I have a piece on Comment is Free about open data and public engagement.
Their title "Citizen science still needs specialism", with the sub-heading "The public can be involved in constructing knowledge. But some data sets are more easily offered for external use than others". Both of which I do kind of say, but my point is more that simply opening data doesn't really work as a public engagement strategy: the people who can use it aren't exactly the public. That doesn't in itself mean open data is a bad thing, but we do need to be aware of the specifics of how we expect such data to be used.
I have a larger p.s. to this piece, with some links to the academic work I drew on, but I won't have time to post that till later this week.