In how to build on bubble-up folksonomies Tom Coates says:
[…] The concept is really simple – there are concepts in the world that can be loosely described as being made up of aggregations of other smaller component concepts. In such systems, if you encourage the tagging of the smallest component parts, then you can aggregate those tags up through the whole system. You get – essentially – free metadata on a whole range of other concepts […]
and goes on to play with ideas for aggregating tags on radio songs into folksonomic descriptions of aggregates of those songs (radio shows, albums) and aggregations of aggregations (a radio station, an artist’s body of work).
Reading it I was struck by a link to something I wrote about a year ago on semantic aggregation and filtering (I’m using aggregation to refer to a slightly different thing in that post) – so from that I would add to Tom’s idea the possiblity for allowing new tags to be added to describe different entities in the aggregation – e.g. directly tagging the shows as well as using tags derived from the tags applied to the songs.
Tom goes on to suggest that by using the links between these emergent tags you could lead people to new-to-them material that reflected the best example of things they may like – “best” being determined in a Wisdom-of-Crowds-like way by the station’s listeners.
The concept makes immense sense from the perspective of a broadcaster that is seeking to create new metadata about material, and to provide listeners with the most engaging experience.
From the perspective of a listener though, I’d like another layer. Alongside the “transmitter-side” aggregation of metadata from the broadcaster based on the tags submitted by their listeners, I’d like a “receiver-side” metadata aggregator that aggregates my tags across all the media I’ve ever listened to over time – and on top of that a way of comparing “my” folksonomy with “their” folksonomies so that I can find new artists or stations that I am likely to enjoy.
