Another visual exper­i­ment on tag­ging: How do indi­vid­u­als use tags — com­pared to the com­mu­nity? Do you use pri­vate lan­guage or are you a main­stream tag­ger? When you tag with “design” — do the oth­ers think it is “art”?

On the left: the per­sonal tags for the per­sonal book­marks — ordered by fre­quency — the con­tain­ing box per is log-scaled so you get an impres­sion of the long–tail posi­tion of a tag. Which means: Often used tags are large, bright and go to the top.

On the right: com­mu­nity tags for the tagged ressources.

In the mid­dle: tags with the same name are con­nected. If a line is hor­i­zon­tal, the ind­vid­ual and the com­mu­nity essen­tially agree on the rel­e­vance of the tag for the ressources. The steeper it is — the larger the dis­agree­ment. If no line starts at a tag, it means it is not present in the other list

So in the pic­ture above, you can see my tags. Same facts you can read from the pic­ture: • “m.a.thesis” is a very often used, but pri­vate tag of mine. • The ressources I tag in gen­eral are mostly tagged with “design” by the com­mu­nity. I, how­ever, use the tag “design” much less often. • “news”, “seman­tic web”, “web­dev” are tags I use often, but not the com­mu­nity. etc.

And often course, you can click indi­vid­ual tags to see what the com­par­i­son is like for sub­sets of the book­marks. That’s espe­cially inter­est­ing for obscure tags like “guru” — you can see what the tag­ger “means” by look­ing at the dis­tri­b­u­tion of the com­mu­nity book­marks (in this case “design — art — pro­gram­mer — artist”). Interesting!

Some more shots:

I wish I could say “click here for the inter­ac­tive ver­sion” as usual — but unfor­tu­nately, del.icio.us offers a JSON API, but did not put a crossdomain.xml file on their server. Which means the visu­al­iza­tion (which runs nicely on my hard­disc) can­not load data when put in the web. Bum­mer. I hope I can fig­ure some­thing out.

So for now — I can only offer a down­load link. Click the index.html. You might have to adjust you Flash player secu­rity set­tings in order to load the com­mu­nity tags. Caveat: The appli­ca­tion is still a bit buggy and pretty heavy con­cern­ing proces­sor ressources.

A very nice paper on the “laissez-faire librar­i­an­ship” often asso­ci­ated with tag­ging vs. more struc­tured seman­tic web approaches. Most notable is that the dis­cus­sion is put in the con­text of Husserl’s the­ory of reflec­tions, inten­tion­al­ity and intersubjectivity.

A PHENOMENOLOGICAL FRAMEWORK FOR THE RELATIONSHIP BETWEEN THE SEMANTIC WEB AND USER-CENTERED TAGGING SYSTEMS

D. Grant Camp­bell Fac­ulty of Infor­ma­tion and Media Stud­ies Uni­ver­sity of West­ern Ontario
Lon­don, Ontario N6A 5B7, Canada

Abstract This paper uses Husserl’s the­ory of phe­nom­e­nol­ogy to pro­vide a model for the rela­tion­ship between user-centered tag­ging sys­tems, such as del.icio.us, and the more highly struc­tured sys­tems of the Seman­tic Web. Using three aspects of phe­nom­e­no­log­i­cal theory—the move­ment of the mind out towards an entity and then back in an act of reflec­tion, mul­ti­plic­i­ties within unity, and the shar­ing of inten­tion­al­i­ties within a community—the dis­cus­sion sug­gests that both tag­ging sys­tems and the Seman­tic Web fos­ter an inter­sub­jec­tive domain for the shar­ing and use of infor­ma­tion resources. The Seman­tic Web, how­ever, resem­bles tra­di­tional library sys­tems, in that it relies for this inter­sub­jec­tive domain on the con­scious imple­men­ta­tion of domain-centered stan­dards which are then encoded for machine pro­cess­ing, while tag­ging sys­tems work on implied prin­ci­ples of emergence.