What Have We Learned from the Omidyar.net Experiment?

Lucy Bernholz, Give and Take, and The Chronicle of Philanthropy speculate on the shut down of Omidyar.net.  If it were my funeral, I would want more grieving and less commentary. The conversation will find new homes, and the friendships will persist, but the end of an experiment that was started with such high hopes leaves sadness and gratitude.  "What have learned?" would be a better question than, "How come they shut it?" What have we learned that will be useful in extending the conversation of giving going forward?

My biggest take-away is that we are tribes and tribes of tribes.  Whatever sites we have going forward to support the conversation of giving need to be built around specific tribal groups and their "owners," or leaders. We need cohesiveness within tribes, some distance among them, and some connections too. The unit of cohesion is larger than one and less than 200. We need ways to include and exclude (or at least socialize)  participants and commenters. We need ways to make things private, tribe-only, or public. On-site email and directories of participants are good. (Ning.com is one site with such functionality.) Celebrity or a draw is also helpful. The Omidyar.net site got going because of the Omidyar name, and their connections, let alone the glamor of their wealth and the possibility of getting some, more so than the the technology. I think it fell apart because the "point system" for posts and participants did not sufficiently filter the flow. They wanted a site that ran itself, one that could be scaled to many sites. They proved in the end that a strong personality, a tribal leader, may well be necessary if you want to invite, please, and keep a cohesive, talented audience.  (Pierre or Pam could have played that role, but their participation waned, leaving paid Omidyar employees who could tinker with the point system but were not given the responsibility to moderate or lead.)  The tribal leader has to have the respect and personal connection to each member. That is why scale matters. A federation or tribe of tribes may be the way to scale to thousands of users. An unpaid tribal leader presupposes passion. And passion is not always corporate in style. Hence of the tacit Omidyar dream of a corporate site that runs itself and scales, while everyone remains upbeat and decorous may be hard to achieve, regardless of the technology. Much praise and gratitude to all who set the site in motion and kept it moving.

UPDATE

At Omidyar.net Peter Rees has started a thread responding responding to the question, "What Have We Learned at O.net?"