I have been a small part of some discussion on Sig’s blog (which spread to Hugh’s blog) regarding tagging vs hierarchical structures (the discussion starts there and progresses through a number of subsequent posts). One of the other participants was Jim Wilde of Advancing Insights, so I checked out his site (which deals with Enterprise Blogging). In his post “Social Software“, he in turn links to Clay Shirky’s post about the growth of tag clouds, which details some interesting insight that Shirky has found about the unforeseen variety found in what and how people tag – it’s a good read so check it out.
Anyway, in the context of the ‘tagging v hierarchy’ discussion, Shirky says a couple of interesting things:
Tagging isn’t voting, in other words, with each user committed to one and only one choice, and the views people share widely are less numerous, considered as unique occurrences, than the views they share with only a few other people, or with no one else at all. Tagging isn’t like getting a group to go to a restaurant, in other words, because there’s no requirement for the users to converge on a single opinion.
Taggers DON’T HAVE TO AGREE with each other! This is one of the things Sig and I have been trying to argue – Sig calls it ‘imprecision’, I likened it to people finding their own pathways to things, but Shirky’s post made it very clear to me – one of the major differences between tagging and tree structures is that tree structures don’t allow for disagreement – it’s my way (if I’m the ‘object-creator’ in Sig-speak) or no way.
Shirky finds another difference:
Taggers are good at characterizing material in ways that search engines are incapable of, and tags are thus good for letting you find material whose characterization does not appear in the text itself.
People tagging his post used expressions/categorisations which (although related, obviously) were not mentioned in the article – taggers discovered the IMPLICIT theme, and categorised that as well as the EXPLICIT discussion. Shirky compares this to search engines which index the actual text, but the same applies to tree structures – by definition they have to be explicit, where tags can be both explicit AND imprecise, even implied … I almost understand some of this stuff now.