From: http://tomgruber.org/writing/ontology-of-folksonomy.htm
Apply semantic web technologies to the data of the social web.
Folksonomy is data that is emergent from shared information.
Folksonomy is not a taxonomy or even a collaborative categorization.
The emergent data from the actions of millions of ordinary untrained folks helps to counter spam-induced noise in search engines. It gives text-based search engines a fighting chance.
Folsonomy and Semantic Web are often presented as alternatives, but this is a false dichotomy.
Shirky's critique equates ontology with information organization. That heirarchical centrally controlled taxonomic categorization schemes are limited (such as the Dewy Decimal System).
Taxonomies do indeed limit the dimensions along which one can make distinctions - local choices at the leaves are constrained by global categorizations in branches - therefore, it's difficult to put things in their heirarchical places - and categories are often forced.
However, Shirky's critique relates only to a very narrow Ontologicial form and methodology. He equates ontology with a centrally controlled taxonomic effort.
[i don't get the distinction here - is the author saying that his proposed tagging system is a broader ontological approach and that many of the current "categorize the world's information" ontology efforts are indeed flawed?
]
Ontologies-as-conceptual-specifications enable multiple, independently developed databases of carefully categorized artifcats to interoperate. And for agents to reason about the differences among the vocabulary used in each of those independent databases.
[ok, is this the point? that the approach should be more conceptual than specific? it should describe relationships without describing these relationships in sentences and words? but isn't that the point of an ontology - to describe relationships via sentences?
]
Shirky's critique is really an attack on top down categorization --as a way of finding and organizing information--
[meaning that the author's proposal is not to use ontology for finding and organizing information? but just for representing relationships between disparate data sets?
]
There should be a common conceptualization of what tagging means - and a way for a service to correlate and connect tag data from one application to another. There must be a way of reasoning about the equivalence or relationship among tagging data across applications. This is an ontology of folksonomy.
The system must identify areas where systems will differ. Ontologies are as much about reasoning about incompatibilities as finding commonalities.
TagOntology is the effort to do this. tagcommons.org
Central is the notion that techniqyues of the semantic web, such as formal specification of structured data and reasoning across disparate data sources, can apply to the social web.
Example
Tagging(Object1, tag1, tagger1, source1)
Tagging(Object1, tag2, tagger1, source1)
Tagging(Object1, tag1, tagger2, source1)
This allows us to say something about a collection of tag data, independent of the specific applications they come from.
Holy moly. I've been wanting to do this idea forever! My friend Ted just turned me onto it... you are a little character that walks across the bottom of your screen and whereever you go, you see other people using the same service. I think I blogged about this idea some time ago... could never figure out though how it would be useful until all of your friends got on the service - so that you could actually surf with friends. It's pretty lame with noone else...
http://www.weblin.com/
