The Provost’s office at UNC issued a call for proposals for MOOCs back in November of last year. I’d been wanting to teach a MOOC for some time at that point, having taken 3 or 4 as a student by then. So I submitted 2 proposals, one for a Metadata course & one for an XML course. Obviously the Metadata course was the one that the selection committee liked, and the rest is history.
But why metadata? I’ve been asked. Because MOOCs are good for courses with broad general interest… they have to be, with thousands of students enrolled. And as I was thinking through what course I’d want to launch as a MOOC — even before the Provost’s proposal — I figured that metadata would be the place to start. As I saw it then, there were 2 important issues: One, what topic in Information Science is most broadly applicable? Answer: Metadata. And this was even before the NSA phone metadata revelations. But metadata is everywhere, and largely invisible, so I thought it would be useful to offer a course to the general public that peered behind that particular curtain. And two, if you were going to MOOC-ify the entire ILS curriculum, where would you want to start? Answer: Metadata. Because everything builds on that.
Now that the MOOC has begun, however, I’m seeing things differently. I guess I was starting to see things differently a while ago, even as far back as when I was planning the flow of the course. But now, if I were going to do it all over again, here’s what I’d do differently.
First of all, Unit 1 of the MOOC is a very brief overview of the concepts around what it means to organize information. My thought was that, before we really delved into the specifics & technical details of metadata, I had to put the function of metadata in context. In other words, Unit 1 is INLS 520, the School’s Organization of Information course, in about 90 minutes. So I realize now that if you were going to MOOC-ify the entire ILS curriculum, where would you want to start? Answer: not Metadata in fact, but Organization. Which is more or less where all ILS curricula do start. So I have to wonder if I’m just brainwashed by the status quo, or if this is in fact a deep truth about how ILS should be taught. I’m going for the latter.
Second, I’m realizing that, prior to a course on metadata, students really need some experience with databases. Ok, sure, you can start to address descriptive metadata without discussing databases. You can start to address XML without databases. But you can’t really get to the roots of either of those without databases. And once you get to Linked Data, it’s all about data structures & how the web is one huge database. So in order for a metadata course address what it needs to address — RDF, the Semantic Web, Linked Data — you need to build on prior knowledge of databases.
So there you have it: My early thoughts on how I would redesign ILS curricula. The first courses a student should take: Organization of Information, and Introduction to Databases. Then metadata.