Monday, September 14, 2009

TVA: Step 3 -- PROFIT!

Tom Vander Ark:

Here’s what should happen next:

· Set standards: a consortium (like School of One, Achieve, iNACOL, COSN, and CCSSO) should set standards for digital content in chunks ranging from short objects and lessons to longer units and courses.

Is TVA aware of the decade or more of work that has already gone into this problem? More importantly, is he aware that despite numerous well-funded global initiatives, the net results are as inviting as a cup of lukewarm spit (or is that why he doesn't mention them)? Has Achieve ever shown the slightest interest in taking on this kind of technical work? If they have, there's nothing on their website to show for it. Does iNACOL, COSN or CCSSO have any capacity whatsoever to do this kind of work? No.

Does TVA have any idea how hard this step is? You're not only talking about the kind of complex technical specifications which, quite frankly, almost never work, but what you're trying to specify is the process of educating a human child from early childhood to adulthood. To have a meaningful standard, you have to have a shared understanding of that process. Good luck with that.

Beyond that, you need to find a group of people with a very advanced understanding of information architecture, educational statistics, pedagogy and politics to do difficult, thankless, incredibly boring work for little reward.

Nonetheless, I'd say, oh, $50 million dollars and five years would get you a good start on the foundations of a solution, if you at least bound the problem to US K-12 education as it usually implemented and build on existing solutions. That's assuming that 2/3rds of the work will be completely useless, which, based on history, is probably optimistic. You definitely can't move faster than that, no matter how much you spend.

The big problem is that for your $50 million all you get is a stack of arcane technical specs and, if you're doing it right, which in the US you almost certainly are not, a set of open source reference implementations for various platforms.

OTOH, the gold star in this work in US K-12 goes to Jes & Co. They've stuck with this stuff through thick and thin. Someone should give them $50,000,000. Since they don't go to the same dinner parties as TVA, apparently, this is probably unlikely.

· Tagging scheme: Then they should develop a tagging scheme by theme, modality, and use license.

Not to seem picky, but if you're going to call this "tagging," (" a tag is a non-hierarchical keyword or term assigned to a piece of information") you're going to reveal how little you understand about what's really involved here. "Tagging" is by definition flat, simple, ad-hoc. It is an insufficient approach to the problem, which requires formal, hierarchical and/or networked specifications. It is hard, and that's why it hasn't been successfully done on a large scale, not because it hasn't been tried. We need protégé for education, not tags.

Let me give you an example: you need to look at an online assessment given to a child six years ago. Since then the standards used by the school have changed twice (new state standards two years later, new federal standards last year), and the technical standards for specifying both the assessment itself and describing the academic standards have also both been revised. Your "tags" have to be able to translate the old data into a form meaningful in the current context. I can almost tell you what you'd have to do to make this work, but it would make you want to stick pins in your eyes. Also, it probably wouldn't actually work.

· Keep moving: in the mean time, School of One should keep chugging along and expanding their pilot with their own framework—we need more next gen examples fast!

· Issue RFP: Groups of districts, states, and networks should issue RFPs for content and producers (or intermediaries) would tag and submit their content. Some RFPs could be entirely open, others a mix of open and proprietary. Some RFPs would be at the learning object level, others may only ask for units and courses.

· Develop pay mechanism: It will eventually be possible to create a series of micro-transactions to pay producers based on use and learning gain.

Micro-transactions? 1998 is on the phone, they want their failed ideas back. In particular, how would this work in a School of One like system? Is a school going to install an black box AI to decide if a student happens to need paid resources from vendors and automatically charge the district?

· Fund service economy: A service economy of taggers, repositories, recommendation engines, assessment providers, next gen LMSs, social learning networks, PD and SES providers, and school improvement services would spring to life with venture. Philanthropic, and i3 support.

It is going to take a LOT of up front investment to get to this. And time, and I just don't see who has the stomach for it. Certainly TVA didn't when he was sitting on the money pot.

This is a lot more difficult than iTunes, but it’s time for an ascendant platform like that and it will only happen with a handful of influential folks set standards and folks that control access to markets to issue forward leaning requests for content. The process outlined above would break out of the flat boring choices we have today and usher in the next generation of engaging personalized content. Alex Kozak’s “open trajectory of learning” would be accelerated by a set of standards and an online economy where open and proprietary content could flourish together. Then everyone would have access to a real School of One.

It is amazing to me that someone would write something like this on a blog promoting his own consulting services, although I guess to be a successful consultant you only need to know a tiny bit more than your clients. Kind of like a teacher staying one chapter ahead of his literature class.

This seems like the kind of problem we will eventually have to solve. I'm beginning to doubt that.

Later... Alternately, IMS and particularly Common Cartridge might be way further along and generally awesome for US K-12 than I think. I doubt it though.

No comments: