Context-free journalism

By Rachel Campbell

Friday, April 15, 2005 12:02 PM CDT

I will be drinking umbrella-laden girl drinks on a Gulfside beach as of next Thursday, so I've been wondering what to do with my column while I'm, uh, on sabbatical. But I think I've found the perfect solution with SCIgen, "an automatic CS paper generator."

The latest in "context-free grammar" technology, SCIgen is a program written by three MIT students protesting conference standards for the (sometimes questionable) papers accepted for presentation. What it produces sounds something like the paint-by-numbers press releases we at the JT have a tendency to receive; only instead of "synergistic" and "opportunistic" and all the other meaningless but really big words publicists love to use, you get something like, "The notion that researchers agree with the visualization of superblocks is largely considered significant." In other words, gibberish. But really really really important-looking gibberish; and, after all, looks are everything.

And should I use SCIgen to virtually write my columns while I'm, um, researching ... stuff, I can be sure the technology is strong enough for a prank, but pH-balanced for credibility: One of the faux papers the MIT students submitted to the World Multi-Conference on Systemics, Cybernetics and Informatics was actually accepted for formal presentation.

MIT grad students Dan Aguayo, Max Krohn, and Jeremy Stribling submitted "Rooter: A Methodology for the Typical Unification of Access Points and Redundancy" and "The Influence of Probabilistic Methodologies on Networking" to, according to Reuters, "(target) WMSCI because it is notorious within the field of computer science for sending copious e-mails that solicit admissions to the conference." The papers were protesting, as SCIgen's Web site puts it, "fake conferences" that "exist only to make money;" but the protest turned into a punk when "Rooter" was accepted and the three authors, so to speak, were invited to pay several hundred dollars in order to present their fake findings.

And those crazy kids have accepted with gusto and, unfortunately, their grad students' incomes. Soliciting donations in order to give "a completely randomly-generated talk, delivered entirely with a straight face," the SCIgen Web site (located at http://www.pdos.lcs.mit.edu/scigen for your lethargic Friday worktime convenience) breaks down their projected budget thusly: "Conference registration fee (2): $780. ... 1 hotel room (2 nights): $260. Estimated airfare from Boston to Orlando (1): $170 (someone has already pledged flyer miles for one). Food, ground transportation, etc.: $100. Randomly-generated talk: Priceless."

But you can put away your checkbook, because they've already garnered $2,311.09. That's $1,001.09 more than their frugal budget requires, so rest assured that our intrepid academics will be able to eat something other than Ramen noodles for once in their lives.

The only problem in terms of my upcoming leave of absence is that I don't write in computer scientish. But then again, when I plugged my name into the paper generator, it spat out the following: "The Effect of Robust Communication on Omniscient E-Voting Technology." I think I wrote about that back in October. This may work out after all.

Online columnist Rachel Campbell, who isn't quite sure what being "on sabbatical" means but thinks that it sounds pretty cool, can be reached via e-mail at rachel.campbell@lee.net.

Special Offer: Get 5 Weeks of the Journal Times for $7!