2 ms·
Yeah, which indicates what a terrible false-positive this is. It's like their spam detection is just pattern/behavior based, and entirely unintelligent, censor
by sneak 7y ago
Yeah, which indicates what a terrible false-positive this is. It's like their spam detection is just pattern/behavior based, and entirely unintelligent, censoring even authorized users doing normal things in projects they participate in.
- shadowgovt 7y agoI wouldn't call it a terrible false-positive. The Bayesian estimate on this behavior is that it's far more likely to come from a porn-bot than a legitimate use case. (Arguably, the use-case in question isn't even what the service would consider "legitimate;" user was trying to tickle the tiger's tail on purpose. If one doesn't want to get kicked off like a porn-bot, it's not hard to avoid acting like a porn-bot).