What Is to Be Tagged?

I’ve circled back to take another look at GoogleBase because of a weblog post by Rashmi Sinha, “The blooming of information architecture at Google: A close look at facets, tags & categories in GoogleBase.” I did this because I’m a sucker for taxonomy, poor me — and because it led to some speculation about tagging in law. And as for GoogleBase, there’s something going on here that isn’t quite clear yet. I feel like that guy in the Tom Waits’ song (well, “song” is an exaggeration, if you know Waits) “What’s He Building In There?” who puzzles and puzzles with some paranoia about a closeted neighbour. (Here’s a possibly illegal 20-second clip.)

Sinha explores the way Google — “Google of the simple search box with a go button,” as she puts it — has embraced information architecture and has now gone all the way to categories, facets, and tags in Base. Google even lets you name your own category if none of the top-level categories fits; though it isn’t clear to me how this is different from simply tagging. Still, it’s a message to Google that their structure is either inappropriate, insufficient or misunderstood. Facets, we know, are like cross-indexing categories: if you go for the category Recipes (and who wouldn’t?), you are given the chance to refine your search with facets such as “main ingredients” and “cuisine.” Again, you can name your own facets.

But there are also tags, though not so-called. Google seems to be blurring or fudging the taxonomic structure here. Perhaps, as Sinha suggests, these Google tags are simply keywords – which, after all is a tag that resides in the text, no? But it seems to me that Google is testing something here, rather than building a store of knowledge.

Maybe Google’s just getting on the bandwagon, because tags are all the rage, tags and folksonomy. I’m comfortable with traditional taxonomy — its categories and top-down organization — but tags are all about bottom-up, organic ontology: let thousands of flowers bloom thousands of times and see what emerges — “tag clouds” apparently — an emergent taxonomy, unpredictable but very present, like the market, democratic in its way, but of unclear utility. (Required reading on tags and internet taxonomy: Clay Shirky’s “Ontology is Overrated: Categories, Links and Tags” and Tom Gruber’s “Ontology of Folksonomy: A Mash-up of Apples and Oranges.”) What makes me feel comfortable is neither here nor there for the internet, and tagging is only going to get bigger. The interesting question is how close will it get to us?

I happened upon an article by Hawaii information science professor Peter Jasco, “As We May Search” [pdf] [Current Science 89(9), 2005, p. 1537] in which he compares aspects of Google Scholar, Scopus and the Web of Science, as citation-based and -enhanced databases. Now citation is something lawyers know a bit about. Sure enough, Jasco discusses Vannevar Bush and Eugene Garfield, and then spells out their debt to Frank Shepard of Shepardizing fame (see Simon Chester’s post below). Which got me thinking: the name of a case or its citation is a lawyerly tag for all of the items within it. Items in the folksphere may have many tags — as many as there are taggers, oddly. Thus, in law-tagging, ‘Case A’ can be tagged with the tag ‘Supreme Court Case X’ or with the tag ‘Ontario Municipal Board Matter Y’ because it may well have been referred to in both.

This is not trivial but it is not news. And to lawyers it must seem forced to treat it as tagging. But it suggests that tagging in the more popular sense need not be as foreign to law as we may have supposed: as with ordinary taggers we care what others think about issues and decisions and rules out there; we care how these are “seen” and labelled; we are swayed by majority views and by authority views. Moreover, like a folksonomy, the common law works instance by instance and is an emergent property, not entirely predictable, always coming into existence without arriving; it’s a bottom-up process operating via solicitor’s letter, argument, negotiation, lawsuit etc. case by case by case. Only when we look back and take stock can we see which opinons/views/labels loom large in the legal tag cloud — in nubibus, as we might say in law.

So now I wonder if we’ll ever see a quicker, simpler version of legal tagging to allow legal views as to potential relevance or value or correctness to gather and be bases for communities as tags are now in non-legal areas. How might it work, make sense, be useful?