(week 5) collabularies

To quote a well known author (me!), from my article 'Folksonomies and collabularies' in Digital Digest #8:

'[Tag searches work best as] ‘fuzzy’ searches (1). An initial, manageable set of search results will provide a starting point from which searchers navigate out to other items of interest. Therefore, one of the most valuable outcomes of social tagging is the way in which it helps reintroduce users to the joys of serendipitous discovery. It facilitates browsing; a valuable search technique which many online information collections fail to support. Encouraging this exploration is important, as ‘the question people ask often does not relate exactly to what they need to know. By facilitating the discovery process, folksonomies may help one find out what they need to know, rather than just what they asked for’(2).

(1) Richard Giles, “Top Ten Tech Trends,” (plenary session at VALA2006, Melbourne, 9 February 2006).

(2) Kroski, “The Hive Mind: Folksonomies and User-Based Tagging”, Infotangle (7 December 2005).

Come to think of it, when I was researching the above article I found many useful websites by looking for links other people had tagged on del.icio.us. Which could be called laziness...but I like to think of it as collaborative learning.

I have to admit that, much as I love the idea of del.icio.us I'm constantly forgetting to make use of the handy buttons that I've added to my Firefox toolbar. I find myself having to go back and transfer links from my bookmarks folder to my del.icio.us account. Which is probably a good thing, as I'm an inveterate bookmark hoarder, and I'd hate to think what my account would look like if my bookmarking trigger finger had free range there! (Scuse the mixed metaphor)

As far as the application of tags to our catalogue goes, I have mixed feelings. I do know that many useful resources are buried under layers of formal/longwinded nomenclature that make them almost impossible to locate (eg if you want to view a copy of someone's will from Victoria prior to the 90s, you must consult the Probate Index Victoria 1841-1992 - an interesting task for the beginner or indeed anyone using our catalogue, as the entry for this index does not mention the word "will" anywhere). By allowing catalogue users to apply their own descriptive language to such entries (in addition to the traditional catloguing taxonomy) we may enable our users to better find the resources they're after. Our users may also illuminate hidden sources (eg a book on a broad topic may included a detailed study of a particular topic or an in-depth statistical appendix, which would not be obvious to those reading the catalogue entry, but which users of the book could tag).

But...would tagging actually work that way? Would we have enough people tagging regularly for the process to endure and remain relevant? We are not a local library with a dedicated clientèle clamouring to share their thoughts like they would with the local book group.

What would be the cost of moderating such a system to ensure that offensive tags weren't being posted? In a study of del.icio.us HP labs found that ‘a stable tag pattern emerges after the first one hundred bookmarks are placed for a particular website’, with inappropriate tags hidden by the sheer volume of frequently used tags and useful tags. I seriously doubt that we would experience anywhere near that volume of use...

Hmmmm.