Books and LibraryThing II

O.K., I’m still thinking about books and LibraryThing. There was a lot of noise about LibraryThing last week, which mostly seemed to come from the slashdotting of its new feature “the unsuggester.” A neat gimmick, and I hope it steered a lot of people to the LibraryThing site.  I think that for people who love books and are always looking for books that they will like, and aren’t willing to slog though bad reading suggestions, LibraryThing is something to take very seriously. Today a short speculative post on metatags, folksonomies, and LibraryThing.

Tagging, as I’m sure you know, is a great way for individuals to organize their books. But it is also an important new research tool that, with the advent of LibraryThing, allows everybody to see how other people organize their books, and what they think about them. This is the beauty of tagging in any environment, whether using del.icio.us to organize websites or tagging on your own blog to make things easier to find in technorati. But there’s another possibility that I’d like to consider, and that is a sort of hybrid combining cataloging systems, which I call user-created, community standardized demographic metadata. Adam Mathes calls traditional cataloging, like Library of Congress or the Dewey Decimal System, expert-assigned systems of categorization—professional- and author-created metadata. He calls tagging a user-assigned system of categorization—user-created metadataUser-created, community-standardized demographic metadata (just rolls off the tongue, don’t it?) is like tagging with a personal twist.  Since we are all experts about ourselves (and sometimes experts on other stuff as well) we can easily apply an additional system of categorization that makes even more information available to others who might share our interests.This is a system whereby we attach personal data as tags. There might be a number of relevant tags that, with some small degree of standardization, would provide interesting tools not only for analysis and research, but for identifying, say, books we might like to read. This method will, in some form, be a large part of the semantic web, like it or not, so this here’s just a first step.  Since I like books, and LibraryThing is relatively new, it is a good place to start with it. Here’s what I suggest:

Those of you who know something about your personality type, use this as a tag on the books you really like. I will tag all of my books with MBTI-INTJ. INTJ is my personality type based on the Meyers-Briggs Type Indicator. I will tag all of my books with the year I was born, b1964, so that people will know that the person who said he liked that book (based on the number of stars I assigned it) was born in 1964.  I will also use the tag r2006 for books I read in 2006. That way I can tag a book that I read when I was a teenager with that date, say r1978, and people will know that it was a book that somebody with my basic personality liked at that age. Finally, I will tag my books with the geographical location where I read the book (on the off-hand chance that southerners, for example, have different reading interests than those from, say, Massachusetts. I don’t know my GPS location, which I imagine this will have to be translated into eventually, but my country and zipcode should be good enough for now.

So, a minimal tagging taxonomy would look like this:

• MBTI-XXXX (Personality-type as defined by the MBTI)
• bXXXX (Birthdate of reader)
• rXXXX (Date at which book was read)
• US01854 (Geographic position of person who read the book at the time they read it—Country abbreviation followed by zipcode.)

If you have any interest at all in this, just look at my library in LibraryThing and look at the tags I have set up there for my books.

One of the nice things about such a system is that when you are looking at the tags of your own library, or of another individual’s library, these do not affect the appearance of a “tag cloud” because they should be the same on every book in either. However, they will affect the appearance of tag clouds for certain books, because certain personality types might like them better, or people of a certain age might.

I’m sure that some people might object to this sort of personal disclosure, and of course it is entirely voluntary, so go ahead, object.  But the thing to remember here is that companies are making a lot of money already from knowing all of this stuff about you. I think that one of the important advantages of the internet, blogging, and Web 2.0 is that they allow people to take some small measure of control over information that has heretofore been owned by large conglomerates. The more we little people know about the world and each other, the more open the sources of our data become, the more power we will have. Not fomenting revolution, here, and I am trying to be realistic.

Oh, and the LibraryThing FAQ answers the question: Is LibraryThing a dating service?  Well, with people tagging their books with their personality type, it will be easy to use LibraryThing as a dating service.  So, I’m sorry about that. . .

Anyway, I’m interested in hearing peoples’ thoughts on this.  Is  anybody already attempting this sort of cataloging, or attempting to start some naming conventions for tagging in some other site or medium?   Please leave a comment so that I can at least clarify the muddy points of the above.