This Blog has moved!

This Blog now lives at Comments on this page are disabled and it is only kept because of my general aversion to broken links. Please update your feed readers to

Friday, January 19, 2007

Heard on TWiT : David Pogue on indexing.

I'm sincerely sorry for database fans hoping for a post on query optimization, but this post will talk about book indexes. You know, books. Made of paper. And ink. With page numbers, a table of contents, and a really old-fashioned user interface which doesn't even offer full-text search.

Books, then. I heard mentioned on TWiT that David Pogue writes his books' indexes (or is that word "indices" ? I never know.) himself. All the hosts seemed to agree this was a waste of time and, well, basically, nuts.

Apparently when you've just finished a technical book and it's time to write an index, the standard way to go is : don't do it. The publisher will handle all of that by himself, either by hiring a "professional indexer" (now that sounds like a rewarding job) or by running the book through an indexing script that automatically generates the index. Some super-scrupulous publishers who think of themselves as catering to the high-tech book connoisseur crowd will, maybe, use a combination of the two.

Please allow me to be the first to say: "Huh ?".

When I'm shopping for a technical book, I have to figure out in a few minutes whether it will help me answer my questions on the subject in a timely manner. And in such a short time, I can't do much more than skim a few random pages, then ask myself a couple of likely questions and try to answer them using the table of contents and, of course, the index. If it's good -- and assuming I had some sleep in the last twenty hours or so -- I'll quickly find an obvious word related to my question, and this will point to a page where I'll find the answer. If it's not good (and I've found most tech book indexes to be pretty bad) I won't.

Now according to an old TWiT I just listened to, David Pogue doesn't do this the traditional way. (Which I should really call "the lazy way".) He believes he can do a better job himself, seeing how he wrote the whole book, and this gives him a better grasp of the content and the kind of question a typical reader might look for in the index.

Now that is a radical idea...

1 comment:

pogueNYT said...

Man, you are SO RIGHT. Good thing I never heard them say I'm nuts to do my own indexes, because I would have blown my stack.

There is NOBODY who can do as good a job on the index as the author, who intimately knows both his subject and his audience.