Consider the eighth category, which deals with stones. Wilkins divides them into the following classifications: ordinary (flint, gravel, slate); intermediate (marble, amber, coral); precious (pearl, opal); transparent (amethyst, sapphire); and insoluble (coal, clay, and arsenic). The ninth category is almost as alarming as the eight. It reveals that metals can be imperfect (vermilion, quicksilver); artificial (bronze, brass); recremental (filings, rust); and natural (gold, tin, copper). The whale appears in the sixteenth category: it is a viviparous, oblong fish. These ambiguities, redundances, and deficiencies recall those attributed by Dr. Franz Kuhn to a certain Chinese encyclopedia entitled Celestial Emporium of Benevolent Knowledge. On those remote pages it is written that animals are divided into (a) those that belong to the Emperor, (b) embalmed ones, (c) those that are trained, (d) suckling pigs, (e) mermaids, (f) fabulous ones, (g) stray dogs, (h) those that are included in this classification, (i) those that tremble as if they were mad, (j) innumerable ones, (k) those drawn with a very fine camel's hair brush, (l) others, (m) those that have just broken a flower vase, (n) those that resemble flies from a distance.I met this quotation in a lecture about classification at UCL. The vivid imagery is a tonic for a tired mind and the quotation is certainly a useful tool to illustrate the deficiencies of a poorly-assembled classification system. The conclusion of the short article from which it is drawn is in fact that it is not really possible to construct a universal classificatory system for the world, and that any attempt will be too subjective and insufficiently comprehensive. But I like this quotation particularly for the way it, especially the extract from a fictional Chinese encyclopaedia, invites the reader to imagine a world in which such an impossible taxonomy makes sense. Can we envisage a Chinese court in which the significance of animals is defined by these arbitrary-seeming categories? And, to sufficiently foreign eyes, how arbitrary would our everyday groupings also seem?
Jorge Luis Borges, 'The analytical language of John Wilkins' in Other inquisitions, 1937-1952, trans. Ruth L.C. Simms (Austin: University of Texas Press, 1964), 101-105, p. 103.
This distinction, between arbitrary-but-locally-appropriate and aspiring-to-universality-but-failing-in-the-details is not a new one in classification, but allowing Members of the Public to publicly create their own classification systems and to assign classifications to collections held by institutions seems to be at the heart of interest (or concern) over the phenomenon of tagging. These are points made in the Clay Shirky article (required Cam23 reading). I've read it, and it's inspired many of the thoughts in this post, but I won't be engaging with it in detail here.
Personally, I love tagging. I've been tagging my personal photos on Flickr for a while as a means of retrieval independent of the time/event-based sets and collections I create for the pictures (personal names blanked out except for a rogue David and Matthew that slipped through the net):
My tags for this blog are a mixture of factual and whimsical. '23 things' is in there as a tag so that I can split posts about things from posts about government cuts, charteship, CILIP and the weather. Humurous content tags 'I only listen to radio 4', 'when I was young', 'outraged of Cambridge' might help new readers to get a flavour of my writing without having to read a post. Or they might, in a few weeks' time, just look naff and annoying. If that sort of thing does float your boat, do have a look at the #regionaltagging discussion [archive at http://twapperkeeper.com/hashtag/regionaltagging] on twitter and discussion and tags on other blogs. These 'personal' tags are one end of the broad spectrum that Shirky discusses with regard to the benefits of letting users define things in their own language.
The Skirky article suggests that we can do away with hierarchy and order in classification, and this is something that I'm reluctant to agree with. Sometimes a hirearchical and/or ordered structure is the easiest to find something, or to evaluate what somethings are available. I do use traditional subject sheadings to find content (I even used the UL guard books once because they have a grand list of liturgical books ordered by type of book, Use, date, etc. and I was looking for a Sarum Use missal from as near to a particular date as possible. I've still not worked out how I could have done that more easily in Newton). And for as long as libraries have things on shelves, it will be useful for the broadly related things to be broadly near to each other, even though we'd need n-dimensional L-space to get all related things into physical proximity.
Combining tagging and traditional approaches is probably the best way forward. This is happening at Ann Arbor District Library, and also on the new British Library catalogue in beta. The success of user-tagging in these contexts rests on users actually adding tags, and it's clearly still early days for the BL. Using tags to discover content is an interesting, almost stream-of-consciousness experience, and can lead to unexpected discoveries, but I think the mechanisms need to get better. I want to be able to search for things tagged with several tags all at once. Can I do this easily now? Not everywhere. I can on Flickr, and very easily Delicioius, but not on Blogger and not obviously at the BL or Ann Arbor. It's not enough to have a search box into which I type some words, I want to be able to highligh multiple items from a list and see if they all apply together to something, and I don't think that this should feel like an added bonus feature.
