I talked yesterday with David Miller, my literary agent and friend. David is the first reader of my drafts (my wife is the first listener) because he has a wonderful head for structure and flow, he understands how books work as an experience for readers as well as how ideas develop in them, he cares about the quality of the book first and foremost, and he’s frank.

David has read drafts of the first two chapters and does not think I have to throw them out entirely. They need more narrative connection and more examples (Examples, my old nemesis. We meet again). And, David suggests, they need a prologue or preface that will set up the book’s problem. David says it needs something like the distinction between the kitchen’s silverware drawer and its miscellaneous drawer in Everything Is Miscellaneous, which gave the reader an easy way of formulating the issue in her mind.

The challenge, of course, is figuring out what that central issue is. I’ve thrashed among several alternatives, each of which would provide a different way of structuring the book. If the book is about the problem that there’s too much to know (which is, after all, the title), then the preface should illustrate that point. If it’s about the growth of networked expertise, then I should present a contrast between the old and new ways of being an expert. If it’s about the restructuring of knowledge, then the preface should give an example of traditional knowledge and new knowledge. If it’s about the socializing of knowledge, then … etc., etc.

I have three times in the past month given a 10-15 minute talk that has gone over pretty well, and that could serve as a setup. It starts with the data-information-knowledge-wisdom pyramid, gives its history, and critiques it on a couple of grounds. It ends by saying that the real problem with it is that it continues our traditional strategy for knowing a world that is too big to know: reducing the field of what’s to be known. But, in a networked world, not only do we have tools available that handle far greater capacity, all that’s available to be known is visible. (I’ve written about this here and here.) The old reductive strategy just isn’t enough any more. I like this way of setting it up, but I agree with David Miller that it still needs a preface, if only because the DIKW pyramid is not as well known as I think it is.

So, I’ve been thinking that I’ll begin it with the contrast between two cases: Darwin unearthing a stubborn fact about barnacles, and the uses Hunch.com wrings out of an endless series of answers to trivial questions about your preferences and experiences. (I had the chance to interview Caterina Fake for an hour last week.) The contrast is between the small world of hard-won facts, and the endless world of tiny, easy facts — nonce facts? — jotted down at 12 per minute. I want the reader to see not only that there are many more facts now, but that the basic role and structuring of facts (and knowledge) have changed.

I’m not convinced Darwin vs. Hunch is the right contrast to draw. My book is not about promoting the particular way of knowing that Hunch.com uses. The book is not about algorithmic combing of huge databases. That’s just one technique for dealing with a knowledge overflow. I’m more interested in the fact that we generate the knowledge overflow, and change our ideas about what a fact is, because we’re now able to do something with facts much smaller than they used to be. So, I have to try to keep the reader from thinking that this is a book about learning without models or about the virtues of massive databases. But, I won’t know how it goes until I write it.

(Why Darwin on barnacles? Mainly because I want to begin with a snarky remark about Thoreau from a comment he made in 1852, which is when Darwin was working on barnacles, busy postponing publishing his evolutionary theory.)