The Silence Penalty

[rda]

Correct me if I'm wrong, but didn't that automatic component exist before? It was applied to robospammers.

The new system is not automatic, as you just told me.

It's unclear if it existed before, talked about this a bit earlier:

http://www.mmo-champion.com/threads/...1#post41494548

More importantly, even if it existed before, it just means we had it bad before, we just didn't know we had it bad (and the abuse was non-existing because nobody knew it could be abused).

The new system is automatic enough to be abused.

[Osmeric]

The new system is automatic enough to be abused.

And so was the old system. It just wasn't being abused.

The QQ now is totally off target.

[Davemetalhead]

It's unclear if it existed before, talked about this a bit earlier:

http://www.mmo-champion.com/threads/...1#post41494548

More importantly, even if it existed before, it just means we had it bad before, we just didn't know we had it bad (and the abuse was non-existing because nobody knew it could be abused).

The new system is automatic enough to be abused.

Yep, it's a wonderful community we have ...

I just hope Blizz punishes those that abuse it - I don't think they should change the system, I believe this is a storm in a teacup and will soon blow over, especially if the abusers start getting punished.

I approve of the silence penalty and believe in the long term it will be a good thing, we just have to get through the painful teething issues.

[rda]

And so was the old system. It just wasn't being abused.

The QQ now is totally off target.

First, you are *assuming* that the old system behaved the same way. It is unclear if this was the case.

Second, even if the old system behaved the same way, like I say, it just means we were sitting on a bomb without knowing. It doesn't make the bomb less dangerous. It just wasn't triggered because nobody knew it existed - and now everybody does, this makes a world of difference.

[Osmeric]

First, you are *assuming* that the old system behaved the same way. It is unclear if this was the case.

Second, even if the old system behaved the same way, like I say, it just means we were sitting on a bomb without knowing. It doesn't make the bomb less dangerous. It just wasn't triggered because nobody knew it existed - and now everybody does, this makes a world of difference.

It means getting rid of the new silence system is not the right solution to the problem, since the new silence system ISN'T the problem.

The solution is cracking down on illegitimate reports. I expect that will be happening if the problem persists.

[rda]

It means getting rid of the new silence system is not the right solution to the problem, since the new silence system ISN'T the problem.

The solution is cracking down on illegitimate reports. I expect that will be happening if the problem persists.

Cracking on illegitimate reports is impossible, because that takes hours for each report and in the end doesn't bring results. This was discussed before, I gave an example of this silence (will slightly modify it here to make it even more illustrative):

Innocent victim: Hi, gusy!
Innocent victim banned (by a raid whose raid leader said "quick, report this guy" in voice chat).

Have fun investigating logs of 40 people, finding nothing, wondering whether it was the misspelling of "Hi, guys" that triggered the reports or maybe the person did something bad in the game instead and that was the reason for the reaction, etc, then deciding that you can not ban those who reported the guy because you don't really understand what happened. (And next time you need to investigate "potential abuse" like this, you just skip it and do nothing, because you know you are likely going to spend several hours again and end up in the same place where you have no idea what happened and have little reasons to ban.)

[Sylvaeres]

He didn't get "reported for saying "I Love Wow"" - he got squelched because he had a certain amount of people report him in a very short time-frame. he could have just out a full stop in chat and got his followers to report - the result would have been the same. It's an automatic squelch until investigated, then a silence is given if upheld. It had zero to do with what he wrote.

The issue is it wasn't investigated before happening, even though they clearly stated that they would investigate each issue accordingly

[Osmeric]

Cracking on illegitimate reports is impossible, because that takes hours for each report and in the end doesn't bring results.

That's obviously bullshit. All that's needed is a button in the review interface for when the reviewer concludes the complaint is unwarranted. If a person gets too many black marks they can be automatically sanctioned. Sanctions can range from loss of ability to report, to being silenced themselves, or even steps up the penalty pyramid.

[rda]

That's obviously bullshit. All that's needed is a button in the review interface for when the reviewer concludes the complaint is unwarranted. If a person gets too many black marks they can be automatically sanctioned. Sanctions can range from loss of ability to report, to being silenced themselves, or even steps up the penalty pyramid.

LOL.

You don't read, do you?

In order to decide whether the report is legit or not you have to spend hours. And in the end you come up empty because the abuse - after some time - looks neutral.

You are talking about some big system of marks and whatever, and it won't work, but I won't even discuss why, you are stopped by the very first step - you can't get the marks.

[Osmeric]

LOL.

You don't read, do you?

In order to decide whether the report is legit or not you have to spend hours.

They are ALREADY reviewing the reports. Making bad reports feed back to bite the exploiters would be almost no extra effort.

You are making no sense, and I suspect you are doing that deliberately. Please attempt to debate honestly on this issue.

[rda]

They are ALREADY reviewing the reports. Making bad reports feed back to bite the exploiters would be almost no extra effort.

You are making no sense, and I suspect you are doing that deliberately. Please attempt to debate honestly on this issue.

What is the process you are suggesting? Check whether or not the silence is warranted by checking the chat log of the reported guy, then if not - put everybody who reported him onto some list of potential abusers and if someone ends up being entered into that list for the 10th time or so, sanction them???

[Osmeric]

What is the process you are suggesting? Check whether or not the silence is warranted by checking the chat log of the reported guy, then if not - put everybody who reported him onto some list of potential abusers and if someone ends up being entered into that list for the 10th time or so, sanction them???

They ALREADY decide if the report is legitimate, when the decide whether to apply the new penalty to the person who was reported.

If this comes back "NO", it's a black mark against the reporter.

This is valuable information about the reporter even if one doesn't find (in chat logs) evidence of deliberate collusion. Do this enough and the reporter should be penalized in some way, for example by reducing the weight given to their reports or ignoring them entirely.

In practice, I expect the number of episodes of deliberate collusion will be small enough that they could be investigated in depth at low cost. The ringleader(s) could then be subject to extra attention.

This is really like any other sort of ToS violation. That it costs effort to enforce the ToS is not an argument against the ToS.

[Mosotti]

http://us.battle.net/wow/en/forum/topic/20747784573

http://us.battle.net/wow/en/forum/topic/20747724644

But it is happening. What do you have to say now?

I'd have to believe them, but I don't. You see, the complains are not done on forums, but on tickets. The only reason to do it on forums is to create attention. As for Asmongold, it doesn't count, it's something else...

[rda]

They ALREADY decide if the report is legitimate, when the decide whether to apply the new penalty to the person who was reported.

If this comes back "NO", it's a black mark against the reporter.

This is valuable information about the reporter even if one doesn't find (in chat logs) evidence of deliberate collusion. Do this enough and the reporter should be penalized in some way, for example by reducing the weight given to their reports or ignoring them entirely.

In practice, I expect the number of episodes of deliberate collusion will be small enough that they could be investigated in depth at low cost. The ringleader(s) could then be subject to extra attention.

This is really like any other sort of ToS violation. That it costs effort to enforce the ToS is not an argument against the ToS.

I repeat the question - what is the process you are suggesting? Be clear. A person gets silenced. What it is that you suggest they do, what exactly they should check, etc.

[Osmeric]

I repeat the question - what is the process you are suggesting? Be clear. A person gets silenced. What it is that you suggest they do, what exactly they should check, etc.

For every person reported, they look at what he said publically over a time interval (say, ten minutes) before the reports. They look for anything that could be objectionable.

They classify this three ways: (1) objectionable, which upholds the new penalty, (2) questionable, which does not uphold the penalty, but doesn't affect the reporters, and (3) innocent, which does not uphold the penalty, and in addition gives each reporter a black mark.

If a reporter gets enough black marks they are investigated in greater depth. If a group of reporters are getting black marks together that flags them for extra priority attention.

[rda]

For every person reported, they look at what he said publically over a time interval (say, ten minutes) before the reports. They look for anything that could be objectionable.

They classify this three ways: (1) objectionable, which upholds the new penalty, (2) questionable, which does not uphold the penalty, but doesn't affect the reporters, and (3) innocent, which does not uphold the penalty, and in addition gives each reporter a black mark.

If a reporter gets enough black marks they are investigated in greater depth. If a group of reporters are getting black marks together that flags them for extra priority attention.

You aren't being clear enough, but fine, I'll deal.

If they only ever look at the specific phrases of the person who got silenced, this is a no go, because this throws away context. In order to determine whether or not the person who got silenced said something silence-worthy it is enough to examine HIS chat log and nothing else - that takes some time, but fine, it's one chat log. However, in order to determine whether or not those who reported that person had reasons to report him, you have to review THEIR chat logs - because it's THEIR chat logs that they saw when they were pressing the report button. So, you are looking at 30-40 chat logs to investigate potential abuse right there. That's an hour or two already, you'll get tired simply switching channels and scrolling / scanning.

But that's just the start of it.

Is this silence-worthy: "Shit, same drop again"? Perhaps it is, right, there's the S-word. What about "Damn, same drop again"? Perhaps not? Gee, you are making a judgement. Good. Let's continue. How about this: "Bloody hell!" You see, the same word is a no-no in some regions and completely fine in others. And you have no idea who is hearing this. Is it offensive? You don't know.

I can continue ad infinitum, it never ends and you will never be great or even good at deciding whether or not the report was warranted. So you will have to compensate for that by only flagging people as "abusers" after they make like 10 fake reports. And if they just make one more valid report in between, the count will get reset (if it doesn't - you aren't being specific here either - you have grave other problems).

And that's not all there is. The moment you ban someone for abuse, you will get tickets regarding that ban, that will eat more of your resources, and this time this will eat personal GM time. And the worse you are at detecting abuse - and you are going to be pretty bad, I guarantee it, because it is a soft matter - the more of those reports eating time you will get.

And that's by far not all either, but I will stop here.

Sum total, this is megahours of time for nothing. Nobody has resources for this, not just Blizzard, nobody.

[Pann]

You aren't being clear enough, but fine, I'll deal.

If they only ever look at the specific phrases of the person who got silenced, this is a no go, because this throws away context. In order to determine whether or not the person who got silenced said something silence-worthy it is enough to examine HIS chat log and nothing else - that takes some time, but fine, it's one chat log. However, in order to determine whether or not those who reported that person had reasons to report him, you have to review THEIR chat logs - because it's THEIR chat logs that they saw when they were pressing the report button. So, you are looking at 30-40 chat logs to investigate potential abuse right there. That's an hour or two already, you'll get tired simply switching channels and scrolling / scanning.

But that's just the start of it.

Is this silence-worthy: "Shit, same drop again"? Perhaps it is, right, there's the S-word. What about "Damn, same drop again"? Perhaps not? Gee, you are making a judgement. Good. Let's continue. How about this: "Bloody hell!" You see, the same word is a no-no in some regions and completely fine in others. And you have no idea who is hearing this. Is it offensive? You don't know.

I can continue ad infinitum, it never ends and you will never be great or even good at deciding whether or not the report was warranted. So you will have to compensate for that by only flagging people as "abusers" after they make like 10 fake reports. And if they just make one more valid report in between, the count will get reset (if it doesn't - you aren't being specific here either - you have grave other problems).

And that's not all there is. The moment you ban someone for abuse, you will get tickets regarding that ban, that will eat more of your resources, and this time this will eat personal GM time. And the worse you are at detecting abuse - and you are going to be pretty bad, I guarantee it, because it is a soft matter - the more of those reports eating time you will get.

And that's by far not all either, but I will stop here.

Sum total, this is megahours of time for nothing. Nobody has resources for this, not just Blizzard, nobody.

You're over complicating things, the system that Osmeric proposed is simple, time effective and workable, there is simply no reason to bring in the level of investigation you seem to believe is needed.

[rda]

You're over complicating things, the system that Osmeric proposed is simple, time effective and workable, there is simply no reason to bring in the level of investigation you seem to believe is needed.

I love these one-liner replies that ignore everything they quote.

Here's my reply to your reply:

No.

[Evilfish]

Haaaaaa! Look at this. Not two days in and the System has proven to be a total failure. BTW you can now "buy" votes to ban people you don't like. It's a little ironic. This was mean to silence assholes but not it's being used by assholes to silence others because there are no consequences for false reporting... I'm curious how long Blizzard is gonna let it go on.

[rda]

Haaaaaa! Look at this. Not two days in and the System has proven to be a total failure. BTW you can now "buy" votes to ban people you don't like. It's a little ironic. This was mean to silence assholes but not it's being used by assholes to silence others because there are no consequences for false reporting... I'm curious how long Blizzard is gonna let it go on.

If there's an API that can be used - even in a roundabout way - to file a report, mass reporting, and, consequently, insta-silencing, will be as simple as doing ready checks.