There are many unpleasant new developments in the world of the Internet to bemoan, but the growing prevalence of spam filters is not one of them. We installed one of the best, SpamAssassin, here at Salon internally earlier this year, and it is a godsend. More recently we installed it on the mail server at the Well so users of Well e-mail accounts could benefit from SpamAssassin’s capabilities.
So why is online-news pundit Steve Outing complaining about spam filters? Outing suggests that spam filters will somehow censor content on the Net because people will avoid using the controversial words that “trigger” the filters. He writes from the perspective of an e-mail newsletter publisher who’s worried that his product is being improperly blocked by the filters, and singles out SpamAssassin as the main offender.
The trouble is, he misunderstands the way SpamAssassin is installed by people who know what they’re doing. One of the great things about it is that it doesn’t automatically delete spam; the way we use it, it tags incoming e-mails as probable spam. The user can then use his e-mail client to filter these probable spams into a separate mailbox for review and deletion. This is, in fact, the way the Well uses it too (Outing misreports this).
A bad spam filter can indeed raise the danger of “false positives” — filtering out e-mail that you wanted to receive. But in truth, there’s little danger of Outing’s newsletter, or anyone else’s, being invisibly trashed by SpamAssassin. You can “whitelist” mail from any recipient you want — basically telling the filter, “Don’t tag this person’s mail as spam, no matter what.”
SpamAssassin isn’t perfect, but it’s a step up the evolutionary ladder. It regularly sifts out hundreds of spams a day from my inbox. And after the first day’s fine-tuning, it hasn’t delivered a single “false positive.” I’m sorry to see it unfairly maligned.
Post Revisions:
There are no revisions for this post.