Metadata and the National Book Awards

Noted in NPR's Thursday, November 21 report on the National Book Award winners was that "a visibly shocked (James) McBride accepted the fiction prize. Considered the clear underdog, he said he wouldn't have minded if any of the other finalists won because they 'are all fine writers.'"

McBride's novel The Good Lord Bird is about a young slave (delightfully named 'Little Onion') who joins the abolitionist John Brown in his anti-slavery mission.

A search on Amazon for "James McBride" features The Good Lord Bird at the top of the list. The next three entries reference the same author, but by number 5 we're offered "Miracle at St. Anna (Movie Tie-in) by MacBride, James" (different spelling of McBride). Soon after Rachel Kushner's spring 2013 novel The Flamethrowers, is featured for sale, as far as I can tell solely because The Good Lord Bird is the seventh title often purchased by people who buy The Flamethrowers.

And what do we call this? Discoverability.

Barnes & Noble fares no better in its treatment of McBride. The second title on its list is called James McBride (Writer). A quick glance signals that it's a study of McBride and his career. Closer examination reveals that it's an 88-page $45 paperback that reprints old Wikipedia entries on the author. Number 4 on Barnes & Noble is "Miracle at St. Anna (Movie Tie-in) by McBride, James." Number 6 is a history of law since the 17th century unfortunately named simply John & Abba Townsend vs. James and Hannah McBride.

And what do we call this? Metadata making books findable on ecommerce web sites.

Kobo gets off to a good start: the first four titles are by our author. Result number 5 is a free "reprint" of an 1826 study called Symmes's Theory Of Concentric Spheres: Demonstrating That The Earth is Hollow, Habitable Within, and Widely Open About the Poles, written by John Cleves Symmes and James McBride (there's no detail on Kobo. Google supplied the publication date and a description of this obscure tome).

I don't even want to begin to describe the abomination of trying to track James McBride on Apple. Because he's active in music and movies as well as books iTunes collapses into confusion.

There's a remarkable (and delightful) difference when I do a search for "James McBride" (in quotation marks) on Google. The first three pages of results all refer to this prize-winning author. And you also learn that the Miracle at St. Anna is not written by a different author with the similar name of James MacBride but in fact by our award-winning McBride. Someone entered the incorrect spelling on Amazon, leading me to assume the book was by a different author.

I think this short saga illustrates two major issues.

The first one is a tough sell, but it's by far the more important: If your name is James McBride or John McNeil or Joe Stewart you'd better change it if you want to maximize your online sales. The average ecommerce search engine does not respond well to a name that's used by multiple authors.

Now of course Mr. McBride is far too advanced in his career to even think of marketing a pseudonym. But for authors starting out, choose a pseudonym (if your parents didn't do you the dubious favor of naming you Thaddeus McIlroy) and then test it against Amazon, Apple, and other search engines. It's tough enough to make you book stand out without the all-too-common author name confusion. (Amusingly I just searched Amazon for Thaddeus McIlroy and I don't appear: it has my books listed only under my abbreviated name, Thad McIlroy.)

The second issue is that the companies selling books online have not (for the most part) married their search engines to accurate and useful metadata. Only part of the blame falls onto these online vendors: in the past publishers rarely worried about supplying "verbose" metadata. They didn't have to. In the pre-online era an ISBN linked to a database like Bowker's was often enough to get a title on sale in a bookstore. Publishers didn't think of metadata as descriptive but merely the minimal data required for a listing. Only now are they beginning to understand the value of assigning robust digital data to their latest and largest titles.

The legacy of bad metadata encompasses literally millions of books. Services like Bowker's can greatly improve metadata records. But of course there's a cost for these services and many publishers and vendors are uncertain if the cost can be justified.

What to do? This is truly an industry-wide (and international) problem that touches authors, publishers and online sellers. Authors don't have access to the data supply chain so we can't rely on them. That leaves publishers and online sellers. Fixing the metadata just for James McBride might take an hour's effort. Multiply that by millions and we're going to be working on this problem for years to come.

Thad McIlroy is an electronic publishing analyst and author whose near-obsessive focus on metadata led him to co-author The Metadata Handbook (with Renée Register).