As the digital landscape continues to evolve, new platforms with new capabilities debut, and new business models for distributing digital products emerge. What doesn’t change is that when you have a product you need to identify it, whether that product is a physical book, an e-book for sale or rent or loan, an e-book in a variety of digital formats, a part of an e-book being sold by itself, or content made available on a Website. To decide how and when to identify digital products, and what kind of identifier to use, publishers have been relying on the Best Practices for Identifying Digital Product Policy Statement created by the Book Industry Study Group’s Identification Committee. That committee—which is made up of extraordinarily engaged and dedicated representatives from all segments of our industry—recently updated its recommendations. The latest version of Best Practices for Identifying Digital Product recognizes the need to broaden the original focus on trade publishing and ISBN assignment. The full Best Practices document is available via bisg.org/docs/BISG_Policy_1101.pdf. What follows explains the highlights of the latest version.
Three changes strike me as most important. 1. Instead of focusing on assigning ISBNs to digital products, the Best Practices policy statement now focuses on various means of product identification, including the use of proprietary identifiers, such as GTIN-13s, and the assignment of identifier by third parties. In-depth discussions deal with defining proprietary identifiers and knowing when they are appropriate for use, and when they are not (for instance, a proprietary identifier should be used only internally and not used as a publicly communicated identifier). 2. Recommendations are now organized around business cases in three distinct sections—Content, Format, and Usage Constraints. 3. The new version provides clearer explanations of usage constraints and of alternative business models such as purchase vs. rent and purchase vs. lend. None of this means that the importance of the ISBN has diminished. On the contrary, we continue to stress that publishers should assign ISBNs to the products they release into the supply chain. But the new emphasis on identification in various ways responds to the fact that a digital product (unlike a physical book) may change in format and/or in terms of usage constraints several times on its way through the supply chain to the consumer.
The Way It Works
At each step of this journey, it is essential for the product to be identified properly so that information about it (the title metadata anchored to the identifier) will be accurately communicated down the supply chain, and so that, after the transaction, sales or usage information will be accurately communicated back up the chain to the entity that provided the identifier. For example, if I assign an ISBN to an EPUB and send it off to Amazon, and Amazon creates a Kindle format for sale through its Kindle Store, Amazon won’t need to assign an ISBN to the Kindle version, but it will assign its own proprietary ASIN identifier to the Kindle Book, and that ASIN will be linked to my ISBN, so I can provide Amazon with updates to the product metadata if needed, and Amazon can tell me how many copies it has sold—and pay me for those copies. Think of it like international travel. I have a U.S. passport associated with my own personal identifier, my Social Security number. When I travel to another country, I am not issued another passport or another personal identifier, but each country has its own way of identifying me and linking that identification to my passport. The identifiers each country uses can all be different, but since they are all linked to my passport and U.S. identity, there is a traceable chain of my travel activity.
“Constraints” Makes It Clearer
Changing a single word in the new edition of Best Practices for Identifying Digital Product helped clarify an issue the BISG Identification Committee had grappled with since work on the Policy Statement began: What exactly is a digital product? In the original Policy Statement, we used the term usage rights. In the Revision, we changed that to usage constraints. Initially, we made the change to match the language used in the ONIX standard for metadata exchange, but a much deeper significance emerged. With a physical product you have an actual thing, something you can hold and observe. You have a package of content, and the package pretty much dictates how the product is to be consumed. It is this package of content, created by the publisher, that is identified by an ISBN, and it is this package that is sold to the consumer, who understands the experiential difference between a hardcover and paperback, and whether a child’s book is suitable for use in the bathtub. With digital products, matters are not so straightforward. Content and package are sold separately and joined only after the transaction, when the digital file is loaded onto the rendering device. Moreover, digital content is licensed in a transaction separate from the sale of the package (the tablet, the e-reader, the computer), unlike the license for the content in physical books and the sale of the package containing their content, which are transacted simultaneously. In the digital world, the exact same content can become any number of distinct digital products, but publishers—well, let’s say most publishers—don’t create the package, the device, that will be used to consume the content. So the content must communicate how it can be consumed, and that is done through the metadata that describes the product, and through the usage constraints, or lack of usage constraints, described in the metadata. Those usage constraints are then applied as the content is sold to the consumer. In one market, the usage constraints may allow printing part or all of the content in an e-book. In another market, printing that content may not be allowed. A retailer might offer consumers the exact same content either with a perpetual license or for rent for a limited period. Although the content remains exactly the same in all these cases, the different usage constraints mean that there are four distinct products. Each one needs its own distinct identifier to differentiate it from others. The identifier could be an ISBN assigned by the publisher, or a proprietary identifier assigned by the retailer, or a combination of both. Can we say then that the digital product is defined by its usage constraints? I think we can, but, as is so often the case with questions tackled by BISG, this issue will probably be revisited right through the next update of Best Practices for Identifying Digital Product. I hope you take the opportunity to download and use the latest version at bisg.org/docs/BISG_Policy_1101.pdf. And if you have questions, please send them along. Everyone is welcome to participate in shaping versions to come.