Collaborative library (ColLib, via S. Downes) sounded very promising to me, since I am hoping for the peaceful coexistence of amateur tagging and professional classification that C. Shirky denies (Aug. 27th). His most plausible argument ist cost: Tagging is cheap. But for whom is tagging cheap? For the providers of the tags – okay. But for the consumers?
Consider the often-quoted example where a topic is difficult to class with a single category, e. g., a newly emerging interdisciplinary field at the border of two subjects. Tagging fans are enthusiastic about the possibility to tag it in both categories. So, for every item there are two tag relationships created that must be processed at browsing time. Classification, in contrast, would perhaps designate one category as the primary “parent”, adding a single “see-also” link to the secondary parent category. Thus the numerous duplicated relationships to the second category are bundled and inherited by a single link. (Additionally, the professional decision for the slightly dominant affiliation to the primary parent, adds value and reduces entropy.) Which version is more economical?
As S. Powers said:
“At some point a human has to intervene with the technology to refine and validate the result. With ontologies, the intervention occurs before the data is captured; with folksonomies, the intervention occurs with each search.”
(Here, “ontologies” can be replaced by “classificaton”, and “folksonomies” until recently seemed to be the hyperonym of “tagging”. i. e., without flavour of organizing the whole world but with emphasis on “folks”.)
Since information has been becoming cheap, we are suffering from information overload and we try to fight it using filters, tags, and categories to organize the abundance. But strangely, we are now, in turn, hoarding these cheap tags as if we were hunters/gatherers who are anxious to miss something:
“Whenever anyone wants to find something and can’t, Amazon loses a potential sale.”
(C. Shirky). Is gathering tags still so important that we must strive for tags overload ?
Shirky himself explains that we are in the post-search world where no stand-alone tool is needed:
“But tags aren’t a stand-alone tool. Tagging is the first post-search tool for information organization; tagging only makes sense in a world where Google has already become normal.”
So finding information (like known-item search or looking up titles) is no longer the main goal of cataloging. But organizing it is what really reduces cost.
Therefore I do think there is a chance for peaceful coexistence and for a fruitful division of labor between tagging and classification. IMO, it could best take place dynamically and vertically:
- Especially for newly emerging fields at a detailed, bottom level, amateur taggers are perhaps closer to the matter than their professional counterparts from LIS staff.
- As time goes by, there will be new or reorganized upper levels in the classification hierarchy derived and destilled from the pioneers’ tags.
(The top-most levels, however, need not be debated or agreed-upon (as failing attempts to “organize the whole web into a hierarchical taxonomy” were trying). Views can easily map one hierarchy to another.)
As for the above-mentioned ColLib, I don’t yet understand how it should work and if the collaboration is meant to be bottom-up and constructing sophisticated see-also’s and mappings. Another possible collaboration of taggers and LIS pros is what I thought of in my earlier (#52) post: create tags like del.icio.us/tag/ddc.xxy/ to integrate the folks’ bottom tags into, say, Dewey DDC upper level classification.