OER Landmines


Some great discussions this past week with some new friends in the Open Educational Resources movement. A few things that came out of it:

  1. Nobody should ever build another lesson repository ever again, because they don’t work. Instead, people should focus on making lessons findable (proper tagging for search engines).

  2. Anyone starting an OER project of any kind needs to plan right from day one for handover from early adopters (who are keen on the idea of OER) to early majority (who just want to teach).

  3. “GitHub for teachers” is a dead end because version control is too high a cover charge for most teachers and learners. (See the previous point: early adopters will use it because it’s cool, everyone else will vote “no” with their feet.) “Wikipedia for teachers” is a dead end because its model of one definitive narrative on any subject isn’t the right one.

  4. The Reusability Paradox states that, “The pedagogical effectiveness of a learning object and its potential for reuse are completely at odds with one another”. Options for dealing with it are:
    1. Fine-grained decontextualized resources that are highly reusable but teach very little and require a lot of work to assemble.
    2. Highly contextualized resources that are pedagogically effective but hard to re-use.
    3. The mediocre middle.
    4. Reducing barriers such as licensing that inhibit remixing (which is not the same as unmodified re-use).
  5. The best way to help at-risk students and create more diversity in professions that lack it is not to create more self-study materials, because that’s subject to the Matthew Effect (the rich get richer and the poor get poorer). Instead, the best way to help people is to support teachers who are already helping those groups. (Quoting Mike Caulfield, it’s like the advice they give you on planes: in event of sudden decompression, put on your own mask first, then help the person beside you.)

  6. Don’t measure success solely based on the performace of auto-didacts: your module might save them some time, but they’ll learn with or without you. Instead, measure performance by looking at general classroom adoption.

  7. Don’t rely solely (or even primarily) on analytics to figure out what’s going on or how well you’re doing, because one of the biggest causes of lack of success is misunderstanding how teachers use material (or don’t). Unfortunately, “open” often means “no support”, which in turn means that the only insight creators get is from a distance. In practice, the best information comes from talking with teachers and listening carefully to what’s going on. Often they won’t have correctly diagnosed the problem, but they know there is a problem, or they are confused about something, or missing something to make the thing work.

I’ve got a lot to think about. In particular, “Stack Overflow for teachers” doesn’t seem to have been tried yet at scale, and once again I’m back to choral explanations.