I still don't understand why there was pushback from the website devs over the obvious solution of some form of Turing Test (with Captcha or it's derivatives being the most common form). I suppose some dedicated spammers will still manually create logins and then quickly flagging/deleting their posts and disabling the logins will be the main line of attack. But the VAST majority of spam posts are undoubtedly from new logins created automatically. If they have a backlog of logins accumulated then again we will just have to find and disable them as they become activated.