I was in the open knowledge conference 2009 last Saturday, giving a quick talk on our time-based video tagging system. The conference was ok, not so exciting as last year I think. Especially the part on Semantic Web and Open Data, it was more like an introduction to Linked Data, somewhere you can see in an undergraduate introduction course.
For this tagging project, I was prepared to take over last year, then followed by an acident and I could only pick it up again last November. We start thinking what is a tag? And it comes out with a phenomenological understanding (which is mainly my part). I think “what is a tag” is not well thought, and I was struck with the complexity of tagging, unfortunately people working too well with web ontology (e.g. Tom Gruber) tend to reduce a tag into part of an ontology.
There are three common understanding of tagging:
1. Folksonomy vs Ontology = Bottom-up vs Top-down (Clay Shirky)
2. Tagging as associationaism vs formal rules (Teil, G and Latour, B., this paper actually brings me back to David Hume, and from there I developed a new understanding on relations)
3. Tagging as social convention, somewhere borrowed from David Lewis
I agree with all, but I also want to add a phenomenological approach (this is explored in detail in my paper on Frege, Husserl and Heidegger:
1. A tag is a complex of noetic acts, a tag denotes, judges, associates, etc
2. A tag materializes the meanings to the user, as well as to the objects
3. A tag is in constant modification of the inter-subjective understanding