Folksonomy Folktales 2010

Yes, this compromise will not accurately reflect how every single user would have done the job and so every user will have some aspect of their categorization overridden by the Yahoo (and other) categorization schema.  And this is also true for every folksonomy as well insofar as it incorporates any kind of relationship between tags.  Folksonomies only avoid this problem by not offering any sort of categorical relationship whatsoever. 

A third factor in the “revolutionizing” of folksonomies might have arisen from a combination of one aspect of our IT/technical culture which is the mechanism of “releases” turned into a metaphor – 2.0 in this case, and the standard marketing hype of everything is revolutionary (from a new soap to a new more efficient way to publish content).  To counteract this particular (bad) marriage of concepts, I recommend reading two books, Metaphors We Live By – George Lakoff and Mark Johnson,  and Death Sentences:  How Cliches, Weasel Words, and Management-Speak are Strangling Public Language by Don Watson.

Conclusion

First and foremost is that Folksonomy should never have been compared with taxonomies, thesauri, or ontologies.  The comparison is essentially bogus in that folksonomies are not a classification system at all.  Popularity is not a semantic structure – to repeat, there is no onomy in Folksonomy.  I give the creator of the term, Thomas Vander Wal, full marks for creativity and marketing acumen, but failing marks for accuracy.

Folksonomies do not organize information, they are aggregates of 1,000’s or million’s of individual acts of cataloging ranked by popularity.  Yes, there are conceptual relationships between articles and web sites that are expressed when two or more are tagged with the same tag and these relationships can grow in very complex ways revealing a great deal about how people think and how some ideas are can be related.  But it’s not clear how the overall set of conceptual relationships constitutes an organization of knowledge.  This becomes particularly clear as the number of tags and sites multiply and the complexity of the tag and community relationships grows exponentially.  Gee, haven’t we seen this process before? 

On the other hand, it is possible to browse through tags and citations and pick up a number of ideas of how other people have tagged a particular set of articles/web sites and thus be exposed to a variety of connections between concepts.  This can be fun, it can be rewarding, but it is ultimately rather limited in its impact.

Perhaps the most fundamental limit of folksonomies for future “revolutionary” impacts is that there is really no mechanism for improving the quality of the individual tags that is consistent with their most powerful characteristic, ease of use.  I just don’t see flocks of users going back and re-tagging or even necessarily getting better at tagging simply by being exposed to other people’s tags, in fact many sites don’t allow it.  The enthusiasts might very well learn to use folksonomies to improve their own tags, but I really wonder how widespread that behavior will be.

And so, the second conclusion is that folksonomies are interesting but not really very revolutionary.  User tagging has been around for a long time and the relationship between free keyword tagging and classification schema or taxonomies has been and will continue to be a dialogue in which hybrid systems will almost always be the best answer. 

So, in my opinion, the future of folksonomies lies not with the revolutionary rhetoric which actually detracts from the real value of folksonomies but with a range of new and interesting ways to combine user-generated tags and taxonomies.  To get an idea of what is going on in the hybrid world, take a look at some of the articles cited in the bibliography.

True, these new hybrid systems of taxonomy/classification schema and folksonomy won’t satisfy the anti-authoritarian strain of thought that runs through folksonomy literature, but I will predict that they will ultimately deliver much more value to many more people in many more environments than pure folksonomy applications ever will.

Bibliography

Ontology is Overrated – Categories, Links, and Tags by Clay Shirky
The standard starting point for folksonomy discussion

Folksonomies: power to the people by Emanuele Quintarelli
A very good summary of the standard claims for folksonomies and the drawbacks of taxonomies

Ontology of Folksonomy:  A Mash-up of Apples and Oranges by Tom Gruber
A good discussion of social tagging, ontologies (distinguished from Clay Shirky’s use to mean hierarchical classification schemas), and a variety of semantic issues.

Collaborative thesaurus tagging the Wikipedia way (v2;2006-04-27) –[[Wikimetrics]] research papers, volume 1, issue 1.  by Jakob Voss
Fascinating discussion of combining thesaurus categories and user tagging.

The Structure of Collaborative Tagging Systems by Scott Golder and Bernardo Huberman
A very good article analyzing tagging behavior and various cognitive and semantic aspects of tagging and taxonomy, including some of the surprising patterns over time within sites like Del.icio.us, CiteULike, and Connotea

Creating the academic library folksonomy:  Put social tagging to work at your institution by Xan Arch
A call for combining library and folksonomy

Cataloging and You:  Measuring the Efficacy of a Folksonomy for Subject Analysis by Tiffany Smith
An analysis of folksonomies at LibraryThing, including various quality issues

Beneath the Metadata:  Some Philosophical Problems with Folksonomy by Elaine Peterson
An interesting discussion of some of the underlying philosophy behind folksonomies (strong relativism)

Structure and form of folksonomy tags:  The road to the public library catalogue by  Louise Spiteri
A good analysis of folksonomy tagging including the overwhelming use of thing tags, and that for libraries to use folksonomies they have to design ways to overcome the problems with bad tagging.

Taxonomy Directed Folksonomies:  Integrating user tagging and controlled vocabularies for Australian education networks by Sarah Hayman and Nick Lothian
A good overview analysis of folksonomies and user tagging, including ideas for combining folksonomy and taxonomy.