The Internet’s battle against itself has been hard fought and bloody. When spam threatened to destroy us, we racked our brains to cobble together filters that recognized the difference between ci4lis and Cicero, cutting spammers off at the ankles; when ads got on our nerves, we setup algorithms to block the javascript, thus freeing us from the oppression of the punching monkey. Whenever evil threatens to poison the cause of online discourse, technology has always been there to help tuck us in at night.

Now, computer scientists are turning their attention to the one last bastion of filth in our little world — the “LOL”. Gabriel Ortiz and a team out of Albuquerque are putting together a filter with the vaulted goal of removing stupidity from the Internet.

OMG!!! LOL

The idea seems almost too simple, you look at a comment and using some advanced pattern recognition techniques you filter out any with too many instances of what I affectionately refer to as AIMese (LOL, OMG, Rofl oh my!). Other things like extended repetitions of consonants (hehhhhh) and strings of incoherent prose would also trip the filter.

Besides the obvious question as to how to tell the difference between a sarcastic quip cast in the voice of internet stereotype and real “stupidity,” I think the deeper question is whether we should go so far as to legislate how people wish to express themselves online. Certainly, moderators have always had the final say when it comes to the content of comments, but when you start letting programmers who have no direct ties to the conversation dictate an appropriate level of “stupidity,” the line becomes a little hazy.

How many otherwise intelligent people use AIMese occasionally to shorten an otherwise poignant comment.

How many really stupid people run their polemics through grammar check a half-dozen times just to cultivate that perfect level of sanctimony.

It is almost impossible to tell the difference between intelligence and stupidity just by looking at the language. Without understanding intent, a filter like this seems to be asking for more trouble than its worth.

Web 2.0 Roundup

Lots of people are behind this idea, and even I really want to see what they come up with. My only warning is that anytime you want to use an algorithm to help decide something as hazy as intelligent discourse, you’re walking a fine line.

Who knows how many LOLs Shakespeare may have used if he had a Slashdot account. What do you think?

[Be sure to subscribe to the RSS feed before leaving. Photo Credit]

If you enjoyed that why not find a job or read our guide to working in the 21st century. You can also join our Kiva team or hire me for your project.