Gnifty

P.S. - If you think this will stop spammers, it won't. They will invite people to groups, they will send ingame mail, they will get creative. What's next, an "/EVERYTHING OFF" toggle?

This detracts from the game in a lot of ways for a lot of people, but the purpose of my post is not to argue the validity of people's desire to have a properly functioning chat system where you can actually send other people tells that aren't already in your existing circle(s). If you think it's stupid to be upset by this implementation, then let's just agree to disagree and call it good. The reason I'm here is that I feel I should post, again, with a basic introduction on how the gaming industry handles this problem, it's not that difficult to do; though it is more difficult than writing the "turn chat off" toggle that we now have. Pasted from this thread: forum.elysium-project.org/topic/29928-please-fix-this-flawed-chat-system/ "Blocking hackers/cheaters/spammers is not the answer, if you show them a wall they will climb it. Subterfuge is the only way to get ahead in this battle. Think back to how blizzard used to do this, they'd analyze data and ban in large batches. They did it that way for a reason. Allow me to explain. Let's take chat muting for example, they can quickly figure out what they did to trigger this, and then they adjust their behavior to get around it. Before you had X accounts spamming Y messages, and then after rudimentary blocking logic you'll likely end up with something like (X*10) accounts each spamming (Y/10) messages. They'll spin up more VM's, utilize more proxies, run more accounts, whatever they can do. That's a crude example, but you get the idea. You can never win this battle without severely crippling your legitimate players. I speak from experience as a serverside game dev, so I'll explain a bit about some better methodology. A basic example of a more effective approach is to let their whispers go through, but only show the whisper going through for them, don't actually send it to the recipient. You can further improve upon this by only letting it go through for the recipient if they're in the same IP subnet or some such, so that it's harder for them to detect if they test it with 2 of their accounts. It's pretty easy to eliminate false positives with detection logic if you aren't also trying to block them, because they get quite cavalier if they aren't aware they've been caught. It gets more complicated than this though, because eventually they figure out your shinanigans; after a while they wonder why noone has bothered banning them. You need to keep the status quo, send them on wild goose chases, make them think that they're fighting the good fight and winning. Ban them once in a while, mute them once in a while, and try not to make the pattern obvious. After a while, these guys end up wasting more time getting nowhere than you do dealing with them. Don't get me wrong, this isn't an IWIN button, but it's the only hope you have of stopping them without breaking rule #1: don't screw over your legit players." You don't have to take my word for it, do your own research, learn from what others have done. Quit trying to reinvent the wheel (poorly).

Blocking hackers/cheaters/spammers is not the answer, if you show them a wall they will climb it. Subterfuge is the only way to get ahead in this battle. Think back to how blizzard used to do this, they'd analyze data and ban in large batches. They did it that way for a reason. Allow me to explain. Let's take chat muting for example, they can quickly figure out what they did to trigger this, and then they adjust their behavior to get around it. Before you had X accounts spamming Y messages, and then after rudimentary blocking logic you'll likely end up with something like (X*10) accounts each spamming (Y/10) messages. They'll spin up more VM's, utilize more proxies, run more accounts, whatever they can do. That's a crude example, but you get the idea. You can never win this battle without severely crippling your legitimate players. I speak from experience as a serverside game dev, so I'll explain a bit about some better methodology. A basic example of a more effective approach is to let their whispers go through, but only show the whisper going through for them, don't actually send it to the recipient. You can further improve upon this by only letting it go through for the recipient if they're in the same IP subnet or some such, so that it's harder for them to detect if they test it with 2 of their accounts. It's pretty easy to eliminate false positives with detection logic if you aren't also trying to block them, because they get quite cavalier if they aren't aware they've been caught. It gets more complicated than this though, because eventually they figure out your shinanigans; after a while they wonder why noone has bothered banning them. You need to keep the status quo, send them on wild goose chases, make them think that they're fighting the good fight and winning. Ban them once in a while, mute them once in a while, and try not to make the pattern obvious. After a while, these guys end up wasting more time getting nowhere than you do dealing with them. Don't get me wrong, this isn't an IWIN button, but it's the only hope you have of stopping them without breaking rule #1: don't screw over your legit players.