3 ms·
I was actually wondering how to detection works. Just checking if all users have the same email domain name (with the exception of a white list such as gmail,
by thecodemonkey 13y ago
I was actually wondering how to detection works.
Just checking if all users have the same email domain name (with the exception of a white list such as gmail, hotmail, etc.)?
Or flagging users that are using mailinator/5minutemail/whatever?
- Amadou 13y agoHe's probably got a handful of tools in the bag, but the first on my list would be to cross-reference IP address with invites. If one ip address signs up all/most of the invites then that is a pretty big clue.
- wpietri 13y agoThat's the kind of thing you don't want to post about. Abusers basically work as hill-climbing algorithms. If you catch them, many will just try again after tweaking something. When in the past I've fought abusers my approach includes: minimizing opportunities to gain information; minimizing signs of progress; lengthening feedback loops as much as possible; increasing the cost of trying to get information; and, when you change things, changing several things at once, so that they have to figure out several things to get back in.