There’s been a bit of ennui about institutional approaches to open content in the blogosphere lately, so it was interesting to see this preliminary data regarding OCW production come in last Thursday. The data set generating this graph is not completely clean yet, but it’s pretty close. The graph represents the number of courses published by OCWC members, as a cumulative total, month by month. Some of the smoothing over the last twelve months is the result of not being able to place all releases of the individual courses in a particular month, in which case they were dispersed over the known release time frame — but, even given that, I think the trends are unmistakable:
The most striking thing to me about this is how quickly other members’ contributions are dwarfing MIT’s piece of this. People sometimes still refer to the OCWC mistakenly as an “extension of MIT’s OpenCourseWare project”. Organizationally, that became history when the Consortium was founded as a separate organization, operating under its own governance, in July of this year. But the data above shows that even before the organizational change was formalized our members’ efforts were the driving force behind the OCWC.
The second thing of interest is the larger significance. There’s a way in which this is graphing the height of an iceberg from sea level. That peak there represents nearly 8,000 courses — but associated with those courses are thousands of professors who have now participated in an open education project, ten of thousands of professors that have been exposed to the concept of sharing these materials, and hundreds of thousands of students in these classes who have learned through the example of their institution that knowledge is a thing to be shared, not hoarded. Add to that the millions of people outside these institutions who have hit these courses from all over the world, and who have even come to expect such materials will be available, and we are talking a massive tectonic-scale shift in the expectations we are creating about knowledge.
Mike, these are very impressive data. Do you have a better breakdown? It would be interesting to see by country, or even by language. I would also love to see statistics on content - how many have videos? Just curricula? Etc.
I think what you wrote here and on your personal blog about the rift between edubloggers and the OCW movement is interesting. Perhaps I have thought some of the same things myself, although not said much publicly. Certainly, I think that the more content is made available, the better - as long as it can be ripped and remixed. My concerns are:
The MIT model seems extremely expensive (luckily, many other schools have come up with much more efficient models, like the dScribes model or including OCW production directly in the normal course development workflow etc),
many courses (and I focus mainly on humanities and social sciences, since those are my areas), are extremely skeletal, often simply a curriculum (ie for most people - a list of articles and books that they don’t have access to), maybe a scanned handwritten page of notes, etc,
there has been too little emphasis on both facilitating and researching (I am trying to do my part on the second one) mixup and reuse of OCWs… Instead of spending more money on adding even more courses, perhaps we should try to build a platform like ccMixter, where one can seamlessly drag in elements from many different OCWs (with permissions preserved, metadata, etc). At minimum, what I would love to see, is trackbacks. If I take material from an MIT course, and post it on my own site, add to it and modify it, it should be possible for a link to pop up under the course - clearly noted that this has no affiliation with MIT etc of course - but still enabling people to see derivatives. Because otherwise it’s impossible for people who build on courses to benefit from the community of people who visit the MIT site (and vice versa).
But I think you make some really good points in your blog post, about how many people this has touched, and the potential for transforming their understanding of open, which is certainly very powerful. It’s clearly a baby step on the road, but a very valuable step nonetheless.
Thanks for the post, Mike. And, great points, Stian.
Stian –
Sorry, I’ve been swamped with the day to day — haven’t had a chance to respond.
The data we have is unfortunately very limited. We could probably break courses down by language, at least in a rough sense, but there are some difficulties even there. We are trying to get better at collecting data, but it’s just been one of those things — for the data to be worthwhile it has to be as comprehensive as possible, and to get that we have to spend a lot of time pushing our members to compile and return it to us. Course counts are a little easier, because they come off the feeds that feed into the search mechanism, and as long as published courses are publishing through RSS they are easy to enumerate.
But yes, we’re trying to get more data — just a little short staffed to get really comprehensive stuff together.
On the remixing — I wonder if we are being a bit constrained by our concept of remixing. Certainly we look forward to a future where this reduces the workload of course creation and where learning objects simplify course creation — but I’m not sure that’s the first step. The most likely first step into reuse is course instructors checking other courses online to make sure their course readings or approach or whatever is up to date. In a way that’s remix, and it’s certainly reuse — but it does not revolve around the digital objects themselves. I know that one of the uses we’ve seen for the MIT materials is this sort of cross-checking one’s own course, I think as a wide library of say “intro to energy economics” becomes available that it will be a common thing for an instructor to browse others courses before the semester and see if there’s anything they might add or tweak — a reading or a course section, for example.
Mike,
good point re the reuse. This is something I’d like to keep us thinking about though
But you are right in that it can happen in many different ways.
I am wondering what your minimum criteria for listing something as an OCW is? Does it for example have to be under an open license? Do you allow non-derivative (which I saw on some sites I visited). As we know, the Chinese OCW sites do not use an open license - are they counted?
And if they are, Fun Den Wang, at a conference in Tianjin October ‘08 stated that there are more than 17,000 OCW courses in China - with 1799 courses of “national status”, 5600 of “provincial status” and over 10,000 of “institutional status”. As far as I understand though, the difference is just in the funding model. How are these courses counted? (If they are?)
Stian — I believe in China we are currently relying on data from CORE about stuff CORE is doing. But I’m very interested in the answer to this question, and will try to get more information on it.
@Stian: I asked around here, and it looks like the count includes appx. 1600 Chinese courses of “national” status. If Fun Den told you that there were 1799 of such courses, it’s likely our figure is slightly out of date here, and we’ll work to get that clarified.
As for the other types of courses, it’s quite possible many of these will end up in future counts, but currently they are not counted as consortium courses, even though they may be legitimate OCW.
You probably know what our definition of OCW is from our site and membership appplication. However, Steve gives, I think, a helpful response on on why there are occasional outliers associated with the consortium:
“The community norm is:
- Materials from courses offered for credit at an accredited school
- All materials published are IP-cleared
- Published free of charge or registration under an open license that permits derivatives
- Published under the university’s name
For reasons historical, tactical and political, not all sites adhere to all norms. In some cases the sites started before the norms were firmly established, in other cases the preference is to include a site that doesn’t quite fit rather than make a group feel excluded. So far as counting CORE courses, we’ve only used the national status ones in the past (which I included at 1600 in the last count).”
Permission to use chart?
I’d like to use the chart in a paper that I’m writing that presents OCW as a strategy that aligns with both short-term institutional goals (e.g., student recruitment), and more general, longer-term humanitarian ideals (e.g., free access to educational opportunity).
Can’t find your email anywhere, so am asking here.
-Norm
This post was interesting. I am looking forward for your next post. Thanks again, Diego Charles @ tinypocketpeople