New Media Mania

To suggest that social classification systems will replace formal traditional classification on the web is to imply that there is no place on the web for bodies of professional, scientific, or other specialised knowledge that possess standardised nomenclature critical for mutual understanding.

Conversely, those who would argue that social tagging has no place in highly specialised fields are being short-sighted. Clearly there is a place for both traditional and social classification systems, and indeed hybrids, on the web.

Shirky proposes conditions in which formal classification systems work well and are necessary.

“Shirky’s conditions” pertain to the type of knowledge being presented and the nature of the creators, guardians, and consumers of the knowledge:-

Knowledge

• Small corpus
• Formal categories
• Stable entities
• Restricted entities
• Clear edges

Participants

• Expert cataloguers
• Authoritative source of judgement
• Coordinated users
• Expert users

The flip-side to Shirky’s conditions is a good description of the web, broadly speaking. Where there is a large, ill-defined body of knowledge, layperson users, and cataloguers that are not particularly expert, and where there is no authority to provide oversight (or it would be a prohibitively large task), rigidity in a classification system will cause problems.

However, clearly specialised knowledge with well-defined audiences can and does exist on the web, such as in industry- or profession-specific knowledge portals.

According to Peterson, using folksonomy to organise information is a decision with important philosophical implications about the integrity of knowledge.

She writes: “There are no right or wrong classification terms in a folksonomic world … Folksonomists are confusing cataloguing structure with personal opinions and subsequent social bookmarking. These are not the same thing, and they need to be separated.”

In that same spirit, I posit that where traditional classification is appropriate and necessary according to an analysis of Shirky’s conditions, a user-centric tagging system can still work:-

1. If it improves the personal experience of the user; AND
2. The tags do not alter the default classification system.

For example, in the case of an online drug reference database, a physician’s tag information could be used to present content to her upon login, as a personalised alternative to the generic welcome information.

Furthermore, to introduce the social aspect of tagging (and its potential in helping to build an online community) the doctor could be given the opportunity to share her tagged content with other registered users of the portal – her “friends”.

Thus, even with a formal classification employed for vital knowledge assets, such as the specific “pharmacodynamics” of drugs, doctors who share the same specialisation, hospital, college education, or political persuasion, might still enjoy a tailored and social experience by using tags:-

1)    To organise content in ways that makes the reference site more convenient for them; AND
2)    To share information with their peers about topics of mutual interest.

A further benefit of offering a user-centric tagging system in addition to (but without directly affecting) traditional classification lies in the potentially useful feedback it can provide a knowledge manager in any future attempt to change the traditional system.

Whichever way we choose to approach the issue of traditional versus social classification systems, a knowledge worker must always be aware of bias, both in the content structure she presents to her audience and the potential for it to creep in from outside.

As Morville put it, “[while] all taxonomies embed bias, the same can be said of search engines, books, blogs, Amazon, eBay, and Wikipedia. This doesn’t negate the value and good intentions of [professionals and amateurs] who labour to improve find-ability, accessibility, and understanding for all … We must all be more aware, as consumers and creators, of the incentives, biases, and weaknesses inherent in all sources and structures of authority and knowledge.”

In the search for the perfect classification system there is no one-size-fits-all. However, the best of both worlds is eminently possible.

Some sources:

Morville, P; May 2, 2007; “(Not) Everything is Miscellaneous”; http://semanticstudios.com/publications/semantics/000167.php

Peterson, E; November 2006; Beneath the Metadata: Some Philosophical Problems with Folksonomy; ‘D-Lib Magazine’, Volume 12 Number 11; http://www.dlib.org/dlib/november06/peterson/11peterson.html

Shirky, C.; (2005); “Ontologies are overrated: Categories, links, and tags”; http://www.shirky.com/writings/ontology_overrated.html