Detecting registration bots

[kermit2]

I use a few lines of code to moderate any first posts which contain a URL. Since the user's post count isn't incremented, if they then make a second post containing a URL, that too is moderated. Sometimes there are false positives, but approving those gives my mods something to do. Obviously the spammers could easily get round it, but the majority just move on to another forum.

Suppose you could do something similar to automatically moderate users with a birth date of 1st Jan 1980, and accept that there are going to be some false positives. Presumably this is only the signature of one particular spammer though.

Or you could just firewall off China