Water-powered cars and grammar checkers

« previous post | next post »

In the 13 September NewScientist's "Feedback" column: a note beginning "There should be a law against it, we grumble", with a report that back in June Reuters distributed a story on the Japanese company Genepax, which claims to have produced a car that runs on "nothing but water". The magazine noted that the claim has been debunked a number of times over the past few years, but keeps re-surfacing. A possible remedy:

We thought for a moment we had a way of stemming the tide of water stories. Surely those clever people who write word-processor programs that put annoying green wiggles under our sentences with notes like "the grammatical passive voice has been used" [nice deployment of the passive!] could add a feature that crosses sentences out in red with the note "this does not happen in the real world". Shouldn't that feature be made mandatory in news organisations?

But then we remembered that when Microsoft tells us off about our grammar we invariably click on "Ignore rule" and proceeed blithely on. Back to the drawing board…

Interesting that the writer takes seriously — or affects to take seriously — the possibility that software could check facts, and gives up on the idea only because users would disregard the software's warnings.

Now, you could devise a simple-minded "fact checking" program that would do a string search, for strings that would have to be hand-entered for each meme that comes down the pike. For some memes, it's hard to see how to refine this search satisfactorily. How, for example, to pick out the meme (discussed many times here on Language Log) that women talk much more than men, without pulling up lots of other stuff involving women, men, and speaking?

In any case, such software would inevitably catch discussions of the meme, including debunkings.

[Other recent postings on language-checking software (other than spam filters and simple spellcheckers): on the spoof software apostrophree and on the software-in-development SpinSpotter (most recently posted about here, with links back to earlier postings).

And now Hamish Ramsay writes about StupidFilter, a proposed attack on idiocy on the web:

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.

Mark Morford wrote about StupidFilter in his SFGate column a few days ago, in connection with the YouTube Comment Snob. Morford takes the creators of StupidFilter to be entirely serious.]


Comments are closed.