it’s a tough life
yesterday i gave my “spotlight presentation” (a 90-second presentation to convince people to visit my poster) and then stood with my poster for a couple of hours, giving my spiel and answering questions. it went well; there were always a couple people in front of my poster, so that was good.
after the poster session, we had the conference banquet. there was champagne, wine aplenty, and quite good food. we also heard an invited talk by david mackay, inventor of dasher, a pretty cool piece of software that allows you to enter text without a keyboard — he also claims that if you have control of any muscle in your body (or indeed, a single neuron), you will be able to enter text using dasher. the demonstrations he made backed up at least the first (i.e. serious) part of that claim.
the thing about dasher is that it makes it easier to enter predictable text (this is to allow high-speed input despite the lack of keyboard and is a Good Idea). this fact led several of us into a quite interesting discussion about whether poetry is harmed if it is typed in using dasher, and whether the training text (the text that is used to decide what is predictable) matters. the predictions that the software makes are based on the previous 5 letters. thus, if you have typed “evident” the “ly” (”evidently”) and “ity” and “ify” (”identity” and “identify”) paths will all be large.
but how much is semantics guided by syntax? will the fact that certain paths become easier to enter affect what people write? will it subtly change their ideas in the transfer from mind to paper?
today’s the last day of the conference, and people are starting to come in for ijcai, another conference that will be held here. i’m meeting up with some michigan folk tonight. tomorrow’s a free day, and sunday i leave.