We might have named it after tinned meat, but e-mail spam isn't much like what's sold in the store. It may be ubiquitous, but it certainly isn't dull. Computer spam is more like the gooey purple creature known as the Blob.

In the 1958 film that inexplicably launched Steve McQueen's career, the Blob terrorized the teenagers of a small town. It was a seemingly unstoppable mass of protoplasm that seeped under doors and flowed through window screens. It slimed its way through any filter.

The Blob was a public menace, and so is spam, which is why it has gotten the attention of Congress. In fact, recent votes to create an anti-spam registry have been unanimous. But governments alone can't stop a spam creature that oozes through every chink in our digital defenses.

So what do we do? I'm hearing from consultants and reading in the trade journals that some companies are considering doing away with external e-mail entirely. That's a huge leap, but it's being justified in the name of productivity, plenty of which is lost when workers find hundreds of spam messages on their computers each morning.

My own ratio of spam-to-legitimate mail is now more than 9-to-1 and climbing. So I took notice when Yahoo announced that it is developing a technology to fight spam by changing how the Internet authenticates the sender of a message. Called Domain Keys, the new software would be made available to the open-source community and would use its connections to percolate through the Internet.

Here's the idea: When a server sent an e-mail message, it would embed a private, secure "key" in the message header. When the receiving system got the message, it would check for the public key that is registered to the sending domain. Assuming that the public key can decrypt the private key, the e-mail is considered genuine and can be delivered. If not, then the message is blocked.

Authenticating a sender won't stop all spam, but it will make it easier to filter out the offending parties. The Viagra ads, the Nigerian scams, the financial come-ons are often from people whose names and addresses have been hijacked by spammers to serve their own purposes. Each would have to prove its legitimacy, and because so many are bogus, much spam would be stopped in its tracks.

Sure, it seems grandiose to reconfigure how the Internet handles mail. But Yahoo is no small company, and I think its concept has a chance to gain traction. The system Yahoo envisages would be free, after all, and the benefits would accrue to the entire user community rather than any one company. If Domain Keys can be implemented in enough places to show it's effective, then support for widening its reach will surely build.

And it's hard to see why a legitimate service would refuse to add a privacy key to its address under such a system. There would be nothing to lose (although, as some have already noted, encryption like this does demand more computer overhead, an issue we may have to face no matter what anti-spam solution we come up with). Compared with giving up e-mail entirely, working with Domain Keys seems a sensible solution that could slow the spam creature down.

Steve McQueen finally took care of the Blob in a memorable scene in which all the town's teenagers sprayed it with fire extinguishers. The Blob's secret was that it couldn't stand the cold. What is spam's secret weakness? Surely it's the deception needed to make it look legitimate. Expose that and spam just might wither, a hope that we can all share for a less junk-plagued 2004.