“Good enough for who it’s for.” That’s one of my favorite lines. Elaine Peterson’s earnest exploration of the shortcomings of folksonomic classification brought it bubbling up to the window in my Magic 8 Ball brain. Peterson criticizes David Weinberger’s essay, “Tagging and Why it Matters,” which is a year and a half old now and standing the test of Internet time quite nicely. (Dr. W. responds to that criticism in a blog post here.)
Elaine, she say,
Weinberger also mentions as benefits financial savings and elimination of bureaucracies of catalogers and indexers…
So there you have it. It’s fundamentally a rice bowl issue. All her Aristotelian window dressing and the heavy concerns she expresses regarding “philosophical relativism” and her concern for the “average user” pivot around the threat she feels that tagsonomies are somehow going to drive out card catalogs and the jobs of the catalogers.
Rest easy, Elaine. The authorial tagging of blog posts and the readership tagging of web pages, the gathering of the tags, the sifting of the tags, the serendipitous appearance of the Portland Rose Garden in a collection of photographs of Marilyn Monroe, all these things are good enough, more than close enough — one might say — for government work, were one to ignore the enormous accuracy required of government workers from the days of the Egyptian pyramids to the Apollo Project. I’ll submit that imprecision is the nature of the tagsonomic game, whereas accuracy and refined structure are the balance points for library classification schemata. Library classification work will be with us a while longer yet, I think.
In the average huge university library there are a finite and inventoried number of volumes and someone has been kind enough to categorize each one by author, publication date, title, publisher, topic, sub-topic, even by size. Each has a place on a specific shelf and a placeholder in a checkout system to track it if it has been removed from the shelf. That stuff works. The great cataloging and classification systems: Library of Congress, Cutter, Dewey and whatnot… they work. And they’re good enough for who they’re for… the patrons, the librarians, researchers, bibliographers. We know what’s available and we know how to find it, or how long we’ll have to wait for it to return from the bindery, or whatever.
The grand folksonomic systems, del.icio.us, FlickR, etc. don’t pretend to address those needs, don’t pretend to that level of taxonomic integrity, but rather function to collate huge volumes of stuff and leave it up to the user to address further curatorial issues.
All this stuff is good enough for who it’s for, indeed it all dovetails nicely card cataloigs providing access to the untagged world of dead trees and profitable eJournals… tags collecting work by new creators. I’m looking forward to tag based bibliographies… the oeuvre of Amanda Congdon, collated and published on DVD with technorati tags and unique google references driving out the old academic formats, ibid, op cit, say what?
Frank,
Mind if I disagree with two points, although I think we actually agree about the second one, and maybe about the first one, too?
1. Judging purely from Elaine’s article, I don’t think her concern boils down to economics. I think she’s worried about the ol’ metaphysics: Folksonomies allow contradictory statements to exist, which means the world isn’t one way and not another.
2. And I actually don’t think the old fashioned taxonomies work very well. We’ve gotten so used to them that theiir limitations are invisible to us. Even the best of them, in which objects are expertly and knowledgably arranged by some important principle of similarity (say, topic) have necessarily forced us to actualize only one of the many, many, many useful ways the objects might be arranged. There are better and worse taxonomies–and it’s no small skill to come up with a better one–but the essence of an old-school taxonomy is that it forces precision along a single axis. Or maybe it can afford 3-4 axes as with a card catalog. If those axes happen to accord with the user’s needs, and if the particular categorizations accord with the user’s way of thinking, then that taxonomy works for that user. But, holy mackerel, that’s a huge limitation. It’s like having a piano with only enough notes to play “Happy Birthday”(tm) and saying it works just fine.
Nevertheless, I do agree that folksonomies don’t give the consistency of classificatory principle, or consistency of worldview (as Clay would put it) that single-source taxonomies do. But I don’t see this as single-source taxonomies vs. folksonomies. Rather, in the digital age (or, as no one but me would say, in the miscellaneous age), we can have as many ways of sorting as we’d like. We’re not moving from taxonomies to folksonomies. We’re moving from a single actualization to indefinite potential…with plenty of room for more single-source taxonomies than ever.
Don’t your lawyers want you to say, “In the Miscellaneous Age™ ”?
Regarding number one, I agree with you and I was just taking the cheap shot. (I also agree with her as you’ve framed her position… the world IS one way and not another. But the nature of a loose and informal arrangement of metadata by non-professionals doesn’t contradict that world view, it’s just a messy way to extend the boundary of our universe of classified information.)
As for number two, I’m back to “good enough for who it’s for.” An argument against a card catalog as a super way to locate one book in stacks containing millions of them doesn’t fly. An argument FOR the value of topical metadata associated with content as assigned by someone in the Library of Congress is a different matter. Being able to find the book is a single dimension of easily fixed value. It’s binary. With the catalog I can find the book. Without it, I can’t. Somehow “knowing” what is in the book based on classification is more elusive, subjective. Elaine suggests that the professionals add weight to this dimension, and the amateurs add little. But this starts to sound like the arguments against Wikipedia. The compilation does not contain all truth perfectly expressed. In fact it contains contradictory and sometimes inaccurate matter. Therefore it is worthless? Not at all, of course it is useful. Just as a card catalog is useful and just as folksonomic clouds of gathered information are useful.
I think we don’t disagree too much, except maybe about fundamental values related to the existence of good and evil, the allegorical significance of the coyote in the old Roadrunner cartoons, and just what the ACME company symbolized. Oh, and another difference might be my stubborn attachment to common organizing principles that enhance a common understanding of the world around us and the systems we use to lend structure to those principles.
But my bottom line in this whole discussion is: as long as the library (or any old information repository) can make accessible to me the book or the data I’ve requested, then I can take it from there. Hooray for tags! Hooray for Melvill Dewey!
Frank, I have to disagree with you about ACME, but only as a metaphysical statement. Whatever that means.
For some reason, on this day before thankfulness, I feel the need to be curmudgeonly. So:
Yes, you couldn’t find a book in a physical library if there were no catalog. But physical card catalogs are so limited that they’re rapidly being tossed out in favor of computer-based catalogs that allow sorting on many more keys. And those computer-based replacements will be absorbed by systems that blow open the metadata gates, letting us search by content (“What book begins ‘Call me Ishmael’?), by user ratings, by user tags, by all of the data and metadata linked to any other book-related data on the Internets. This is something we agree on, I’m pretty sure.
So, then I think the question is: What are the unique benefits of card catalogs and of single-source taxonomies. There are such benefits. They’ll add to the mix we’re brewing. But I don’t think ultimately they’re going to be anywhere near as important. After all, Amazon has a vital economic interest in helping you find new books, and it’s top-down taxonomy is just about buried…
For organizing and retrieving information, I like computers better than card files, but those oak cabinets with row after row of drawers each fronted with a little brass handle and a label frame, the handy writing boards you can pull out that permit the jotting of a note, well… when the last of those is gone, replaced by computing machinery, the world will be a poorer place.
On a brighter note, ACME… umm, that is AMAZON just sent an email to inform me that my copy of ISBN 159420120X has shipped.
First, the idea that terms assigned from taxonomies enforce precision along a single axis is not true. In the Library of Congress Classification system, single call numbers were assigned because the physical book could only be in one physical place. But one could assign how ever many classification numbers (as opposed to call numbers) one liked to express different facets of the content. Far more often this was done through assigning multiple human readable subject headings, but in principle they are the same. The limited number of terms applied was based on both economics and, to a certain extent, principle: librarians tried, with the tools available to them, to express the relative importance of a given topic in the work, and therefore improve the signal to noise ratio in search results. Of course it’s flawed, but it continues to have value.
Librarians have been there, done that, 15 years ago already. And got clobbered for it by the likes of Nicholson Baker. Honestly, we can’t win for losing.
I love Amazon to death, but they are starting to run into the same sorts of problems libraries have. Have you seen the “Style” drop-downs on an individual product page, which list half-a-dozen different products? Figuring out what the reviews actually apply to, what you’re actually buying, is becoming harder and harder.
I am actually all for both formal organization and tagging, but I share some librarians concerns that cost-conscious administrators see tagging as a way to eliminate formal organization activities, or at least reduce them to the point where their value truly will be neglible.
I expect tagging to be overtaken by commercial interests. Today, if I search for information about a brand of old piano, my first 75 hits in Google will be for sales of said piano, moreover sales that were over months ago!
A recent search of mine turned up as its first result: “Looking for Formal Specifications? See the wide selection and low prices on eBay!”
I expect tagging to suffer the same indignities that library subject headings suffer as language changes. Libraries update their headings to spare societies’ sensibilities. What happens with tags? Etc.
What many librarians fear is not the loss of their jobs. They fear the unthinking destruction of something valuable in the name of something new, the implications of which are not fully understood. And they are getting just a little ticked about people “discovering” concepts that have been the foundation of librarianship for over a century, acting as if no one has ever thought of them before.