The first part of the paper explains the inherent conflict of interest involved in legislators having responsibility for publishing legislative data. He examines the example of roll call votes, looking at cost and implementation across both Congress and state level state-funded databases. The second part of the paper examines citizen assemblies as a corrective mechanism against situations where legislators have an inherent conflict of interest.
I've been excited about the contents of Jim's research for several weeks, especially the first paper with such direct bearing on Open House Project priorities, so I'm happy to share it here now that the papers are public.
(and if I've portrayed anything inaccurately, I'm sure Jim will correct me.)