Artificial Stupidity

(For the original article, in TIMES ONLINE, click the title above)

PAGES of computer-generated gibberish, containing such gems as “contrarily, the lookaside buffer might not be the panacea”, have been accepted as an academic paper at a scientific conference in the United States in a victory for hoaxers.

Convinced that many scientific conferences would accept almost any research for the right fee, three students at the Massachusetts Institute of Technology celebrated yesterday the submission of their gobbledegook masterpiece, Rooter: A Methodology for the Typical Unification of Access Points and Redundancy.

Jeremy Stribling, one of the students, said that he and two graduates were convinced that many academic conferences had few or no minimum standards because their sole purpose was to make money. “We decided to test the limits,” he said.

They wrote a computer program to generate nonsensical research papers, complete with “context-free grammar”, charts and diagrams. The program randomly selects and assembles sentences, then drops in impressive-sounding verbs and nouns. “Many scholars would agree that, had it not been for active networks, the simulation of Lamport clocks might never have occurred,” the paper asserts in its introduction.

“Certainly, the usual methods for the emulation of Smalltalk that paved the way for the investigation of rasterization do not apply in this area.” The students submitted Rooter, and a second paper, to the ninth World Multi-Conference on Systematics, Cybernetics and Informatics.

Mr Stribling said that they targeted the conference because it is notorious for sending e-mails to solicit admissions. An accepted paper usually attracts a fee. Nagib Callaos, a conference organiser, said that the paper was taken on a “non-reviewed” basis — meaning that there had been no feedback .

The students have raised more than $2,000 (£1,060) over the internet so they can attend the conference and give, as Mr Stribling said, “a completely randomly generated talk, delivered entirely with a straight face”.

An exercise in academic deceit
# We ran four novel experiments:

(1) we dogfooded our method on our own desktop machines, paying particular attention to USB key throughput

(2) we compared throughput on the Microsoft Windows Longhorn, Ultrix and Microsoft Windows 2000 operating systems

(3) we deployed 64 PDP 11s across the Internet network, and tested our Byzantine fault tolerance accordingly and

(4) we ran 18 trials with a simulated WHOIS workload, and compared results to our courseware simulation

Taken from Rooter: A Methodology for the Typical Unification of Access Points and Redundancy