On ISBNs

Back in my Bowker days (1986-1996), when we first realized we were going to have to expand the ISBN to more than 10-digits, there were numerous debates about whether the ISBN should still retain meaning:  the country/language code and the publisher prefix.   Professionals in the field argue this both ways.   When we were considering establishing a system for digital content numbering, the web consultants we worked with argued against building any meaning into IDs.   I always liked the idea and wanted to maintain it.

In the media industry, there was a standard using the UPC system that assigned a media digit to the number (the number before the check digit.)  That's why most CDs have a "2" before the check digit.   Using this schema, you could tie all of the different media versions of an intellectual work together with the same number, except for the media digit and the check digit.   Unfortunately, not all the media publishers adopted the scheme and even among those that did, it wouldn't have handled the case of a letterbox version of a DVD and a full-screen version.   But in theory, I always thought it was a great idea:  imagine if the ISBN were the same for a mass-market, trade-paper, hardbound and even an audio version, except for the media identifier and the check digit (and ignoring that frequently paperbacks are published by different publishers than the hardbound and therefore, it would have a completely different number anyway.)  

I had always thought that the DCI should have been an extension of the ISBN.   The ISBN would have represented the parent work and the various derivations, elements and digital versions of the work would have extensions of the original number.

One of the things that's troubled me about the implementation of the ISBN-13 is that until you start using the 979s, you don't gain any numbers.   And once you do use the 979, if you use the same publisher prefixes for the 979 as for the 978, the core of the number (except for the check digit) could be the same for two different books from the same publisher.  I would find that confusing, but I suppose the standards committees did not.  To my way of thinking, making the ISBN-13 totally consistent with the EAN made it less useful than it could have been, although "Bookland' has always put a smile on my face.

The Whitakers, who along with Emory Koltay in the U.S. were responsible for the development and implementation of the ISBN, were always opposed to using the ISBN for anything but printed books.   I always disagreed with that and the fact is that there was no way to control this anyway because aside from what most people think, the ISBN Agencies do not really assign numbers - they assign a prefix and rules - the publishers assign the numbers.   My feeling back then was that anything that was circulated in a library (except perhaps for periodicals and sheet music, which had their own numbering systems) or sold by a bookstore should have had an ISBN.   So when you state, "ISBN's for everything," I completely agree.

I find it baffling that publishers don't want to assign ISBNs to digital content since either the rights have to be tracked or the sales have to be tracked and any system they're using is going to require a unique identifier anyway, even if it's automatically generated.   As long as you're using a unique identifier, it might as well be the ISBN.

Michael Cairns has a great bit on Persona Non Data about the ISBN today as well.

Michael Cairns has a great bit on about the ISBN today as well.

Michael Cairns has a great bit on about the ISBN today as well.

Michael Cairns has a great bit on about the ISBN today as well.

Michael Cairns has a great bit on about the ISBN today as well.