Leverage points in Digg

I've been working on a post that applies Donella Meadows' 12 leverage points to social software systems. (Meadows was an influential environmental scientist who founded the Sustainability Institute.)

Luckily the recent Digg controversy illustrates several of these leverage points, so I thought I would write up a quick summary. The controversy involves Digg's friends feature (here are mine) and the practice of automatically digging one's friends' submissions (sometimes called "circle digging").

"Circle digging" is controversial because it has given a small user population significant control over which stories make it to the Digg homepage. The Digg algorithm also favours submissions by people who have had popular stories in the past, creating a feedback loop that reinforces the influence of top diggers.

What follows, then, are some obversations about how Digg--or any large-scale social software application--is bound to resemble a complex system like an economy or ecosystem.

Meadows' leverage points are places in complex systems where a "small shift in one thing can produce big changes in everything." The recent changes in Digg--as well as people's suggestions for fixing Digg--line up fairly well with these leverage points.

For example:

• Digg changes the system structure by altering the algorithm to reduce the impact of "circle digging"
• They will also change the rules of the system by recognizing top contributors, and recognizing top Diggers for the diversity of their diggs (these are both changes in incentives).
• Peter Abilla suggests that Digg can be more democratic by controlling who has access to what information, specifically "do not show profile or # of votes for up-and-coming dugg articles."
• He also suggests changing the rules of the system to remove incentives for submitting popular stories ("do not profile Top Diggers or elevate anybody higher than anyone else"). This is a structural change as well, since the Digg algorithm favours submissions from Top Diggers.
• Joshua Porter suggests limiting the gain of positive feedback by making it harder to digg stories.
• In the past I've suggested strengthening negative feedback loops by making it easier to bury stories.
• On Joshua Porter's blog I had an idea about sharing aspects of the Digg algorithm with top contributors to build trust, another version of controlling who has access to what information.

There are no magic bullets here, just a slightly different way of thinking about social software like Digg. The suggestions even reflect different goal states--Joshua's and Peter's lean toward promoting quality submissions and indepedent voting while I'm a little more interested in maintaining community.

For more details see Donella Meadows' paper Leverage Points: Places to Intervene in a System (PDF, 91KB), "an exploration of what kinds of actions really make a difference in changing the behavior of a system."