Home
Jun. 9th, 2005 @ 08:50 am more on folksonomy
Little did I realize what I was getting myself into with the digital libraries site at del.icio.us:
http://del.icio.us/raharris/

Some changes:

1. My original folks-onomy included make (digital library or something else), model (collections, projects, centers, etc.) and some subject descriptors (history, literature, women, etc.). But is also included geographic locators: Regions > (Americas, Africa, Asia, Europe) > (U.S., Namibia, China, U.K., etc.) > even states in the case of the US. But then I got to thinking -- since this is online one can access an Asian site from the US and a Eurpean from Africa, so why is geography important? I guess I included it because my blighted old arse was born and brought up in the pre-cyber world where geography was an important indicator of access. If something was in Europe one could potentially fly there, but most things behind the Iron Curtain were off limits. In cyber reality geography becomes less important, and in this case not significant enough to make it into my system of classification.

With some exceptions. When the location is a subject matter rather then a mere geographic locator it is included. For example the Illinois Alive! site concerns the history of Illinois and thus earns a tag by that name. However the Digital Projects at the University of Illinois  (Urbana-Champagne) holds collections not relevant to location and the Region > Continent > Country > State classification is eliminated.

2. Originally I was thinking in terms of digital libraries, but the more I poked about the more I became aware of another important classification: digital archives. The first concentrates on text resources while the second collects material artifacts, sound and image recordings, etc. The US Labor and Industrial History WWW Audio Archive (Cambridge) stresses, as one might expect, sound recordings. So I've added another major type to the digital.libraries designation, digital.archives The more experienced browsers among my readership (which as far as I can tell is comprised of just one person: hello pruneprisms!) will note the problem right off: while it might be important I make the distinction between archives and libraries (I think it is) what is to be done in the case of collections that straddle the line? For example the Emma Goldman Papers (Berkeley) contains both text and material artifacts.

My solution thus far has been to enter both designations in such cases, but that is somewhat akward. What are the other possibilities? The only logical option that comes to mind right off is that I could come up with an overarching term and sub-designate collection type: text, material, images, multimedia, etc. While that sounds good I see two problems: first, what term would overarch both libraries and archives? Second, with over a hundred entries I'd have a lot of back-tracking to do . . . maybe I can just convert existing tags for digital.libraries . . . no, that won't do it . . .

Which brings me to my final point for the day -- two actually. 1. It would have been better to have come up with the folksonomy before I started applying it. The problem with that is that you don't know what you'll need until you are knee deep in it already. 2. There is just a shit of a lot out there. Some of it is indeed shit, some not. What to include and what not? One of the criteria I've used is that anything involving history or literature, but especially history, is in right off. Science oriented sites and the like get a much more robust screening. One result there is that only the best of the science sites have made it into the list, while all sorts of history sites have.

What other criteria could I use besides personal interest? Perhaps the quality of the site. But how does one assess that? It's difficult (despite the fact that above I claim to have done it for science sites). Does one go by the looks of the site, how often it is maintained and updated, navigability, quality of content? And what are the benchmarks in any of those cases? The use of metadata to describe the site? I'll tell ya', some of the metadata sites stink and some that don't employ are great, IMHO.

So anyway, considering these questions is fun. I imagine if I had more formal experience in the subject, but it's interesting to work these things out on one's own -- Cheers, Robert
About this Entry
monksville, punts, bridge, four dogs, lookleft, oxford, raharris, walkway
Leave a comment
Permanent Link