Joe Wikert complains about the general lack of indexing in e-books, pointing out that text search isn’t really the same thing in his Average Joe post. Now a month later, he comes up with some concrete suggestions.

Now, what he suggests is obviously an excellent idea. But, as we live in a world dominated by the marketplace, I wonder if anyone will ever believe that they can make money by providing the sort of hyperlinked index Mr Wikert proposes. A lousy index in a print book, or even no index at all, has probably never led to the loss of a sale — criticism of author and publisher to be sure, but if they want/need the information in the book most readers are just going to have to put up with it.

Mr Wikert is suggesting that we could have an automated sort of index generator built into our reading device, or software. I suspect the problem with this would be not the filling of the bath, but the regulating of the flow so that you didn’t get inundated by ever more recondite and (for your immediate needs) irrelevant information. Set a computer to parsing text, and it’ll parse till the cows come home.

I think the problem with indexing and e-books may just boil down to the nature of the beast. Just as a print book is better at propping up a table or keeping a door from closing, so it may be more suited to index-use. We use an index in a couple of ways. You are reading along, and you think “This guy mentioned James VI’s sleeping arrangements before. What was it he said?” So you go to the index and look up “James VI, favorites”; and Bob’s your uncle. Or you are researching a topic, and you look at “Parliament, financial control”. (Notoriously of course scholars use the index to check whether they are mentioned.) With a Wikert hyperlinked index you risk spending the rest of your life tracking down references if once you press the magic button.

As he points out, one of the strengths a human indexer brings to the table is an ability to index synonyms: thus not just “sleeping arrangements” but bed, bedroom, bed chamber, night companion, sheets, four-poster etc. etc. might all end up indexed under “sleeping arrangements”. But God save us from the retrieval of every such sort of cognate from the depths of the web. The main virtue of the human indexer is the ability to distinguish what is relevant from what is not, and to arrange the material into a logically coherent system which enables the reader better to access the information in the book. Computers are as eager to search for “the” as they are for “bedfellow” — be careful what you ask for.

With an e-book you are always a little uncertain just where you are. You don’t have that intuitive notion “I’m about a third of the way through” though of course there will no doubt be a little note telling you you are in fact at location 2018 out of 7981. Precision here is the enemy of intuition. I may often recall that the earlier discussion of one of James’s boyfriends was on a recto, in the top third of the page, and it was a while ago that I read it so I can conclude it must be back quite near the beginning. And I can often get there without bothering with the index — some of the things I might want to look back at wouldn’t be indexed anyway, e.g.misspellings. Thus even if it’s not in the index, we have ways of making the book talk. Such a facility is not offered by an e-book. Everyone says the distractions offered by the web are poisoning our reading capacity (something I don’t agree with) and trolling through an e-book in an attempt to find an earlier reference is a massive distraction. I even find clicking on a note call-out distracting: off you flash to a different, though superficially identical location. Maybe I’m just not used to the switch and would get better at it if I tried harder, but I suspect it may actually have something to do with the nature of the beast.

None of this should be surprising: e-books are obviously different from p-books. The fact that much of what’s available now is derived from a preexisting p-book is probably confusing us as to what’s possible and what’s not. Maybe we haven’t been at it long enough to have recognized that indexing may just be a p-book tool. For this reason, and the display of tables, mathematics, foreign languages especially in different scripts, as well as the nature of the use to which they are put, academic monographs may just work better in print.