Some thoughts on collaboration related the unrevdb stuff, the uviz
stuff, the http://www.bootstrap.org/ conversations and climbing too.

From email with Kathryn.

--------------

I've been reading a lot of rock climbing lit lately. Rock
climbing is an extremely competitive sport in a very weird way.
There are no spectators for big wall stuff (like El Cap) and
teams do not get recognition, individuals do. A result of this is
that "spray" is an important part of the culture:
self-aggrandizing explanations of accomplishments. No one was
there to see it, except your partner (and often they can't see
you and besides their role is very subservient when you have the
sharp end (the lead of the rope)), so the only way to let people
know about it is to work it into the conversation.

When collaboration is required in climbing it is necessary to
build a team but that team is frequently a group formed for the
particular occassion, a cause (let's get to the top of this
thing, I've worked out we need X people with Y skils). The group
usually disperses. Parnerships last, sometimes.

Knowledge collaboration seems much the same: lots of spray, real
collaboration only with clear goals, and then only when there is
clear division of labor. Linux gods have clearly defined domains;
particular pieces of the source that _belong_ to them.

Successful collaboration in work or school has happened from the
enforcement of arbitrary (in the dictatorial sense) goals, but
still involves clear and continually clarified division of
responsibility and expertise.