who might automatically react in a similar way, "Why are you speaking English with your English-speaking friend in a private chat, anyway?  You should be speaking [insert other language here] so that our auto-censor works properly!")

I don't know the exact circumstances of the situation on Chess.com, but urge everyone to keep in mind that there just might be cirucmstances you haven't thought of, and mindless use of a mindless bot should be tempered with human oversight.

Thanks.

I like Martin_Stahl's suggestion, by the way.

censor picks up certain words / phrases

 doesnt pick up

GROLLOX YOU NUMPTY NO WAY CAN YOU KILL MY FOCCACIA QUEEN

 use your imagination its also good fun to come up with new words

"It took my friend and I a moment to remember that we had been chatting in French, and apparently the auto-censor didn't like the French word for "enough".  When I pointed this out to the administrator, I was told with holy righteousness "the French word for 'enough' is very offensive, you know!""

?? What this administrator had been smoking ? "enough" = French "assez", which means nothing else than "enough", and isn't close to any offensive word.

Martin's suggestion is a very sensible one, although it may be easier to automatically turn off the auto mod for games between friends.

The auto mod does record the phrase for which it kicks someone, but the whole chat is not recorded, so the mods will normally take that into account. I don't believe that they review all of the auto mod's actions, but if an account comes to their attention via a report or something they just happen to notice, it'll be there for them to see when they review that account's history

The real question is why can you not say a curse live, but you can leave horrible messages on their main page for anyone to look at!

Yes, even francophones have trouble figuring it out, isn't that so?  Apparently it was too similar to the English word "asses", as in "We're going to kick your ..."  I guess it was put in to prevent people from trying to be creative when circumventing the censorship for the "extremely offensive <sarcasm>" word "asses".  Oh my goodness, I've said it TWICE in this post now!  Oh, no, the end of the world is coming!

That's a reply to all the comments that say, "If you want to curse, simply ... (whatever)".  Okay, but if you DON'T want to curse?

So ... automatic to an extent is okay, but not too automatic, please.

The script is fairly simple,  in my opinion. I can't recall the specific phrase I had when I got warned once, but the last couple of letters of one word combined with the first couple of the next, caused a hit. 

Oh yes, if you consider two different languages, there are a lot of innocent words in the source language that might be offensive in the target one.

For instance the English "bit" well known in computer science might be understood (and pronounced) as the French "bite" meaning "d.ck" - just replace the dot with an "i"  :-)  Language bots are so stupid, this can be seen everyday by reading automatic translations on Facebook, for instance, which are quite often laughable.

You fork out money to pay for the things that your level of membership says you'll get, not a whole load of other things you think you should also be allowed.  It's not difficult to avoid getting censored, it's never happened to me.