0Tim Berners-Lee's TED Talk about Linked Data went up yesterday. Always a delight to hear him. The big message, at least for me, is:

0

0What's RAW DATA NOW? It's the idea that we are sitting on tons of data that can and should be openned up on the web. We don't need no fancy interface for it. We don't need no web site or explanation to force it into a 'proper' context. We don't need no worry about ohmygoodnessmewhatwillhappentoallmydata! We just need a handy open license like CC-"knock yourself out" and let's get the data out there!

0This is exactly the spirit at work when Dave Pattern released data from the University of Huddersfield Library (posts about it here and here, our podcast with Talis here ). He saw that he had a goldmine of data, and put it out there.

0Closer to home, this is a spirit that universities desperately need to start embracing. It's a huge cultural shift -- university faculty and administrators tend to hug their data even more than most people. That's all the sadder, because they're the ones sitting on some of the most useful and interesting data out there. After all, university faculty are the most highly skilled thinkers and innovators out there. That's their JOB, ferheavensakes! It's great to see more and more of them blogging. The next step is to get at their data and connections.

0Following up on the talk, Tom Heath put out an offer to collect compelling stories and reasons for people to get their data out. I'm hoping that will get off the ground -- I'll certainly be working on it. The story of the Huddersfield Library data is one good example. But there are a lot more that I'll be pursuing, particularly in the realm of higher education.

0 This is why I'm working on this funny Giant EduGraph idea (see also here). Take, for example, all the really wonderful associations between ideas that are expressed in just a course syllabus / reading list. The fact that a professor has associated a number of books together for a course tells us that, in the mind of that thinker, they are all related somehow. Even if it is not obvious, the fact that a very smart person has put them into a collection will tell us a lot about the significance of those works. That's data that needs to be exposed, so that others can share in contemplating how they are related. Okay, that might be a bit too abstract to be deeply compelling. But once we get some working examples of that exposure, I think really neat things will pop out.

0 Here's something a little more immediately compelling. Whenever I talk about the Giant EduGraph idea, I start with a sad story of wonderful possibilities for interdisciplinary connections being lost (this is the story I tell in slides 3-14 of the slideshow below, from my talk at ACCS). A while ago, I was working with a chemistry professor who was teaching a course called "Mad, Bad, and Evil Scientists". The course was looking at various representations of science and scientists in literature, film, art, etc. It's a given that one of the first reading was Frankenstein. So here's a chemistry prof teaching this work of literature specifically from the viewpoint of how it represents science and scientists.

0Now, Frankenstein is a pretty standard reading in intro to lit courses. So it's a good bet that over on the other side of campus in the English deparment other teachers and students are reading it. Look! Exactly what we want! Interdisciplinary connections! Sciences and Humanities sharing ideas and perspectives! Dogs and cats living together!

0BUT! How would they find each other? There's no easy way to discover different courses studying the same texts, or same ideas, or same time periods, or same people. The part that makes me really cry is that WE ARE ALREADY RECORDING THAT DATA IN A CENTRALIZED WAY! Orders for the bookstore go out WEEKS before the semester starts. The bookstore has all the data about who's reading what. The data goes in, and overpriced textbooks come out. None of the wonderful connections are realized, even though faculty have done the paperwork (and faculty love paperwork!) that could facilitate it. I want faculty to get an email from a service that tells them anytime someone in the university -- hell, ANY university -- is teaching the same book. Or concept. Or idea. If the data were out there, that'd be possible. The technological hurdles are really low compared with the cultural hurdle of openning up the data. Makes me cry.

0 There's another story that I think is compelling. Really, it's a use-case I imagine for a Giant EduGraph. Prospective students have to make decisions about what school to attend on some really limited info. They get a good feel for the campus -- that's why groundskeepers are so busy in the spring just before prospective students visit -- but how much do they learn about what their intellectual life would really be like? I doubt that they leave with a stack of course syllabi. And if they did, I'd be silly to think that they'd read them to get at the important data.

0Now, let's imagine that the data is out there, and someone has built a nifty interface onto it (NB -- this lets the data-people do the data, and the user experience people do the interface). Alice is looking into universities to attend next fall, and her favorite book is Frankenstein. She checks the Giant EduGraph for classes that read Frankenstein at a school. A chemistry class and an English class come up, and she notices that one of the topics that the chemistry class covers is 'vitalism'. She's never heard of this, but since the Giant EduGraph is part of the Linked Open Data cloud, the abstract of the Wikipedia article on it is right there. (Notice, our prospective student is already learning!). Sounds interesting, so she checks to see what other courses teach about vitalism, and sees a history of science course that also hits that topic.

0 But she really liked the campus she visited two weeks ago, and so she checks the Giant EduGraph to see if there are courses reading Frankenstein or studying vitalism.

0 "Let's see what courses that I might like are there. . . (click, click, click. . . ). Oh. Wait....oh. hmm. It looks like that school hasn't opened up any of their data. Ohwell! Let's see what other topics that history of science professor teaches about. . . "

0Yeah. Let's get RAW DATA NOW!