From reader JD in Montreal:
Archive for HLT
This speaks for itself, I think — the first referral from the new Cuil search engine that I've noticed in our referrer logs:
(Click on the image for a larger version.)
Here on Language Log we have recently been twitted by readers who believe we have been insufficiently attentive to celebrity linguists, in particular Noam Chomsky (I find the idea that we should choose topics for our postings using personal fame as a guiding metric bizarre, but there it is). Mark Liberman has now responded, linking to a frivolous Facebook group pitting Chomsky against Labov. We've been frivolous about Chomsky before, in a posting about Ali G's interview of him; in two postings about Chomsky as the object of sexual arousal; and in a posting quoting Woody Allen's "The Whore of Mensa".
But (as Bruce Webster suggested to me) we seem not to have discussed the famous Chomskybot, which has been around in one version or another for about twenty years.
Powerset is a search engine that allows users to express their queries as phrases, rather than a few keywords. It uses natural language processing (NLP) technologies to analyze the verb-argument structure of a query and deliver more focused search results, initially just from Wikipedia. Powerset has attracted interest from the NLP community, as its services promise to demonstrate the value of NLP – and of language analysis more generally – in extracting information from the trillion or more words of text on the web. On Tuesday, Microsoft announced it has acquired Powerset, and that Powerset will become part of Microsoft's Search Relevance team. I hope this takeover means that natural language search will become mainstream, scaled up to the entire web, and used far more widely than before [Powerset blog|Microsoft Live Search blog].
I'm in Marrakech for LREC2008, and so I should be blogging about all the interesting things going on here. But instead, at least today, I'm going to post the first installment of something that's been on my to-blog list since early April.
At the P.I. meeting of the DARPA GALE program, there was a panel discussion, chaired by Philip Resnik, on on the topic of Chinese-English machine translation. Or rather, according to Philip's introductory slide, on the topic "What the @#$% Do We Do About Chinese-English MT?"