Spam now comprises an estimated 46% of all e-mail, straining mail servers and users' patience alike. The Chronicle of Higher Education published an article last week (subscribers only, alas) describing spam-fighting strategies under consideration at major universities. Note the phrase "under consideration." It's well known that university faculty and administrators naturally wreathe themselves into Rube Goldbergian committees that ponder, debate, massage, and complicate such policy issues until, as with Dickens' Jarndyce and Jarndyce, nobody remains who remembers the original point. But the Chronicle did manage to extract some pre-release beta versions of several universities' anti-spam strategies, and publishes them in its article. Stand by to watch academic bureaucracy in action:

Strategy #1: Shutting down e-mail open relays on campus servers to prevent spammers from hijacking the machines.

Ah-ha! Administrative ploy number one: propose as an innovative solution something that should have happened years ago. Of course, this doesn't address the problem itself -- very little campus spam actually originates on campus servers, most of it arriving via insecure machines in Asia and South America -- but offers the pretence of action. Which, in academic circles, is generally indistinguishable from the real thing.

Strategy #2. Directing students with spam complaints to a campus Web site with answers to frequently asked questions and articles about how to avoid spam.

Administrative ploy number two: Ignore the fact that excellent informational sites on this topic already exist outside the academic system (Abuse.net and CAUCE.org are only two of the best) and set about reinventing the wheel. The ensuing site will be pretty but vacuous, useful only as evidence of administrative "concern." And everyone knows that "We are very concerned about this problem" rarely means "We will do something about this problem."

Strategy #3. Offering seminars each semester on how to use the anti-spam filters that are built into some desktop e-mail programs.

Only a true adminocrat could come up with this one. Seminars to teach students and faculty how to use Eudora's email filter! Naturally, this strategy fails to mention that most desktop email programs have weak, inflexible filters that can't effectively fight spam. Not to mention that desktop email programs filter spam on the desktop, meaning that the spam first has to be downloaded to the desktop, meaning that spam still creates bandwidth congestion, fills the /var/mail spool, and stops real mail from getting through to overflowing inboxes. Tutorials on procmail, a powerful server-side mail filter, are presumably not in order.

Strategy #4. Setting limited blocking filters on the campus-mail gateway.

This is a controversial one. The Chronicle reports: "Some college officials say that concerns about violating the principles of academic freedom, privacy, and the First Amendment make them reluctant to block e-mail messages based on their content or to 'blacklist' the Web sites of known spammers." How strange that college officials, who care not a whit for academic freedom, privacy, and the First Amendment in their relentless pursuit of ideological and intellectual homogeneity, now invoke these principles to protect the rights of spammers. Seemingly, administrators believe that students should see Viagra ads, debt consolidation scams, and pictures of 26-year-old horny teen virgins -- but not the works of enormously important conservative and libertarian intellectuals. One must have some standards, after all.

Strategy #5. Closing down individual campus e-mail accounts, if requested, to put an end to spam attacks.

Surely this is a brainwave. Closing down e-mail accounts is a surefire way to stop spam. But open a new account for the same person and the spam flood will begin again in weeks, if not days (the mailto: links liberally scattered throughout campus Web sites are handy targets for spammers' automated address-collection devices). Add to this the fact that many people will be unwilling to surrender an email address that they've held for years, and you have something that looks like a solution, but really is not.

To engineer a totally spam-free campus, maybe universities should just eschew email entirely and go back to circulating typed memos. Possibly some adventurous and technologically progressive schools could even switch to carrier fowl, thus flavoring their campuses with the spirit of Harry Potter. After all, that first wave of Potterholics will be ready for college in just a few short years and will expect message delivery via owl.

Strategy #6. Installing a firewall to block spammers from searching campus servers for e-mail open relays.

Ah yes -- another "innovative" proposal that should have been implemented years ago. Strategy #6 is really a complement to #1, and will be equally ineffective at fighting the 99.8% of spam that originates from non-university systems.

Strategy #7. Offering an alternative "filtered" mail service, in addition to regular campus e-mail, for faculty and staff members and students who want to avoid spam.

Creating "filtered" accounts in addition to regular campus mail will surely create administrative havoc. Most academics of my acquaintance can barely cope with one email account, never mind several. This option seems cumbersome at best, hopelessly unworkable at worst.

I have an e-mail account with Panix.com, a truly excellent New York City internet service provider. To help its users fight spam, Panix has installed on its servers the spam-tagging program spamassassin and the server-side mail filtering system procmail; Panix customers can then deploy these powerful tools in any manner they might choose. For example, one might choose to live with the spam and leave one's account entirely open; or to filter spam into a separate "junk mail" folder and sort through it later in case some "real" mail got wrongly tagged; or simply to kill all spamassassin-flagged messages before they even enter the mail spool.

In other words, Panix hasn't tried to centralize its spam-prevention efforts. It hasn't tried to make executive adminstrative decisions about What Should Be Done. Instead, it has put powerful, flexible tools in the hands of users and lets users do the work of configuring their own accounts per their individual preferences. Ultimately, the difference here is between a leftist university system grounded in the ethos of central planning and a libertarian computing culture that naturally inclines toward personal responsibility and individual, decentralized action. Having tailored spamassassin and procmail to my needs and preferences, I get almost no spam. My academic friends, mired in debates about the ethics of centralized spam-filtering, receive so much junk mail that their accounts are fast becoming unusable. Who's winning here?