Thomas “Father of Folksonomy” Vander Wal is giving a talk at the Univ. of North Carolina symposium on social software. (I gave the opening 15 min talk, as per my previous post.) [As always, I’m paraphrasing, typing quickly, missing stuff, getting things wrong…]

He defines tagging and then quotes Rashmi Sinha: “The beauty of tagging is that it taps into an existing cognitive process without adding much cognitive cost.”

Thomas thinks the Wikipedia article on folksonomy is now much better than it was. [I haven’t seen it, but that means I’ll probably disaree with it.] He defines it as the result of personal free tagging of pages and objects for one’s own retrieval, usually done in a social environment.

The value of folksonomy is “derived from people using their own vocabulary that adds some explicit meaning, which may come from inferred understanding of the information or object.” He says people aren’t caregorizing so much as “placing hooks” so they can re-find stuff.

He explains the folksonomy triad: Object, metadata and identity. (Flickr tagging doesn’t allow identity, he says.) Between object and identity and interest. Between identity and metadata is vocabulary. Between metadata and object is definition. From identity can come community. Between community and identity is definition. Between community and metadata is terminology. Between community and culture is object. [Here’s a pdf that has many of Thomas’ slides in it, including this diagram.]

He stresses the importance of folksonomies and taxonomies working together, the folksonomy recognizing gaps in the taxonomy. There are business tensions, though, around who controls the naming, the known value of building taxonomies versus the unknown value of folksonomies, and consistency vs. emergence.

He says research shows 0.5% of Net users tag things. People tag because their own use and value comes first, it adds perspective (missing metadata, emergent vocabulary, personal descriptors), refindability (aggregation of info, task-based aggregation), and because it states interest. “Every tag is sacred,” he says, perhaps purposefully echoing Monty Python’s Every Sperm Is Sacred.” He warns us against the impulse to “clean up” a folksonomy.

Q: Is anyone tagging with images?
Thomas: Platial.

Q: Any research into the efficiacy of short descriptions vs. tags? Tags vs. narratives. And few tags are verbs…
A: Perhaps it’s driven by how people search, and they search mainly for nouns.

A: Right now, we only have text boxes. When we have transcription, people will talk narratives. [Good point. When writing, tagging is cognitively easy. When speaking, talking in tags takes focus. It’s easier to spew.]

Q: How about spam?
Thomas: Spamming is hard when you have identity.

Q: Who controls? At K-fed.com, they’ll probably strip out the “should be working at Wendy’s” tags and Slashdot strips out the “gay” tag because it’s applied to every article.
A: “Every tag is sacred” is an ideal, but it helps if you have identity.

Q: Tagging for retrieval is difficult since you can’t know how you’re going to want to search for something.

[Tags: unc thomas_vander_wal folksonomy taxonomy everything_is_miscellaneous ]

Categories: Uncategorized dw