Alright, so there's a few points I'd like to address.
First off, Blockland is largely a kids game. Many of the older members here have either been here for a long time or were brought in by friends, I believe it's rare that an adult goes out and just buys this game. So, since we can categorize Blockland as for the most part a kids game, I'd like to avoid those sort of discriminatory terms in a
public chatroom. My personal belief is that there's better way to say things and you don't need to use that terminology, but I understand that's unfortunately not a shared belief. For this reason, you're still able to say whatever you want in direct messages. We have no interest what users do and we're not trying to censor anyone, however it's not appropriate to foster that language and behavior in the public chatroom of a game full of children. Discrimination filtering is not going away.
We've had some internal debate about the best way to go about handling this, of course. I've stood by a reactive system (letting the message go through and then handling it) as opposed to a proactive system for most of this. If someone's intention is to use discriminatory language, a popup saying "You can't say that!" won't stop them. They'll repeatedly try different permutations until the message goes through. However, if we allow the message to go through and then apply the 5 minute ban, the same damage has been done but we were able to catch it and stop it. Hard filters just offer a (not so difficult) challenge to work around.
GlassBot isn't meant to be a substitute for actual moderators, but instead just to aid some trivial cases. We can't expect moderators to sit around watching a not-so-active chat 24/7. Typically if you're running Blockland, you're doing so to play the game and not to babysit some indecent people in a chatroom. There's also only a small handful of people that I'd consider responsible enough to moderate, and those players coincidentally rarely actually play Blockland. Moderation does not fall to a bot, a bot just takes care of the easy cases.
We do have a rules command. A server update was issued yesterday with some adjusted rules and thresholds for GlassBot.
A summary of what GlassBot does:
- GlassBot issues warnings. First warning is a 5 second mute, second is 30, third is 60, fourth is a kick, and everything past 5 warnings is exponentially increasing ban times. Warnings decay over time.
- Any message that is longer than 5 characters and more than 75% caps is considered yelling
- Any word that is more than
60% 80% similar to a discriminatory word is considered a discriminatory word. (Similar meaning that the sent word fully contains a discriminatory word from a list we've put in) - Sending more than 5 messages in a 5 second span receives a "typing too fast" warning
- Repeating the same message 3 times in the past 5 messages within 60 seconds receives a "stop repeating yourself" warning
We've tweaked these since the launch of Glass Live and these seem to be the most appropriate values thus far. I'm open to some more tweaks given justification, but really these are simple guidelines for a chatroom. Any normal conversation should not be impeded by
any of these limitations, and if the way you communicate does you may wish to revise how you talk to people.
