SPLPAC

The Society for the Promotion of Long Prepositions,Adverbs and Conjunctions wishes, henceforward, to exist, notwithstanding its lack of positive ontological status heretofore. Moreover. it regrets and plans to remedy its previous delinquencies in this area, but nevertheless accepts that its existence may not continue for long. Contrariwise, it sees itself as a lexical mayfly skittering over the surface of the language, and is OK with that. Anyone know where the nectar is?

Are you a linguist interested in taking CS classes at OSU?

If so, I have two options for you:

Option 1

If you want to prove that you know some computer science and to demonstrate the ability to do your own programming (all scientists should want these skills), then take the

Graduate Interdisciplinary Specialization/Minor in Applied Software Engineering

which requires you to take:

1) CSE 502: Object-Oriented Programming for Engineers and Scientists (can be waived if you have some programming experience)

and

CSE 688: Applied Component-Based Programming for Engineers and Scientists

2) One of:

a) CSE 767 Applied Use-Case-Driven OOAD for Engineers and Scientists

b) CSE/ECE 794R: Applied Enterprise Distributed Computing for Engineers and Scientists

3) One or more other elective courses to make up to 15 credit hours:

I strongly recommend that linguists taking this path choose their electives as CSE 630: Survey of Artificial Intelligence I and CSE 680: Introduction to Analysis of Algorithms and Data Structures.

Everything about this strikes me as recommendable. If you have the choice, maybe you should do 2b in preference to 2a.

Option 2: If you want to compete for postdocs and other jobs that need demonstrable CS training, Masters' degree, Research track is your only sensible option.

http://www.cse.ohio-state.edu/grad/ms.shtml

This is a much bigger commitment of time and effort, but gets you an extra and very  solid academic qualification.

It is exceptionally hard to do this unless you have the energy, time and funding to devote an extra year or more to your studies. The 7xx classes in the Masters program are difficult. We are talking here about classes that substantial numbers of well-prepared computer scientists are prepared to fund themselves to do, because they judge that it will improve their earning potential. They are correct, but both you and they should expect to work very hard to realize the potential.

Similar options will be available once we move to semesters.

The Well-Designed Child

John McCarthy (who, in 1960, wrote the first paper on Lisp), has written a provocative piece called "The Well-Designed Child". It's been published in Artificial Intelligence recently (see below), but seems to have been mostly complete by 1997. The point of the piece is to advocate for a kind of common-sense nativism in understanding human and animal capabilities. McCarthy points out that the world has a number of properties that the well-designed child would not learn ab initio. These include object persistence, the tendency for objects to continue exist even when unseen, and the consequences of gravity, which make it reasonable, for many purposes, to conceptualize the world as mostly two-dimensional.  In the same way, he says that the human bias to perceive objects in terms of natural kinds is such a powerful structuring device that evolution ought to build it in as an assumption rather than requiring each child to learn its effectiveness anew.

For linguists, this article could be seen as a contribution to the debate about linguistic nativism: whether the human child has an inbuilt language faculty. But McCarthy touches on that only in passing, being much more concerned to demolish the tabula rasa hypothesis as a basis for the design of child-like robots, and to rehearse some claims about what the language of thought might be like.

The thing I really like about this article is the way it emphasizes the continuity between human capabilities and those of other animals. According to McCarthy, many of the things that make us human are similar to or even identical with the things that make dogs canine, cats feline, mice murine and octopuses octopoidal (although we would expect sea creatures, who live in a genuine 3D environment,  to be out of line with us on the stuff about gravity). And those things are largely about the styles of conceptualization which allow organisms to be effective in the particular world in which we find ourselves.

References

John McCarthy, The well-designed child, Artificial Intelligence, Volume 172, Issue 18, Special Review Issue, December 2008, Pages 2003-2014, ISSN 0004-3702, DOI: 10.1016/j.artint.2008.10.001.
(http://www.sciencedirect.com/science/article/B6TYF-4TMJ41M-2/2/6076a1f21080a46b5bb52900faf763c7)

Alice in Wonderland

Well, I liked it. I'm glad we didn't get discouraged by the reviews and go and see "Up in the air" instead. My favorite character, the creepy (though "good") White Queen.



Tim Burton was not at all faithful to the originals, but made a suitably weird world for Mia Waskikowska's Alice to inhabit. Some of the characters are like Lewis Carroll's. Steven Fry's Cheshire Cat was just one of many good efforts in small parts.



This  Alice is a modern teenager, who resists the conventional roles that get thrust on her. She's not a Victorian maiden, and she's a pretty reluctant Hollywood action hero. Her dress in the potion clip deserves to make somebody a few millions in sales, and all the costumes and visuals are up to Tim Burton's usual standard. I especially loved the army of playing cards.

In this movie the IMAX 3D was just OK, for me, not a big thrill like the 3D from Avatar.

Helena Bonham-Carter's Red Queen is BIGHEADED, IMPATIENT, AND UNREASONABLE. Apparently she based the performance on her imperious toddler daughter.



Somebody (not Lewis Carroll) made up a lovely bloodhound for Timothy Spall to play.



The bloodhound produces the best line in the movie, spoken by a deeply cynical talking horse. I won't spoil it …

Heard Keith Devlin on NPR ( http://www.npr.org/templates/story/story.php?storyId=12463231 ) saying that the original Alice was partly a sly mathematical satire. Let's just say that Tim Burton's Alice has less math in it than Mean Girls. Actually, Wasikowska's Alice is a bit like Lindsay Lohan's character in that film, and there's one incidental plot touch in Alice that Tina Fey would have been proud of.

Bottom line: fun movie, great visuals, plot maybe a little conventional, enjoyable performances by a swarm of British stars. No math, chess or puzzles, but that's OK. I'd be happy to see it again.

Chart Parser initial release

I just put up on Google Code a version of the chart parser code I use for teaching computational linguistics. Feedback welcome. The slightly cruel example sentences about pigeons in cages being … um … reinforced are a throwback to my grad school days in experimental psychology. No pigeons were harmed in the making of this parser.

http://code.google.com/p/chartparse/

Teaching in Tübingen

I just spent a very happy couple of weeks teaching a compact (=insanely intensive) course from Croft, Metzler, Strohman's Search Engines: Information Retrieval in Practice. Really good textbook. Teaching four hours a day for two weeks is actually feasible! There were some nice projects: two groups added Named Entity tagging and one made a Russian version of Galago.