Spam filtering for dummies

MessageLabs is evil.

Not really. I’m sure most of the people who work there, and even most of the people who direct the company are very nice human beings who treat their families well, don’t kick dogs, and host backyard barbecues.

Nonetheless.

As many Advice Line readers know, I also send out an independent e-letter once a week called Keep the Joint Running (KJR). It goes to a list of voluntary subscribers only. It is, from a spam perspective, entirely kosher.

It doesn’t matter. So far as I can tell, something about it has flagged my ISP’s SMTP server as a spam source.

That, by itself, would be okay – these things happen, and there’s no definitive test of an e-mail message that clearly separates spam from legitimate messages.

Here’s what isn’t okay: There is no way to contact MessageLabs to discuss the problem, inform them that they’ve made a mistake, and get the flag cleared.

MessageLabs is accessible to its subscribers. It isn’t accessible to its victims. Period.

I’m seriously tempted to have my attorneys decide whether identifying my e-mails as spam when they aren’t constitutes libel. I won’t do it, of course – the cost of pursuing the complaint would exceed anything I’d get out of it by a huge multiple.

Except, of course, for the satisfaction I’d get.

I don’t have any advice to give about this, other than to explore alternatives to MessageLabs if you’re looking for a commercial spam identification system.

Not that MessageLabs will fail to identify spam. Quite the opposite. I figure, their system is easy. If you flag everything, some of what you flag will be spam.

Sorry to sound so grumpy about this. Mostly, it’s because I’m pretty grumpy about it.

– Bob

Powered by ScribeFire.