The eBook Test: Smart Digital Books Metadata Notes #5

Via France: Beyond Google ... The ways of collective intelligence

This is an interview with Pierre Lévy, of the University of Ottawa.

I'm quoting the original French via Google Translate, without tweaking, so the syntax is rough but the ideas are still understandable.

It must be understood that the major funders of the W3C. The main objective of return on their investment and return the maximum value for their shareholders, not to develop the collective intelligence of mankind. If their goals through an increase in collective intelligence, so much the better, they are not against a priori, of course, if not, too bad! The result is a focus on this very focused on the so-called "click stream marketing, and therefore the revenue per click.

Emphasis added by me.

Understand that. A world ruled by Google and its ethos is one ruled by a marketing scheme: revenue per click. Anything else is secondary. And how to increase revenue per click? Offer as many damned clicks as possible. Hence Google Book Search: trillions of other possible clicks, from which Google makes it daily tens of millions of fractions of a cent that add up to a collection of billions of dollars at the end of the year.

Finally, new modes of computation and social organization of collective intelligence emerge from the spontaneous activity of Internet users.

One of the first of these was Dave Winer's RSS. More will sprout up as an increasing number of people wake up to the fact that having any one point of the Net in centralized hands is detrimental to the overall good. (My own use of Twitter as a source of research through Favorites has revealed to me that I cannot leave such data solely in Twitter's hands and just this week started feeding Twitter into Google Reader [Google again!] and finding ways to locally store my tweets.)

I'm looking in the same vein, a digital encoding of meaning, whose effect could be an extraordinary increase in the power of expression and interpretation in the hands of users and their collective intelligence.

Emphasis added by me.

Yes. In my original post, the entire point of smart digital books containing metadata was to encode the meaning within the text.

And this:

We must not forget either that a huge amount of metadata are not organized by ontologies. I am thinking particularly of tags produced spontaneously by Internet users on their blogs, on applications like Flickr, Delicious or YouTube, or to "hashtags" used on Twitter.

I would expect, and made room for in my original post for, user-contributed book metadata too.

(Not quoted above, but of personal interest to me, seems to be the less-than-favorable reaction some of his prior work received in France. I'm very surprised by this. France, which gave the world Balzac, Baudelaire, Hugo, Nerval, and others, not seeing the point?)

Lévy has been working on something called Information Economy Meta Language. There's a website for that. Pierre Lévy is also on Twitter.

There is also OWL: Web Ontology Language, which seems to fit into RDF. With OWL and RDF working together, what support should there be for EAD, then?

Unfortunately, when it comes to smart digital book metadata, people tend to immediately think fiction. And while there is applicability to fiction (it is no less vital to fiction than to any other kind of book), the first thing to consider is non-fiction. I recently read Outliers by Malcolm Gladwell. It was a startling book to me personally because it turned out to be a missing puzzle piece that fit together so many other books I've read over the years -- years meaning decades. Outliers would be the perfect expression of the book metadata idea in that it would -- with a complete dataset (which it's not possible for me to create, not recalling the titles of every book over those decades) -- reveal how one book could relate to so many other books; not just ones in the past, but ones not yet written too, that seemingly have zero connection to it.

That connection between present and future books is also part of the concept expressed in this post, Defining the Big Shift:

From knowledge stocks to knowledge flows. We are moving from a world where the source of strategic advantage was in protecting and efficiently extracting value from a given set of knowledge stocks -- what we know at any point in time. As knowledge stocks depreciate in value at an accelerating pace, the focus of economic value creation shifts to effective and privileged participation in knowledge flows. Finding ways to connect with people and institutions possessing new knowledge becomes increasingly important. Since there are far more smart people outside any one organization than inside, gaining access to the most useful knowledge flows requires reaching beyond the four walls of any enterprise.

Emphasis added by me.

To take Outliers as an example, the smart digital book metadata model would have connecting metadata added to the book as other books in the future are published. I could, for example, be looking at a passage in Outliers, go to its metadata, and see two new books have been published that somehow connect to that very passage. If I'm interested, I could then immediately buy and read those books. This is also an example of Axioms 3 and 7. Note that I wouldn't have to rely on reviews (which would be likely to miss such a connection) or slow Word of Mouth. The connection would come to me.

Metadata used dramatically in the movie Fight Club: