06.02.05

Peer Production and Structure

Posted in Uncategorized at 1:11 pm by Todd

Spurred by Chao’s recent post, I rolled up my sleeves and tucked into Coase’s Penguin, a discourse on the emergence of Commons-Based Peer Production. CP was authored several years ago, but I’d never seen it before, and I have to confess that it was really rather mind-blowing. I had a dozen “aha!” moments, as CP really captures some core shifts in paradigm, changes that I’d seen and, perhaps, understood only intuitively. CP really helped to coalesce some my thinking.

Now the problem is that I’ve got Peer Production on the brain: everywhere I look, I see problems that might be amenable to a Peer Production type of solution. I’m sure I’m going overboard, and in time Peer Production will seem like just another arrow in the quiver. But for now, it feels like the arrow in the quiver.

At any rate, I thought it might be interesting to apply some of the ideas outlined in CP to some of the areas I’ve been exploring. Let’s look at the public event space, for example, as one that is near and dear to so many of us.

Events call for structure. They are entities that have fairly well-formed metadata associated with them. They are interesting to wide swaths of people (essentially everyone). They are also ephemeral. There are an infinity of authors (potentially everyone) and a number of fairly new aggregators. Let’s look at Upcoming.org as an example.

If you accept (and by now I hope you do) that the ability to do interesting things with events is substantially hampered until you can get some kind of explicitly structured online representation going, then you can view Upcoming as a kind Peer Production model for accomplishing that task. What do I mean by that? Loads of event hosts (see Bottom of the Hill as a favorite example) provide online listings of their events, but in general not in a usefully structured form. But if you go to the appropriate page on Upcoming, you’ll see they’ve managed to create a nicely structured page of events at Bottom of the Hill — this is in RSS. And if you do a view source of the page for a single event, you’ll see that the event is marked up as an hCalendar event. (Aside: They seem to have gotten the “dtstart” title wrong; it’s empty on this page.) How’d this happen? How did we go from a completely unstructured page to a highly structured one?

That, dear reader, is the power of Peer Production. The answer is that the structure was created by the devoted users of Upcoming. By filling in a simple form for each event, Upcoming’s users have created structure for those events, given each one a URL, and in doing so enabled further processing down the line.

While this does demonstrate something of the power of Peer Production, I can’t help wonder whether it’s really got legs. Part of the problem is that this endeavor that Upcoming is implicitly demanding of its users never ends. There are always more events, and until someone keys them in to Upcoming’s interface, they don’t exist in the structured world. So what’s the incentive? What does Upcoming offer me in exchange for my manually keying in event data? That part of the value proposition isn’t really clear to me. In contrast to, say, NASA’s Clickworkers project, or Linux development, I don’t think Upcoming offers its users much in the way of hedonic or social/psychological benefit. And there’s clearly no money associated with it. So can you expect Upcoming’s users to continue indefinitely to key in events?

A second problem I see is demonstrated clearly by an event that had been scheduled at Bottom of the Hill, but then, apparently, was canceled. Yet the event continues on in Upcoming. Do all of the people (4) who had expressed an interest in this event know it’s been canceled? Hard to say.

There are actually two issues conflated here. The first may be a simple glitch in Upcoming’s implementation: is it possible to cancel an event that you’ve entered? The answer appears to be no. This needs to be fixed. The second (and more troubling problem) is that my retyping information that already exists online creates an information gap: if my source (which is probably more authoritative than I am) makes any changes, I have no way of knowing about those changes. Thus I’ve accomplished the questionable goal of making less reliable information more widely consumable. Hmmm.

Here’s another example of potential weirdness: ArborBlogs blogs about an event (presumably one that ArborBlogs itself — whatever it is — is organizing). Included in the event description in the blogged announcement is a link to an Upcoming event. Which one is authoritative? If the details of the event change, will the change be reflected in both places? Since the answer is “if ArborBlog remembers to make the change in both places” then I take that as a no.

Enough complaining. What’s a better solution? Would it be possible, instead of motivating a Peer Production regime to re-create already-existing data, to motivate that same regime to accomplish something more sticky and more reliable? That is the million dollar question here.

More later.

Permalink