- Essjay - wrote:
The problem is not what the users do from AOL IPs, it is the way AOL *assigns* IPs. Right now, if you are an AOL dial-up user, you will be assigned a new IP every 15 minutes, or every page load, whichever comes first. Literally *every time you load a page, you will have a new IP*. That makes blocking problematic AOL users *impossible*. By the time a sysop can discover and block a given address, the vandal has jumpted IPs at least once, potentially a dozen times. Additionally, no sooner than the block button is clicked, a legitimate AOL user with an account will be assigned that IP, and will be blocked. This is not as much of a problem on smaller projects, with fewer editors, fewer vandals, (and fewer AOL users as a result), but it is a serious problem on large projects like en.wikipedia, where there are thousands of AOL users.
That's certainly not the impression we gained when we researched this a couple of years ago. Has something changed recently? AOL's information on this subject certainly hasn't changed:
http://webmaster.info.aol.com/proxyinfo.html
My understanding on this subject is that in the US, AOL uses a proxy cluster with load balancing by a hash of the destination URL. Thus if you block an AOL IP which edited a particular article, you block all AOL edits to that particular article.
-- Tim Starling