I was talking to someone else in the real world about this briefly (came up through a converstaion of internet cybe bullying) and they told me that in some chat rooms on the internet, that if you swear; just once, you are automatically booted.

Even if mods are in these chat rooms, then this "bot" would automatically boot people for swearing just once.

Now I am not suggesting that - but it works on that site, people think about words twice before typing. I think it should only be in use when no moderators are logged into the game.

If it works on other sites, then surely it can work here too.

Yes, it does - as long as people do as they are constantly encouraged to: Send complaint so the staff get the logs.

That's all there is to it.

Lol chill you no me.. am just messin

It's not a discussion about whether it can be done or not. Adding automatic boot based on the filter isn't hard or much work at all.

The question is whether or not we should.

The filter is far too imperfect for this - as previously explained. Nor would such a mechanism be able to apply the level of discretion that is usually required.

Lets revisit this topic after you've had a couple months as a mod under your belt and I'm sure you may have a more sensible and balanced view on this

fair enough...it might just be that

one more point of argument (its just not me without one), nothing is perfect, and as I have said, its to complement the current system...

I rest my case for a couple of months - it may be bumped then...unless someone wishes to keep going on this.

Idea: anyone who uses a variety of swear words in one sentence or chat room "entry" could be automatically booted as it proves they're up to no good.

However if your in a PM with one of your mates, whats the matter swearing then? That IMO would be a flaw with the automatic boot.

Agreed. Although someone might PM a stranger with random profanity, which wouldn't be so agreeable...

well while swearing is not allowed on the game - I suppose friends do like to add in the odd word here and there.

It may be a slight backstep on my stance - but maybe dont count a strike IF the talk is in a PM window - to a friend, and the swearing is VERY limited??

or too confusing?

In the same respect I know that me and a couple of friends are the same in a game when its just us which is surely the same as just in a pm!

and what do you mean "very limited" surely.. you swear or you don't lol you cant "half swear"!

ok, when talking to a listed friend in a PM window - I ment like make two words one strike.

But I agree - you do or you dont and it should be consistent.

So whats the different between a private game and a pm window?

and before some smart bum tells me ones a game don't bother lol you no what I mean

^^ there is no difference - the rules apply no matter what area of the site you're using.

Implimenting this would be no problem, as it is already implimented for guests.

Warnings are issued, and if ignored, they are automatically booted.

It has not been implimented for members because people could make innocent mistakes (due to the inhuman nature of the chat filter as already mentioned) and end up losing clan, tournament or ranked games by making a simple typo.

So my suggestion - same criteria as guests (3 offences) would cause the player to be muted and a chatlog sent.

This way you wouldnt lose out if it was a mistake as you could carry on playing, you wouldnt have a "black mark" against you as the chatlog would show the context, and if someone was being abusive, other users are automatically protected.

I completely agree with Damien on the need to distinguish between public chat vs private messages to between friends. The "family friendly" aspect may also often be inapplicable for PMs as you long as you know who you are talking too.

The problem is that there's virtually no foolproof way to implement a system that automatically takes this into account. The definition of friends or "people you know" is much more complex than the friend mechanism used on funkypool.

Muting is a more realistic option yes, since a flawed base for triggering the reaction would be less of a problem since the reaction is far less severe.

but I still dont quite see how it is flawed.

I must be missing something jan - because it doesnt matter how it comes about - but swearing three times in say 15 minutes (regarless of how good the filter is - it only blocks bad words) can not constitute a punishment.

I cant see how it can be mistaken. No interpretation is needed for swearing. If it was for content of conversation - then yes, cant be done.

So if the filter is triggered, strike one. Only keep count for say 15-30 minutes - then people will not be booted for accidental typo's.

As spinner says in the old post - no one accidently has three typo's that turn into swear words in 15 minutes - and on strike two, you would be more careful.

I just dont see how it is flawed jan - sorry

NB: spoiler there for easier reading as i make a side note

Yes it is. Context can greatly affect how a given word is, and should be, interpreted.

And as many have already pointed out, including Dave; the filter filters words that often comes as a surprise to users. Good examples of this are combinations of letters that stand-alone would form a bad word, but being a completely legitimate part of a larger word.

The simple point is that the automatic boot system would rely entirely on the performance on the filter. Any flaw in the filter directly impacts the automated boot system. So by making the consequences of violating the filter more severe, you also increase the requirements for the filter to be intuitive according to average user's expectations.

Flagging things is quite ok, automating reactions on the other hand implies requirements to the information base far beyond what the filter current represents.

Lol i got booted before without warning.

No one says you necessarily should be warned before booting. That is a different discussion entirely.

The issue is that a booting should be subject to discretion only humans can have, and not based on blind automation - especially considering how bad the current filter is.

After reading through the first thread and the second, on a whole i think its a good idea, Sorry jan i cant see the flaws that you see, if its implemented right it should take off its self,
What you said about blind automation, i think it can be done why cant a computer that can interpret swear words through a filter not interpret that into strikes..?