I don’t know how I missed this. Once I found it, though, I kept checking to see if it was an April Fool’s joke.
Common Tag is a format that embeds defined tags in a document or object. That’s right — tags that are already defined; not free-text tagging. In other words, they’re providing a way to link to a subject heading. And are they using existing thesauri? No. Right now, they are using two: Freebase, a community managed database of concepts; and Wikipedia.
So, here’s the problem as they see it. When people tag, sometimes it isn’t clear which definition of the word they are using, e.g., orange the color, the county, the fruit. In addition, they found that people tag the same thing using variations, e.g., nyc, newyorkcity, newy0rk. Do these problems sound familiar? They are the same issues the library field dealt with years ago.
I don’t have a problem with the project as a whole, but to re-create controlled vocabulary like this when you can at least start with one like LCSH seems . . . silly.