If nobody has major objections to the proposed system design or the considerations in RFC3, I'm going to start working on it tomorrow afternoon, so that we at least have the platform to do it from. @Undo any more comments in particular on the system design would be awesome, since you know more Rails than I.
Just in regards to this - running it under your own account (Undo) would make the flags binding, no? That seems quite a bit more dangerous than simply running it under SD. Or do you propose to only run it on non-SO sites?
@DJMcMayhem we've pinged 'em several times over the past year with little response, so at the moment the plan is "they're unlikely to not trust us, so let's get going on a basic level and then bump its power upwards if they want us to".
@Undo A big part of that, I think, is because I showed up with stats showing that what I was doing was sane. Doing the same thing here, and augmenting that with metasmoke history, should make meta happy.
@Art update for future-you: Added the properties to User and Site (no assocs), created FlagSettings scaffold and got the permissions and tests and everything right. As some syntactic fun, you can do FlagSettings["flagging_enabled"] and it'll work how you'd think it would.
@Undo @ArtOfCode The misleading link filter I wrote is really bad at finding spam (almost all false positives), so I added it to FireAlarm and will remove it from Smokey.
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, bad keyword with email in body, blacklisted website in body, repeated URL at end of long post: SAP BPC ONLINE TRAINING COURSE by Rebeccan4411 on stackoverflow.com (@dorukayhan)
For a project I'm working on, it'd be nice to be able to retract flags through the API.
We allow people to retract flags from the web interface. Could this functionality be extended to the API? It could work such that you receive a 'Retract Flags' option on calls to /questions/{id}/flags/options...
@SmokeDetector I left a comment on there but I removed it because the troll probably likes to be noticed. It's likely that a quick glance will not reveal that this is a copy/paste of part of the existing answer.
[ SmokeDetector | MS ] Bad keyword in body, blacklisted website in body, link at end of body, pattern-matching website in body, repeated URL at end of long post: Gurus influenced this product? by mizzashelly on drupal.stackexchange.com
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in link text in body, bad keyword in title, bad keyword with email in body, blacklisted website in body, repeated URL at end of long post: TERADATA Online Training by tejawath12 on stackoverflow.com (@dorukayhan)
@Undo did you check with SE if this was OK to do: This is a way to funnel some rep to Smokey so it can hit the assoc bonus. Nothing to see here, just some sockpuppetry...? Feels like over the edge for me.
[ SmokeDetector | MS ] Bad keyword in body, link at end of body, pattern-matching website in body, repeated URL at end of long post: trim off a couple of pounds by nanagne on askubuntu.com (@ThomasWard)
tpu- by tripleee
[ SmokeDetector | MS ] URL-only title, bad keyword in body, bad keyword in title, link at end of body, pattern-matching website in body, pattern-matching website in title, repeated URL at end of long post: garciniasecretdietabout.com/slim5-fr/ by user631179 on askubuntu.com (@ThomasWard)
@Jan and @angus, I can't say I personally approve of bounties for rep transfer either, but he's right in that nobody actually gets any benefit out of this apart from Smokey hitting the assoc bonus.
which is kinda necessary if we want to flag things, since you need 15 rep.
Don't ever leave that to a bot owner's discretion ;-) (unless you want it to be allowed globally, in which case we'd need MSE to allow it. Or, you know, we could ask for forgiveness like we're going to
@JanDvorak I wasn't planning to leave that to bot owners; someone else can judge if it's legit or not. AFAIC, our reason is legit - we're doing a project that requires 15 rep globally to work out.
@JanDvorak I keep saying it, but: spammers don't care about us. They probably don't even know we exist. Spammers are people who sit at a computer and post some piece of text they've been given as many times in as many different places as possible so that they make as much $ as possible. They just don't have time to try to get around anti-spam measures; that's why checks we put in place two years ago still work.
Not them, either. They're attention-grabbers, not evil viciously intelligent block-evaders.
You can predict all you like, but I would happily bet money that the absolute worst that can come of this is someone posting a meta question to complain, which will promptly be closed as a duplicate of the one someone's already asked.
