(hoisted from the archives)

Integration is hard, and it’s harder when you didn’t build the pieces yourself. Lately I’ve been trying to push people to think bigger about the sharing of OER, precisely because integration sucks.

What do I mean? Well, your slides would be a lot easier for me to use if I had the question bank you used for tests — I am testing on what we do in class, right? That question bank you use for tests isn’t all that useful if the questions react to a book that I don’t use in class, so I’m going to need the book as well. That analytics work you’ve worked into your course design doesn’t work unless they use the discussion board the same way — so can I use that too? Also, I’ll need the homework problems, the worked problems, the clicker peer-instruction questions. And if you could give me some data on how well students usually do on all this stuff, all the better.

In other words — I want courseware that is pretty much the whole course, but open so I can change it as I start to see where it could be improved. And, if it could be managed, I’d like you to make it really easy for me, drop it into an LMS that I can fire up on day one.

As your average professor, I don’t want to have to integrate 80 pieces from different places before I can even run it once. I want to improve it over time, but that’s different.

This is where textbook companies are going already. Fully integrated packages with the assessment baked in. It’s where education is going.

So if we are going to have an open educational resource movement that matters to current institutions, we’re going to have to start thinking a lot bigger.

I think of this, incidentally, along the lines of the old scripting vs. object-oriented debate. There were a set of people that argued great objects would take over the world, allowing a set of integrators to quickly develop solutions from building blocks. Then there were the WordPresses and Moodles of the world that said — forget the building blocks, here’s a soup to nuts product you can run on day one, but change any way you like.

Soup to nuts won, because most people have pretty similar problems — they just want to change 10% of something. That makes time spent integrating a lousy investment for most people. Integration is in many ways harder than building stuff, which is why people using OER often report no time saved. Sharing bigger things solves that for most practitioners in a way they can understand.