Langa Letter: Silent Censorship

Beware this column! No, there's nothing scandalous in Fred Langa's advice, although some poorly-made E-mail and content filters would think so.
Good Intent, Bad Practice
The problem with content filters doesn't lie in the concept or with those who use them. Most people who install and use filters are entirely well intentioned.

At the corporate level, businesses want to keep employees on track, without the distractions of inappropriate E-mails and sites. They also want to prevent distribution of material that may be offensive or deemed harassing. At home, parents want to try to keep kids from being exposed to the raunchier aspects of online life.

These are admirable goals. But the examples above show that most content filtering simply doesn't work. In fact, I regard many commercial content filters as near scams. They work so crudely that whole swaths of legitimate and utterly benign language, whole chunks of the dictionary, are deemed offensive. That's nuts. But worse, many of these tools are so easy to defeat, through nonstandard punctuation or spelling, that anything (no matter how vile or inappropriate) can sail through the filter without a hitch.

The net result is backwards. Many filters represent a technology where the bad stuff can get through, while the good stuff is filtered out! How wrong can something be? How utterly useless? How silly? How counterproductive?

I know I'm not alone in this feeling: See, for example, The Electronic Freedom Foundation's Censorware site, especially the section halfway down the page on "Why Blocking Technology Can't Work."

Are Some Filters Better Than Others?
Lest I be accused of painting with too broad a brush, let me state that not all filters are created equal, and some may be less bad than others. For example, reader Matt Whitenack (who works at an ISP) found a high-end filtering tool aimed mostly at businesses that seems to do a much better job.

I agree, there are a lot of poor content filtering programs out there... [But] we've found a service called Postini, which does the filtering for us. However, instead of just filtering the mail and sending a bounce message, it allows our users to view the mail that was filtered and then decide whether or not to accept it. In the time I've spent researching these services, both server side and outsourcing, this is the only one that allows the users to control the level of filtering, or simply turn it off completely. Oh, the other plus to this is that it doesn't cost an arm and a leg.

At the end-user level, other readers have recommended PocoMail, MailWasher, Junk Spy, SpamWeasel, SpamCop, Email 911, and other content/spam filters, services, and tools.

If you must use a filter, it makes sense to try to use one of the better ones. But I still have yet to find one that I feel is worthwhile.

One reader told me that my problem was that my priorities were all wrong, and that we all should accept filters, even with all their flaws and limits, because some filter might protect some children, somewhere, someday. I disagree, because I think content filters do far more harm than good.

What's your filtering experience? Where do you weigh in on the subject? Do you use filters, and if so, which ones have you found good or bad? Do you use tools to circumvent filters? Join in the discussion!

Related InformationWeek Articles