Panel 4: Conceptual Frameworks for Social Classification

An examination of authority in social classification systems [PDF]
Melanie Feinberg (University of Washington, USA)
http://www.slais.ubc.ca/users/sigcr/sigcr-06feinberg.pdf

different kinds of social classification: my tagging for me, vs group tagging
- questions raised in their study...

- where is the social in social classification? What is the process for reaching consensus? doesn't it have different kinds of socio/political characteristics that emphasize the individual?

Reference to Wired/long-tail and sales: but economic arguments of the long tail aren't just about sales, they are about the incremental costs of carrying additional inventory in digital vs electronic environments.

- where are the means to define community-based sematics in the current social-classification tools? What defines the community?

- what are the complementary benefits of traditional classification and indexing?
- what do people need to know to be able to create systems of organized knowledge / requirements of information literacy?

jb: what does consensus have to do with it?

- emergent patterns might actually be representative of communities themselves.

A phenomenological framework for the relationship between the Semantic Web and user-centered tagging systems
D. Grant Campbell (University of Western Ontario, Canada)
http://www.slais.ubc.ca/users/sigcr/sigcr-06campbell.pdf

- new ideas drawn from the discussions in this workshop [jt: in the perishable aisle?]
- paper grew out of a larger study of the semantic web, which itself changed in the context of web 2.0
- how is social tagging different from the semantic web? what is the relationship between libraries and the library community to this?
- Foucault as the baseline -- but starting with Husserl and a dialogic relationship between the individual and the world, between a person and concepts, that uses concepts held internally as a referent. Shared inter-subjectivities, community of shared concepts, underpin collective understanding. "overlap sufficiently so that we can converse"

- conflict between inter-subjectivity defined by control (vocabularies, ontologies) and one defined by emergent semantics (tagging), that pits consistent activities by trained experts which create contrived and elaborate structures against against loose, evolving structures created by end-users of information.

- these two visions seem to be in conflict / represent different polarities, and that raise some concerns:

- web 2.0 is cool, sexy and cheap -- jt: what a dangerous triumverate! -- (but the sematic web might be closer to the kind of structured systems that we need in libraries)
- blog and IA lit. suggests that the semantic web is a failure and that tagging is a replacement
- 'emergence' is a troubling concept - self-organization seems a 'miracle' -- but are the circumstances right for this thing to grow in a good way? might this be presented as an alternative to what libraries do?
- expert knowledge is valued for a reason. diversity is necessary

q: marcia bates - research on folk-classification vs classification that's done by experts
LSCH volumes have headings and rules but many people don't use the instructions for how to apply subject headings.
a: response explores relationships between individuals and experts

q; corinne -- need to respect the localized expertise

jt: diversity is good; people are tagging because they feel a need to add something / augment data.

marcia: complexity in vocabulary needs to be represented in retrieval

jt - are tagging and the semantic web in conflict? --> I asked...

Social tagging and the next steps for indexing
Joseph T. Tennis (University of British Columbia, Canada)
http://www.slais.ubc.ca/users/sigcr/sigcr-06tennis.pdf

- what are the similarities/ differences between social tagging and subject cataloguing (both as species of indexing) compare Fordist and post-Fordist approaches
- what is subject indexing? What is subject description?

Image for indexing -- posts of an old pier - order, forethought, investment
images for tagging -- gnarled tree - organic, live

tagging moves beyond content to process
- multiple purposes require thicker descriptions
- need a broad range of frameworks for indexing

grant: are we saving our buts by "scrambling up to the metalayer"? I'll be able to sort it if i can only get above it...
joe says 'yes' -- needs to do this to understand; there is a purpose behind the need for 'indexography'. need to have frameworks to be able to compare what's going on.

jt: when is social tagging not subject indexing? why is the assumption here that tagging is only about content analysis, when so much work about tagging identifies it as situationalist and process-driven?

q: how do you separate classification/clustering and indexing/retrieval functions in the analysis?

emma: there is a lot that can be learned and then applied using tag-systems as the basis for understanding tagging systems and complex systems

marcia: we are at a point in the development of the approach that is temporally defined. things are changing, and we don't know where it is going. it's difficult to generalize when the change is happening.

q: is a tag like a citation? is it a reference that validates something, or an assessment of validity.

Jonathan Further Wraps up
- sig/cr next year?

- visual resources
- expertise
- visualization
- web 3.0
- collaborate with sig-use and explore both sides of ir -- see http://iats-coe1.missouri.edu/~siguse/index.php
- giving away: problems my research offer you [things that someone else could be able to pursue]
- tagging for yourself or tagging for others
- tagging isn't always indexing