Blacklisted - Filtering Spam


For the last four days or so, my primary email server has been listed in SpamCop.net. The way you find out about such things is your email starts bouncing at some sites. As near as I can tell, getting listed in SpamCop.net is as easy as someone sending in a report. Getting out is a lot harder. SpamCop.net ages the reports and removes addresses after 48 hours of the most recent report, so the standard line is "stop spamming and in 48 hours your email will start working." Of course, that only works if you have any idea at all why you've been listed.

A number of companies, including Sento (I'm on the board) and even the State of Utah, where I was CIO, use blacklists like SpamCop.net to filter spam from their inbound mail stream. The problem is that SpamCop.net, by its own admission, and other blacklists are very aggressive in who they add to their list--for example, you can be added to SpamCop.net on a single, unsubstantiated report. The result is that if your enterprise uses a blacklist to filter email, you're likely throwing away quite a lot of legitimate email every day. A better strategy is to flag the email with a header and rely on downstream users to decide what to do with it.

Spam is on the verge of making my email communications unreliable enough that I no longer depend on them. It used to be that I trusted email explicitly because email was designed to deliver or return and error no matter what. That's no longer the case and I feel that I've lost an important tool. I don't blame companies who filter or blacklists--they're just a reaction. I can't even really blame the spammers since I think its too much to expect good behavior from certain people in a system that has no accountability. The problem is simply poor design. Until we fix some of the basic problems with email, everything else is just a band-aid.