Open Curation
I attended a Betaworks brown bag talk yesterday, which featured Megan McCarthy of Techmeme. She was basically explaining her workflow as the editor of Mediagazer and Techmeme, and how she ‘augments’ the algorithm that does most of the heavy lifting there.
A lot of questions ensued: people wanted to know how it worked, what she saw on her editorial dashboard, when would she step in (to pull in new stories that are important but too young to have gained much attention, picking a better story as the top of a pile-up, and so on), how many times a day did she intercede and so on.
I asked her if they had considered making the curatorial gestures publicly visible, so we could see their activities. She wondered ‘Why would anyone want to know that?’ To which I answered, ‘I want to know everything’, semi-facetiously.
But I do think it should be visible, and not just for the edification of those viewing the resulting page at Techmeme or Mediagazer, although they would get a direct benefit perhaps. I was really thinking about meta analysis of curatorial activities by other curators — human or algorithmic — where the sort of curation or the source of curation is extremely relevant.
Imagine a curatorial tool, called Cyur (pronounced ‘cure’), one that is examining stories in my upstream: looking into those that I follow on Twitter and the sources in my RSS feeds, for example. Some of those sources are agents like Techmeme, a curatorial system, itself. But unless Techmeme shares the curatorial actions, Cyur would not be able to know that stories pulled from obscurity by Megan tend to pass my interesting filter more frequently than other editors there.
So, on both points, I think acts of curation should form another part of the stream, as metadata, or editorial gestures, just like retweets and reposts, which is what they are like. And I think its odd that she doesn’t see why we would want to see it.
