peterme.com

Happy New Year from Adaptive Path

In what is becoming a tradition, our 2005 inaugural essay lists resolutions from each member of Adaptive Path. There's some good stuff in there -- check it out!

Separately and additionally, don't forget Jeff Veen and I are teaching our "Make Your CMS Work For You" content management workshop in our offices in San Francisco on January 25. Use promotional code FOPM and get 15% off the registration fee.

Heading to Ol' Blighty

From 9 Jan to 17 Jan, we will be visiting London. Staying in Barbican with a friend. Wondering what on earth to do. Not for a lack of things to do -- frankly, there's too much!

If you're there (or will be) and want to hang out, email me. Would love to hook up with peterme readers.

Mob indexing? Folk categorization? Social tagging?

For folks interested in free tagging systems such as those seen on del.icio.us and Flickr, Adam Mathes' essay, "Folksonomies - Cooperative Classification and Communication Through Shared Metadata" is required reading. It's probably the single most thoughtful discussion of the issues at hand.

Among other things, Adam takes me to task for using "ethnoclassification," because classification schemes require applying single, or mutually exclusive, classifications to an item. Citing Elin Jacobs' "Classification and categorization: a difference that makes a difference" essay, he argues that what we're seeing is categorization.

However, considering its presence in the title of his essay, Adam seems perfectly happy with the term "folksonomy" which, if anything, is even more wronger than classification. In Jacob's article, taxonomy (from which "folksonomy" is coined) "is carried out within the arbitrary framework established by a set of universal principles," and is listed as a subset of classification.

Yes, it annoys me that "folksonomy" is becoming accepted because, 1) it's inaccurate and 2) it's ugly. I'll cop to "ethnoclassification" being insufficient. But where does that leave us? Other terms I've seen, like "distributed classification" are clunky and inaccurate, too.

Elin Jacobs' essay suggests some possibilities. (I can't recommend you read the essay -- clarity and brevity is not the good professor's strong suit. But if you want a crash course on this stuff, it's a pretty complete work.) Jacobs discusses postcoordinate indexing, defined elsewhere as "A method of indexing materials that creates separate entries for each concept in an item, allowing the item to be retrieved using any combination of those concepts in any order." This is exactly what we're doing with free tagging.

Which lead to the title of this post. I somehow want to combine a synonym for "multiple of people", such as "folk", "mob," or "social" with some accurate description of the activity, "indexing," "categorization", "tagging."

(Though I recognize, sadly, that "folksonomy" will likely stick.)