The Silence Penalty

[Gramul]

I like the idea of community dictated moderation. My only concern is that some might use it as a way of abusing others. For example, someone could get their linkshell to constantly report someone they just don't like.
If they can prevent that kind of abuse, then I'm all for it.

[Dustytome]

I like the idea of community dictated moderation. My only concern is that some might use it as a way of abusing others. For example, someone could get their linkshell to constantly report someone they just don't like.
If they can prevent that kind of abuse, then I'm all for it.

I would hope they still keep watch over it and if abuse is happening, a much worse penalty befalls the people involved.

[NolLacnala]

I like the idea of community dictated moderation. My only concern is that some might use it as a way of abusing others. For example, someone could get their linkshell to constantly report someone they just don't like.
If they can prevent that kind of abuse, then I'm all for it.

The system they're implementing will need to not be fully automated and will require GMs babysitting the reports, since I've read several stories of GMs warning people for abusive language, when the only players they cursed at were gold sellers. This implies the gold sellers actually report people for harassment lol...

[KitomiSaitichi]

I like the idea of community dictated moderation. My only concern is that some might use it as a way of abusing others. For example, someone could get their linkshell to constantly report someone they just don't like.
If they can prevent that kind of abuse, then I'm all for it.

There is an easy way to mitigate this.
The frequency of messages, the number of users telled to, the discentrality of spammers and the number of previous interactions initiated by the player to the spammer before the first tell received are some easy factors to consider.
Further first-level passes and checks include a low account age or moderate account age with no high level or an inactive account that suddenly logs on as red markers.

From fact by observation, we know that bots consistently send messages on a periodic basis to multiple players. A normal player is more sporadic with their tells, and usually only whispers one or two people at a time at most. Spammers might use locations of cached players' names or simply an observer that parses everyones' names nearby, while players typically tell to people they met already or had interacted with on a per-frequency basis much higher than spam bots. This could be from dungeons, from free company invites, from trading - whatever the case. Spam bots are more likely to be lower level players, simply by virtue of mass creation being the more economical choice when choosing to spam messages than initiating any effort in obtaining a certain character level to seem like a 'real player'. Organic players don't just pop up one day and tell many random players. They do quests, they move around and stuff. Metrics like dungeons done, quests done, achievements done, etc can correlate to a certain player profile. Deviants after sufficient analyses of these bots can come up with a profile that just doesn't add up, creating a complexity factor hurdle that needs to be overcome with better programming to skip through these hoops and loops of red flag heuristics, effectively lessening the impact of spam bots altogether.

After a player-based 'vote', we could add an automatic test that followings through the aforementioned metrics/profile that is being weighted against, and then finally confirm it to be a 'bot' or not and quickly silence them (while having them on their screen show as spamming messages). We could also add a second metric that determines the reputability of each said player after confirming some random samples of 'autoconfirmed silenced bots', and players with low reputability, either stemming high correlation of a subgroup like [A, B, C friendship; A, B, C free company; A, B, C linkshell) ganging up on people or reporting 'organic players' could simply just have their report weights be zero.

Kicking AFK players is not a systematically sound solution. Secondly, 'spam bots' must mimic keyboard input, thereby negating the 'AFK' arguement completely. Organic players do AFK at random intervals, they do not follow any predetermined threshold or steps. However, from a probablistic perspective, people are more likely to be AFK after a certain period of time from logging in more so than when they just logged in. Since players logged in typically don't feel burned out from playing yet, and they are more likely to have less time waiting on them like going to work or having to do other responsibilities, factoring in that most people decide to block in their game time during the day when that time is a leisure block and as you progress through it - there are more likely to be interruptions like calls, and whatnot.

Curbing RMT activities is also possible. If the marketplace is a place to transfer currency, then we can infer that if we take the 7, 15 and 30-day trending prices of various items, and take the weightings of the price by each respective unique buyer, and set a % threshold over the average and median price like 5000% on item X, then we flag all these buyers who sell 10 gil priced items for 100000 gil or something like that. Additionally, a very disproportionately low amount of people would ever set their selling price to higher than the lowest going price, although people do put them at the median 50% percentile of the list (i.e. a list of 10 prices), or better - rarely do they set a price the highest of the market and over a magnitude of 10 times or more unless they're buying out the market. Secondly, it's more likely that these gil sellers are using these proxy measures because of the level of convenience and lack of tracability. Players that do more than one purchase transaction of this type, are, more than likely not by mistake, and are bots themselves.

[Sanghelios]

I like the idea of community dictated moderation. My only concern is that some might use it as a way of abusing others. For example, someone could get their linkshell to constantly report someone they just don't like.
If they can prevent that kind of abuse, then I'm all for it.

Well of course they can. If a group of people continues to report you, you can just report that to an GM and they will give them a warning/ban. Easy as that.

[Niwashi]

Well of course they can. If a group of people continues to report you, you can just report that to an GM and they will give them a warning/ban. Easy as that.

You don't even need to report it to the GMs since they're the ones who get the original reports, so they'll already know when false reports come in. (And that's a good thing, since you also wouldn't be able to report it. Who would you even report? It's not like you'd know who had falsely reported you. You might not even know you'd been silenced.)

One person reporting a message that's not RMT spam would probably just get a warning. (After all, maybe they just mis-clicked when trying to report the name below yours, rather than having deliberately tried to harass you.) But that's also why a single report wouldn't silence anyone, and it takes multiple reports against the same person before that happens. If 20 or 30 false reports all came in against the same person, it's not going to be a simple error. The only way that can happen is if they all conspired with each other to harass that player. A harassment conspiracy of that sort justifies a permanent ban on all their accounts rather than a simple warning.

[MaeIsMean]

I like the idea of community dictated moderation. My only concern is that some might use it as a way of abusing others. For example, someone could get their linkshell to constantly report someone they just don't like.
If they can prevent that kind of abuse, then I'm all for it.

If you read how Blizzard is doing it, the person doesn't get silenced until a GM reviews the reports. It's not automatic. So abuse will be easily spotted by a bunch of reports on someone for no reason. Unless the person was being abusive or spammy, the GM wouldn't silence them.

[Berethos]

If you read how Blizzard is doing it, the person doesn't get silenced until a GM reviews the reports. It's not automatic.

That's how they claimed they were doing it, that it would be Report->Investigate->Silence/Not Silence based on said investigation.

What it's turned out to be is Report->Silence->Investigate->Repeal/Uphold.

There's a lot of irate forum posts over on their boards right now because of that.

[Dumbledoremd]

The pj salt will be real haha good i hope square dose introduce a similar system to help lower the rmt spam

[Greydor]

Or, even better, if they had a "silence" voting option in the vote kick window. There are some good players that shouldn't be on chat. Instead of kicking them, let's just keep the volume down.

Then again, I would be super sad if people voted to silence the raid bongo spammers during faceroll 24-man runs. I crack up at that.