Andre Engels wrote:
2006/3/19, Mark Williamson node.ue@gmail.com:
The current situation seems to be that all AOL users are blocked pending cooperation from AOL in trying to stop vandalism. This seems like a good approach to me.
Not to me. It's not the IP's task to micromanage the online behaviour of their users. And why single out AOL? Why not block ALL users until their ISP starts to cooperate?
We single out AOL because they are the only ISP we know of that categorically refuses to append XFF headers to proxy requests. Today, before I read this mailing list thread, I wrote this page on the subject:
http://meta.wikimedia.org/wiki/XFF_project
One of the ISPs on the trusted list (EscapeNet) is only there because I emailed them and asked them to change their proxy configuration. Their response was fast and friendly. My hope is that other ISPs will follow their lead, and that eventually the problem of shared IP addresses will be mostly isolated to AOL.
As for AOL itself: sad to say, but we might need to put up some entry barriers. Bug 550 is an extreme example of this: a feature which would let sysops restrict access to trusted users only. Email confirmation with an aol.com address required would be better.
-- Tim Starling