In science we learn and do research based on the findings of — and conclusions drawn by — others before us. (”Standing on the shoulders of giants” ring a bell?) That is the very essence of cumulative science. To make the obligatory car analogy: What if the car manufacturers had to reinvent the wheel and the engine (steam first) every time they wanted to make a new model? That would most certainly make Thomas Kuhn cry, and we’d be driving ridiculously old-fashioned cars.

Yesterday my professor told me that he couldn’t actually provide us with access to various papers on the subject of the course he’s currently teaching, because certain publishers have hired students to spy on the teachers at the university and report back on any copyright infringements they notice. Usually the teacher would upload a PDF of the relevant article on our intranet, thus allowing students to download (and print), but now we’ll have to obtain these papers by other means, which may be any of the following:

• try to get hold of some obscure copy of some obscure journal containing the article in question
• pray that the publisher has made it available (either for free or relatively cheap) on the web
• get in contact with someone who owns the relevant issue of the relevant journal containing the relevant article and hope they’ll let us xerox it (and still infringe on the copyright, but this time without the professor’s help)

… and that’s only the articles.

A lot of the texts we use in linguistics come from various collections, where a number of people contribute to the work published in book form. Now, we have some good libraries here, but they rarely have more than a single copy of a book (well, maybe some older editions), which means that if every student had to read an article from a particular book, they’d actually have to either borrow the book or read the article at the library. Think about it… 24 students all wanting to read the same article (which they’re not allowed to xerox) in the same book, of which the library only has one copy.

One could, of course, buy all the journals, collections, etc… But that would be so ridiculously expensive that studying something like linguistics would cost a fortune. Handing in a decent paper based on good research would quickly run into the thousands of euros. And we’re expected to write 2 or 3 of those papers each semester…

Now I’m thinking Science Bay (or maybe Science Nova) — a marriage between Pirate Bay and Discogs containing nothing but torrents of scientific articles neatly categorized, with detailed meta data, and searchable in every imaginable way. It would even be possible to provide BibTeX entries for every article, and even provide one huge, downloadable BibTeX database for the entire content of Science Bay. When uploading a torrent, one would be able to either fill out all the relevant meta data form fields for the article in question, or simply paste a BibTeX entry containing all the relevant/required meta data. It wouldn’t just be a solution to a problem, it’d be awesome! Searching for and actually finding relevant articles couldn’t be much easier! I’m not talking about sharing entire books or even just chapters of books. Just scientific articles from journals, collections, and what have you.

Only problem is, where would one get such a project hosted? We’ve all heard about the legal trouble Pirate Bay has faced through the times and I certainly can’t afford to hire a lawyer to fend off pissed off pubilshers. While I could probably easily do the code for the project, I don’t have access to decent servers, bandwidth, Hungary, legal advise, etc…

But isn’t this how science is really supposed to work? Everyone sharing what they find and contributing to a greater cause. I know some scientists like to live like rock stars (I’m looking at you, Hawking! :-p), but that’s no excuse for their publishers behaving like retarded record labels.

Indeed it has. I’ve been (and still am) busy studying, working and doing stuff, like social interaction with other members of the species homo sapiens - and studying for assignments. In the beginning of January I’ll have to hand in 3 assignments, Language and Culture (actually just anthropological linguistics), Methods of Linguistic Description and Theory of Science.

Language and Culture

I’ll try to reject the strongest version of Whorf’s theory of linguistic determinism based on studies of color terms in various languages - studies done by e.g. Berlin, Kay, Heider and of course Whorf himself. Going to be pretty interesting - and it’s always fun to criticise Whorf.

Methods of Linguistic Description

Not entirely sure what I’ll be doing here, but probably something computer-related. We’re allowed to do some sort of alternative project, so of course I’ll try to find something that involves some programming, as that is what I’m hoping to combine my linguistics with at some point.

Theory of Science

Based on Morton Winston’s Did a (Kuhnian) Scientific Revolution Occur to Linguistics? and Kuhn’s own works, I’m gonna write a discussion of the first-mentioned paper and my own view on the subject. Also something I’m actually looking forward to, as I really like Kuhn and (a lot of) his ideas.

So as you can understand, I’m busy, but in the good way - the really enjoyable way, actually. My blogging/coding/internet stand-still will probably last a while yet, but who cares, it’s not like anyone actually reads this crap anyway. :-p

Oh, I nearly forgot - a top [something]!
Top 3: Great Stuff

1. Cowboy Bebop
2. DJ Format (especially the 3 Feet Deep video)
3. Last Life in the Universe

I forgot this point in my last entry…

As some of you may remember, I was running this word recognition experiment for a linguistics assignment I was supposed to hand in in January, but, as you may or may not know, I never wrote the paper, which I still have to do. So I was thinking that I could use some more data (a lot more) and decided to let the experiment run for a while longer.

So, I hereby urge everyone to head to http://lillesvin.net/ling/wr/ and do the experiment. It’s really easy and it only takes 3 minutes, so there’s absolutely no excuse for not doing it. I really need as much data as possible, so apart from twisting only your arm I’ll now ask you to tell all your friends (and enemies) to go do the experiment, link to it in your blog/on your website, print flyers and throw them out of a plane over Tokyo, hack slashdot.org and replace it with a link to the experiment, spam everyone in your address book… I don’t care as long as you’re just doing something!!!

The experiment deadline is April 15 2005, so there’s not a whole lot of time, I know, but I should still be able to get a usable amount of results. I hope you’re willing to help me out - if not, remember what Woody Allen said about egoism…

In advance, thank you sooooo much!

Apparently someone has been reading/watching Ghost in the Shell (Stand Alone Complex)…

Make sure to check out the videos, they’re really cool. Especially these two.

Ok, now it’s time for everybody who gets by here to go to: http://lillesvin.net/ling/wr/ and help me out on my linguistics assignment by doing this little experiment.

The experiment is about word recognition and will probably take about 5 minutes to complete, so please spare me some of your time and give me some data to work with, ’cause I really need it… And fast… ;-) The deadline for the experiment is December 1st.

If you want to, you can help me by telling all of your friends to go here and do the experiment and/or link to it from you homepage/blog/whatever.