[attention conservation notice: LONG post ahead that is probably too linguistic-y for some people, and certainly not linguistic-y enough for others]
OK, so just to make certain all of my readers know what I'm doing here, I'll recapitulate the classes I'm taking:
- Introduction to Computational Linguistics (aka CompLing) This is being taught by Steven Bird, Ewan Klein (Ash's MA advisor) and Ed Loper. It's a really general introduction to tasks that are the bread and butter of a computational linguist's day: automatically tagging words in a text for their lexical category (e.g. noun/verb), parsing sentences into hierarchically organized representations (a lot like when you used to "diagram sentences" in elementary school), and stuff like that. It all takes place using NLTK, the Natural Language Toolkit, which is a series of modules that the lecturers developed for the Python programming language. Since I find the CompLing stuff interesting, and Python is my favourite programming language, you can well imagine that I'm enjoying this class.
- Poverty of the Stimulus, Machine Learning, and Language Acquisition (aka PovStim) This class is devoted to a careful examination of Chomsky's (in)famous "argument from the poverty of the stimulus", which is pretty much at the core of the claim that children come to the task of language acquisition equipped with at least a bit of innate knowledge about how languages work. The lecturers are Shalom Lappin, a smart (Canadian!) dude who does philosophy and some computational linguistics, and Alex Clark, a (REALLY) smart dude who does machine learning of language. Both of them are great pedagogically, and the class is definitely forcing me to at least revisit some assumptions that I've left unchecked for several years.
- Computational Psycholinguistics (aka CompPsychoLing...I know, these are pretty clever, eh?) Taught by Roger Levy, who looks like he's about four years old (and his hair's way better than that picture would make you think). This class is probably the toughest one for me, given that I'm almost completely unfamiliar with the relevant body of literature, and I'm having to double-check a fair amount of the stuff that the authors assume of their readership. I'm certainly learning a lot in this one, though...the best was a paper that I came across in the bibliography of one of the readings that takes a close look at the relation between theory construction and computational modeling, and whether/when modeling is useful. Fun stuff.
- Introduction to Simulation as a Tool in the Study of Language (aka Simulation Blah Blah Blah...yes, I actually call it that) Taught by Andy Wedel, who got a PhD in Molecular Biology, then decided Linguistics was more fun and went back to do a PhD in that. This is definitely my favourite class. Andy's a great guy...very approachable (he often has lunch with my group and hangs out with us at social events) and a great teacher, who's clearly passionate about his area of research. I'm seeing his name crop up a lot in the bibliographies of papers by people whom I respect, so I think he's (i) on to something, (ii) definitely a strong up-and-comer. The take-home message of this class is that computational simulations of a particular sort can show that a lot of properties that were assumed to be part of our internal grammar in fact emerge independently through successive generations of language transmission between adults and children, given biases in the way we perceive and produce words (e.g. certain kinds of sounds are harder to hear, and certainly articulatory combinations are harder to make). A fair amount of the reading is stuff that I've come across before (when I started getting interested in complex systems theory and the Santa Fe Institute), but this class is forcing me to actually read the literature carefully and with a critical eye, so it's all good.
OK, so that was perhaps unnecessarily long. In any case, I'm definitely loving every minute of my time here, notwithstanding the fact that there's quite a bit of work to do. In fact, I have almost 1000 (!!!) pages of reading this month, and it's not exactly light stuff. But whatever...that's what I'm here for, right? So I spend several hours of every day reading. Anyway, that's the academic side of Life here.
On the social side, there's been some stuff going on, as well:
Saturday, July 7th: The first of two institute "parties" scheduled for this month. It was at The 750, the only pub on campus. The place has a weird vibe to it...acoustic tiling and fluorescent lighting on the ceiling make it look like an office building that gets converted after everyone goes home for the day. Nonetheless, we had a good time...there was a band (of linguists...named Dead Tongues...seriously, that's not even groanworthy) and I got to hang out with some fun people and chat with Ash a bit.
Wednesday, July 11th: Impromptu trip to San Francisco with Dave, Bryan and Claire! I had a great time. We took the Caltrain to SF and got there just past lunch time. The first thing we did was head over to Chinatown and grab lunch at a place The Chinese Restaurant (what else?). It was actually not that great, but we were all a bit hungry/grouchy, so we didn't shop around much. After that we went down by the water and took a trolley/cablecar all the way across town to The Castro, which is the gay district. Having Dave & Bryan around made this fun...we went to a bookstore and a couple of shops that might have reduced a lesser man to tears. Heh...good times. After that we wandered around town some more looking for a place to eat, and finally settled on a Columbian/Salvadorean place that Bryan remembered from a previous trip. The food was outstanding and cheap, and we hung out until after closing time. On the way home, we missed the 10pm train and were forced (boohoo) to go to a nearby brew pub to kill some time until the midnight train. I had a pretty decent red.
More to follow...a bunch of us went to Palo Alto for a non-cafeteria-style dinner over the weekend and hung out a bit, and there was a party in the courtyard last night...stay tuned.