Taxonomy
David Weinberger has taken on taxonomy as a project this year. A book will be forthcoming and the truth will be revealed. Meanwhile we can only wait and wonder, and maybe pick up an $80 copy of next month’s Release 1.0. Today David points us to Clay Shirky’s comments on Louis Rosenfeld’s observations regarding "Folksonomies." Some of the discussion here relates to the relative merits of metadata taxonomies (think Dublin Core) versus "Folksonomies" (think FlickR and del.icio.us).
From Carl Linnaeus’ work categorizing life forms according to kingdom (plant or animal), and then hierarchically by phylum, class, order, family, genus and species to modern phylogenetic systematics… from the DBMS data dictionaries of the 80s to the XML Document Type Definitions of the 90s and more recently the XML Schemas of the 21st century, we’ve been categorizing, coding, classifying, and otherwise discriminating one thing from another. Geologists do it with rocks and minerals. Linguists do it with languages and phonetics. Hell, everybody does it.
I wonder if there isn’t a useful distinction between "searching" and "finding" lurking in here somewhere. A large bank might be interested in searching and finding its entire agricultural equipment lending business collateralization. Folksonomies might not be the best approach for making this happen. Different loan officers might call those Massey Ferguson Corn Pickers by different names. Maybe another axial distinction might be between "want" and "need." I want to find information on free lance writing and kill fees. I don’t need all the information. Folksonomies might be good enough for what I want.
There will be problems, chiefly with commonality of tag assignment I think. But many hits on a topic can be sufficient, when all hits aren’t required. Of one thing we can be sure, the term "Folksonomy" is here to stay. We’ve needed a new neologism for a while now. Is "new neologism" redundant?
