[mfyvie]

Originally Posted by nocte

Imagine this situation (we accutally have this issue):

There?s a Club with a public internet terminal. Many forum users log in there.

The Club?s owner has an account too as well as many more fake accounts to promote his club.

Sometimes the owner uses his own PC ? so we can be quite sure that in that case all multiple accounts are fake accounts.

Sometimes the owner uses the public PC ? in that case we can?t distinguish between normal users and the owner?s fake accounts.

So let?s imagine the owner clears the browser cache on his private computer and logs in again with several of his accounts. In this case we have to wait and watch again for a while till we can guess whether he?s using the public or the private PC (in other words: if we can ban all multiple accounts or if there may be normal users among them)

Did I make myself clear? When public computers are involved it may be very comfortable to have system data as well.

OK, I can see your point here. Problems like that are very hard to track down. Having the IP address or extra information about the PC might not help in a case like that - they might be identical installations. One thing you might like is the logging feature, if the accounts are being created or used in quick succession this is much easier to spot in a log than by searching for and examining the profiles of individual users.

I've extended a few extra things based on suggestions. IP address first on cookie and last on cookie is now reported, as well as each individual first / last IP for a user on a given cookie. The browser user agent string is stored and reported, as well as the list of accepted languages from that browser. Screen resolution is a little harder to capture and requires the use of javascript. I don't really want to go down that road at this stage just for that information.