I know this reads more like stram of consciousness than something well organized. But I had to dump something out here… it’s getting cramped inside my brain.
I hope you’ll forgive the reference to the Bush who’s vision we all respect. I think there’s something big about to happen. Actually, it’s already started. I won’t manage to sketch the whole here, but hopefully enough of it for the flavor to be distinct.
You will recall Adam Bosworth’s amazing keynote (audio). Joseph Hardin recently pointed me to a great commentary on another of Adam’s posts about the platform of this decade.
I’ve been struggling to understand what we do with the next generation of social software. Of course, I’m interested in facilitating formal and informal learning interactions, but I’ve come to believe that no piece of social software targeted primarily at education is ever going to have a significant impact. There are currently some discussions about centralizing where social software (esp. for open educational resources) will go next.
Think about the social software systems people are actually using in numbers these days:
Blogging, like wordpress Social bookmarking, like del.icio.us Photo sharing, like flickr.com Social networking, like friendster Social search tools, like technorati.com Mapping tools, like frappr Comment tracking, like CoComment Question and answer software, like Yahoo Answers
With the number of users of these tools, and the others in each of these categories, why on earth would we create a new tool in any of these spaces? When critical mass is the single most important part of a network, why would we build another ‘walled garden’ collection of applications? Why would we trade tens or hundreds of thousands of users in existing systems in favor of an “integrated” collection of these tools?
Lost Boy’s lists as social content engines (thanks againg to Joseph for pointing this out) comes really close, but falls down on the rebuilding issue. He says:
However for really serious use I believe we need something more: a service like del.icio.us that allows storage of arbitrary properties against web resources. And it seems obvious to me that such a system would be based around an RDF data store. Jo Walsh has already created an RDF version of the del.icio.us API, and this tutorial on creating a social content engine in RDF provides some additional thoughts.
(Now a PhD student and I are working on a system sort of like this, but you’ll have to wait another week or so to start playing with that. But he and I are clear in our belief that this system is only one part of this larger framework I’m describing here.)
The RDF part of Lost Boy’s statement is a key insight that I’ll return to in a moment.
I don’t think we need new systems in spaces where there are already incredibly popular tools. We have lots of existing systems that people already like and use. It’s worth asking - what are people doing when they use these various systems? They’re creating what Mimi Recker and I called “nonauthoritative metadata” (hey - it was 2001 and I wasn’t tenured yet! =) The advent of social software and tagging systems have turned all of us into metadata creation engines. And what did Mimi and I recommend one do with nonauthoritative metadata? Make recommendations.
Many of us already aggregate this data via RSS and ATOM for viewing and reading. Maybe you even use one of those newfangled web 2.0 desktops to read many of your feeds in the same place. Having access to all that data in one places is, granted, more convenient than having to chase it all over the web. However, these systems generally only show you the data from a remote location. Why don’t we store all this data in one spot?
Then why don’t we provide easy access to this data as RDF? There is admittedly some work to do here for each service, but nothing insurmountable. Then you, and me, and everyone else in the world can start writing agents to make (lowercase) semantic web-like uses of this data. Sure, the semantic web supposedly runs on a high octane mix of controlled-vocabulary-ontology-taxonomy mayhem, but who says we can’t run a useful system on folksonomic data? Sure, it will be a much more “interesting” problem. On the other hand, since you’re never going to have a large collection of non-folksonomic metadata to play with, what choice do you have?
A collection of information of this kind will provide amazing data for agents to recommend new content and new people to you. This seems like the future, so we’ve started working on it at COSL. Someone else may do it better, but I think we’ll have a blast along the way, and hopefully make sure that the things we need to support learning stay in there, just far enough beneath the surface that they won’t scare people off.
