Monday, July 5, 2010

Institutions vs. collaboration

Clay Shirky has an interesting talk on TED about institutions vs. collaboration, which, more or less, is the subject of his follow-on 2008 book: "Here comes everybody: the power of organizing with organizations".  


This talk is a variation on the general theme: 'organizing without organizations'.  His proposition is that there is value going untapped because, as he calls it, the "economic framework" of collaboration--that is, the cost of coordination and contribution--exceeds the value of coordinating the contribution of the Nth contributor.

He makes a distinction between "institutions" and ad hoc groups. Institutions are any type of formal organization  with economic, legal, and process structure that has the ability and imperative to shape the contribution of its members; groups are unstructured associations and efforts of people with some common affinity but are otherwise uncoordinated

His idea is that since the cost of coordination has fallen through the floor, the talent and contribution of groups can be tapped economically. He posits that the tools of coordination--he mentions specifically meta data tools like tagging--can be provided easily to participants as a by-product of infrastructure. In doing so,  the work--and the cost--of coordination is borne by the participants themselves.  Think: linux developers and wikipedia.  Thus, Shirky says, we now mine the value of the Nth contributor that heretofore was uneconomic.

Perhaps Shirky has a point, but he steps around a lot of issues in his talk.  Shirky doesn't address assembling or satisfying requirements, assuring the quality of the Nth contributor, nor the security, validation, and integration costs of the Nth contribution.  In fact, wikipedia is not free: there is a large formal institution that governs content, and as the number of contributors has increased, so has the cost of quality.

Shirky does make one interesting point to ponder: institutions operate on the most economic portion of the 80-20 rule.  If 20% of people make 80% of the valuable contributions, then hire those 20% and forego the 20% of contributions by the other 80% of potential contributors. Thus, the 80% constitute the group for which the cost of coordination is institutionally uneconomical. But not no so for coordination afforded by coordination infrastructure and volunteer participants.

This idea, of course, is the so-called "power rule": the Nth contributor makes 1/Nth the contribution of the most prolific contributor.  This means as the number of contributors increases, the peak of the curve gets sharper and the tail gets longer, as we see in the figure below.


 However, that necessitates setting up exclusionary boundaries--now pretty much required because of security--and a "professional class" that is somewhat less agile (there's that word!) than their non-institutionalized compatriots.

Delicious
 Bookmark this on Delicious



Are you on LinkedIn? Share this article
with your network by clicking on the link.