Gruber had posted this afternoon about a great-sounding, open-source project titled StupidFilter meant to help filter out stupid comments on the web. From the StupidFilter site:
The solution we’re creating is simple: an open-source filter software that can detect rampant stupidity in written English. This will be accomplished with weighted Bayesian or similar analysis and some rules-based processing, similar to spam detection engines. The primary challenge inherent in our task is that stupidity is not a binary distinction, but rather a matter of degree. To this end, we’re collecting a ranked corpus of stupid text, gleaned from user comments on public websites and ranked on a five-point scale.
However, when I tried the demo out, I was very disappointed. I thought I’d start easy and enter:
First post!
Which is just about the dumbest comment I could think of which might appear online. The response from the online demo?
Text is not likely to be stupid.
Uh-huh. We clearly have disparate definitions of what constitutes stupidity. Good luck, guys. I am really rooting for this to work. If the trolls, flamers, and idiots know they’re being ignored, then they really might go away. We’re just not there yet if I still ever have to read “First post!”
So, Yikes! I thought I’d try and lob them another slow and soft pitch to see if I had jumped the gun with my two-word gimme. This text gave the same “not likely to be stupid” result:
You’re and idiot! I cannot believe that you’d ever agree with Bush and/or Obama! You should die you Nazi and/or hippie!
Could someone please give an example of what is stupid text, then?
You would not believe how tempted I was to leave a “First!” comment to this post. :-)
That’s an excellent project, and really hope it can be made to work. I tried “you suk!”, which was deemed “not likely to be stupid”, and “u suk!”, which was considered likely to be stupid. They’ve got a long way to go.