Can our libraries be digital if the books are not?

Maybe a new genre will be necessary for e-books to become popular successes — you know, something along the lines of a multi-pronged narrative where the order of what you read and what the characters know depends on the reader making choices: "(A) Bill lets the stranger walk by. (B) Bill engages the stranger in conversation." Such books already exist — my third-grader has several Goosebumps titles that R.L. Stine has concocted that use this structure — and it is because these are all text, all narrative, that electronifying them will preserve the book essence.

But, really, if this were done right, it wouldn't be a book at all but something more akin to a CD-ROM adventure or a video game. And I can imagine applications that would take your input — say your financial circumstances and retirement goals — and then offer text-heavy guidance as to your best investing strategies. But the more you leant on the application and interaction, probably the more interesting and useful this could be, and it would probably be better off if you didn't approach it as a book at all.

Now a library isn't restricted to serving books to patrons — obviously periodicals, music and videos can be checked out in most libraries, and the rationale for offering online access seems akin to that of offering reference librarians: You need some information? Still, any Hill acolyte will say that a book — reading — has an almost unreal capacity to engage the reader (magic Hill says) and let's not ignore the formula that has been devised over these centuries as we seek to expand the delivery vehicle.

So if I focus my argument on making books for the digital library instead of discussing the broader topic of ways of delivering information, it's because I agree that books play a special role for us. Let's not re-invent everything just because we can.

But there's so much that could be done that isn't being considered, even within this narrow slice.

Take the matter of the subject index.

Today in a city the size of, well, my hometown, Montclair, NJ, which has about 35,000 residents, there are many sizeable libraries — the town library on South Fullerton Street, the university library/libraries at Montclair State, the high school and three middle school and six grade school libraries. A fair-sized library — maybe a couple thousand books — at Union Congregational Church and surely similar ones at other churches and synagogues. The library at the art museum. And there must be some individuals — I know some candidate professors — with specialized collections as large as Union Cong's.

If you were able to take the index out of any of these print books and merge it into a site-specific subject catalog, what a detailed and powerful search tool that would be. The patrons at any one of these libraries would find that incredibly useful.

Of course, the resulting subject card catalog would be enormous and making physical cards for all these index entries would be taxing, not to mention sorting them, so let's skip ahead in this gedankenexperiment and make our unified-index subject catalog from electronic files. Of course the books themselves don't need to be electronic. The subject index is simply going to point to a location in a resource — and I use that word advisedly — and that might be a print book or even a periodical, if it's been indexed. And of course Topic Maps and RDF have paved the way for this sort of thing to be done rather handily, if only the indexes were available electronically.

Publishers, naturally enough, do have all their indexes in electronic form, all their content for that matter (I exaggerate I suppose, but no publishers can typeset their texts by non-electronic means without increasing their composition costs ten- or twenty-fold). No one has asked for the indexes to be released separately from the print book and in electronic form.

Yet imagine it — the full contents of a library could easily be put into a unified index (well, apart from fiction and other non-indexed titles). And Montclair's index, specific to its titles, would be individual and different from those in Verona or West Orange or Bloomfield just a couple miles away.

Perhaps there aren't enough libraries to persuade publishers to go through the extra step of making indexes available in RDF or XTM format. What about individual book collections? In my own case, and this is on the small side for someone in publishing, our home contains several hundred books, half of them children's books. I'm an e-book enthusiast, with well over 2,000 individual titles (and more than 10,000 with all my duplicate formats) but if I had just the indexes for the Egyptology or XSLT or art or Renaissance or chess books, I would be ecstatic. Even though I have fewer than a dozen in all of these subjects, these little mini-collections would be vastly more useful if I could do one lookup and then go only to the book or books that had material to answer my question.

Almost from day one the key issue in electronic publishing has been publishers' concern about lost sales because of "piracy" or digital copying. Yet if libraries — not just digital libraries — would push on something like releasing indexes to print books in electronic form, where no "piracy" or digital copying could occur, think how much farther down the road to our future we would be.