Interesting question, but I don't think it does! I don't know why ross calls it a power law, because a power law is usually used in ref to the long tail, and is a) differently shaped and b) used to map a distribution. Perhaps he could call it a commitment curve, or engagement curve....

Either way I think it conflates some important distinctions. First, it puts these very different user modalities on a curve -- some don't belong, some are inseparable (reading/writing?!), and the assumption that they are independent of user personality is a bad idea. There are lots of different kinds of users, and some of them together are required for healthy and active community participation. Pundits need their fans. Experts need their critics. Connectors need their status seekers. And so on. So to map engagement on the basis of the "engagement depth" of a user action doesnt seem right to me.

Besides the assumption that these are comparable, is the assumption that they correspond to an outcome: collaborative intelligence. If collaborative intelligence differs from collective intelligence simply in its engagement depth then the whole curve is tautological. In other words collaborative intelligence is then just a label -- not an outcome.

There's the matter of relationships (moderate, share), which are fundamental to social media and which are grouped in here with user actions (read, write). Doesnt make sense. Write or blog, comment, message, tweet? Writing is only how they're captured, but not what they *are.* A confusion of the message with the medium here, perhaps. Relationships required for moderation are totally different than relationships involved in tweet followers -- wouldn't a curve want to distinguish between these types of relationships? Better then just to say "collaboration," which is a commonsense term for interactions -- and knowledge, if that is the outcome of topical social interactions.

I'm just getting started, so I'll stop now before I get ridiculous. ;-)

I think the best way to approach this is through user types, motives and incentives, and social bias introduced into content and conversation by the use of questions/answers to get visiblity, presence, status, etc.

The answers space is really interesting -- the ux for the asker has a relevance it doesnt have for the answerer -- the asker wants a question answered (or wants to ask a question!); but the answer hasnt had that question in mind, so what's her incentive to answer (cred, rep, status, visibility, flirts, etc)... Could one design a question space -- a social hunch.com -- for social affinities? Do our questions give us away?
Permalink | Report