Projects: language

This afternoon I went to UB:2 and pulled out a stack of books on language, linguistics, and computational linguistics. After reading through them for an hour or two I left with three (An Introduction to Language, Svensk Grammatik, and A Prolog Primer). I want to teach the computer Swedish and play with machine translation. I think that being a bit of a novice in the language (Swedish) is actually a benefit in such endeavors, as I actually think about the syntactic constructions rather than simply having the language come completely `naturally.' This afternoon I wrote a simple Lisp program that conjugates verbs (trivial, really, but it's a start). I think it would be really neat to create a parser for the language. Of course, that's a huge and non-trivial task, and I'd be completely happy with a limited subset of the language. The parser would start with a basic definition of the grammar (ie some kind of BNF) from which it would start parsing a sentence. The program should be able to try to figure out unknown words by looking at where they fall in the sentence structure and by looking at the word ending, etc. A more immediate goal is to write a program that, given subject-verb-object and a tense, will make the adjectives and verbs agree with one another in the appropriate way. (Of course, I'll probably never actually get around to doing it.)

projects

I feel like my head is going to explode, I have so many projects in mind right now that I'm dying to work on, or, in some cases, have someone else work on. I just started hacking out some thoughts on some of those projects but ended up writing a veritable disseration instead. Ugh! On the other hand, I now have the kernel of a possibly publishable article. Okay, I will summarize the rant: Linux needs extended attributes, access control lists, and filesystem-tracked mime types! We need to recognize the importance of metadata! We need the HURD's pluggable filesystem stuff. LiveJournal needs to export its content in XML format, and in RSS. LiveJournal needs subjournals, so that I can have a "technical" journal, a "travel" journal, etc. They're offering a wonderful service, but I just realised that it's one we don't need, because it can be done better in a distributed fashion! Oh yes, and I hope someone will reply to the previous entry about the electroencephaloscope project. And, I need more time!