Discussing the future of the ISBN as a key identifier for digital products
An exchange of views and ideas about perspectives on the ISBN as the key identifier in the publishing supply chain in the digital age and the possibility of additional options for handling variants and granular content took place between the International ISBN Agency, the MVB (Marketing- und Verlagsservice des Buchhandels, Germany), and Klopotek on 5 March 2014.
- Stella Griffiths, Executive Director, International ISBN Agency
- Michael Castner, Head of Product Management, Klopotek
- Wolf-Michael Mehl, Managing Director, Klopotek
- Ronald Schild, Managing Director, Marketing- und Verlagsservice des Buchhandels (MVB)
- Steve Waldron, Vice President of Business Development, Klopotek North America
Ronald: I’d like to thank everybody for attending. The purpose of this meeting is to exchange views and concepts about the role of the ISBN in the digital age, especially with regards to identifying granular content, such as individual chapters, or variants, such as books and e-books in different formats. Klopotek told us that some of their clients have issues with using ISBNs for this purpose and, in an article, said that software systems should not be what they called “ISBN-centric”. Stella, could you perhaps first tell us about your opinion about this and your thoughts regarding the future of the ISBN?
Stella: We certainly don’t dispute the idea of having a system that is identifier-agnostic. In fact, we very much support that because what we have found is that sometimes the ISBN has been too successful for itself. It’s been around for almost 50 years and it was a very quick success within the book supply chain. I think publishers clutched on to it because it really worked as an identifier even when having this type of international identifier was a new idea.
On the other hand, publishers also used the ISBN for purposes for which it was not originally intended. They saw its potential as a fantastic identifier and thought it could be used to identify not only books but also book-related merchandising. We know that many publishers’ systems are run entirely by using ISBN. We agree with Klopotek that people should use identifiers in the most appropriate way, and thus ISBN is not always the right option. Essentially, I think it comes down to this differentiation: What do you want the public to know, and what do you as an organization want to know privately and keep private? For identifying data about a contract or its rates you should not use an ISBN. Instead, it would very much be in the interest of all publishers to have their own proprietary identifier to drive their systems. Then, when they want to exchange information publicly about a particular edition or format of a book, they should use an ISBN.
Steve: I agree. One issue with making information available publicly is, however, that many of our customers have to use different ISBNs for content-identical products. For example, e-books in the EPUB or the Mobipocket format: these are just variants of the same product. Some people in our customer base are frustrated about that.
Stella: If the formats are different then they really are separate products, but is it the case that publishers are worried about using and having to pay for so many ISBNs to identify all of their different products?
Steve: Yes, this is certainly a problem for many independent and small publishers but also in the academic sector. We, of course, accept that the ISBN is the only game in town for identifying products, but – obviously – there is now a much greater variety in products than some years ago.
Michael: Perhaps I can give you an example. Two or three weeks ago I had a workshop with a large international academic publisher with a huge number of products about doing chapter-based business. They want to sell their chapters individually, and their average publication has 20 chapters. To sell these chapters individually in digital formats for 1,000 books, the number we are talking about is 20,000. In addition to this, they want to make their chapters available in the PDF and the EPUB format, so the number rises to 40,000. This is a huge amount if every chapter needs an individual identifier. They are afraid of being forced to spend a lot of money if an ISBN has to be used for every chapter.
Stella: That’s a very interesting example. There is one question I’d like to ask about what they’re planning to do. Is it really the case that they want to make each of these chapters available uniquely, or is their plan something like this: People might want to construct their own individual sets? If people create their own e-books or sets of content, the ISBN is probably not the best way to identify these customized products. But I would assume that these fragments, which users can construct, will in most cases consist of more than one chapter. So I think it would be helpful to know more details about what this publisher is planning.
Michael: Well, they say that they want to sell chapters individually. Of course, they also want to offer the possibility to create customized products. I think they have no worries about that. What they see as a problem is that the amount of sales items for which an ISBN as a unique identifier is required is increasing to an extent that it becomes too burdensome for doing chapter-based business.
Stella: ISBN can be applied to any level of qualifying product that is available separately. So, ISBN can be assigned to an individual chapter or even to a definition within a dictionary if the publisher wants to make these available separately; while these products are much smaller than what we might think of as a “traditional” book there is no minimum requirement length and so they will qualify for ISBN just the same. However, I wonder if they want to make these individual pieces of content primarily available through their own website rather than in the supply chain? This is another important question. ISBN is of great value if you are making these individual pieces of content available through more than one channel. If the intention is to sell from one single source there might be other options than ISBNs.
Wolf-Michael: I think it is not absolutely clear to them what they really want to do. They are exploring their options. They have not made up their mind which platforms they want to use or if their content should be sold exclusively through one source.
Steve: There is a large US-based educational publisher who can serve as another example of one of our customers who want to split up their content into chunks and chapters which can be sold individually. They want to use multiple platforms and partners for this. They are worried that they’ll have to use a huge amount of ISBNs to get this done.
Ronald: Stella, at this point of our talk it might be a good idea to summarize our views and what has been discussed to far. Would you do this for us?
Stella: Sure. Essentially I think that the ISBN as an identifier is pretty much format-agnostic. It doesn’t matter if you have a print or a digital publication which you want to identify. The important question to ask is: do you want to make your products available publicly in the supply chain? An ISBN is probably the most trusted identifier for this. If you want to sell your products through multiple channels in order to maximize your sales opportunities and chance for discovery, then an ISBN should be used.
On the other hand, a concern expressed today is that if you choose to produce more granular products – potentially in many different formats – then the challenge to keep track of all the related metadata is potentially far too great to use individual ISBNs – or, indeed, any other identifier. The other worry that has been voiced is that it can also be expensive.
Ronald: I had the impression that the worries that were mentioned had to do with costs. Nobody said that ISBNs are not suitable for identifying digital content or smaller pieces of content.
Michael: This is correct.
Steve: Many of our customers – an American professional association who use our software is another example – are selling their content below the traditional individual product level. Still, when you get into the main channels and want to sell it, the use of identifiers is getting prolific. I think the ISBN would hugely benefit if it were made more adaptable to the digital world in a way that publishers don’t have to use individual ISBNs for every format or chunk of content. This is why some of our customers are looking for alternatives.
Ronald: But what kind of identifier is being used instead of the ISBN?
Steve: Some use DOIs [Digital Object Identifiers]. Others have been experimenting with UPC [Universal Product Code] and GTIN [Global Trade Item Number]. But, obviously, ISBN is the most powerful identifier that exists and I also think it is likely to stay that way if ISBN becomes more flexible. But just because ISBN is the only game in town, will it always be the only game in town? This is the kind of question we get from our customers.
Ronald: Steve, I think you’re right, but the GTIN and the ISBN are not so far away from each other. The question is always: What do you want to achieve? Are you looking for a unique identification possibility? Then the ISBN is the right choice for you. To my knowledge, the DOI should be used for a different purpose. The DOI helps you to track content and follow the re-used content on its way through all the different kinds of media. So you cannot really compare them, as they have different purposes.
I think that Klopotek believes that there is great value to the ISBN, but their customers have so many workflows which are new and related to digital content and granular content, and these workflows require additional identifiers. The ISBN cannot handle all of this; it puts too much pressure on this identifier.
Wolf-Michael: I fully agree with that. A publisher of scientific journals, for instance, cannot use an ISBN for identifying each individual article of an issue. So, for me, it is obvious that you need different identifiers.
On the other hand, just using a DOI instead of an ISBN would not help but would make things even more complicated. DOI is not just an identifier; it is associated with a URL as well, so managing it in a system is more complex. Putting the costs issue aside for a moment, I think that there is no reason why ISBNs should be replaced for identifying and why ISBNs couldn’t be extended to identify individual chapters of books.
Michael: What we have in mind would be this scenario: To be able to use an ISBN but to have additional information to determine what this ISBN is really used for. Currently, the problem is that you need an ISBN for each e-book format of each title. And it has been decided not to use the same ISBN with a format identifier. So, for the moment, everything works, as there are usually only two or three formats. But we believe that when it comes to selling chunks of content, there’ll be so many individual elements a publisher has to handle, so assigning ISBNs to all of these chunks will not work.
To sum this up: if we’re talking about a unique identifier as such, the ISBN is fine. But for handling content at a granular level it would be easier to have an ISBN plus some additional information which determines what you really want to deliver at the end of the process, such as: this is a chapter of this book.
Stella: For each individual product that you make available, you will need some kind of identifier. If you don’t use ISBN as a unique identifier, you – or our downstream provider – will still need something else. I think we’re in a situation where we actually have to recognise the need for different types of identifiers, each doing a different job – for example: identifying a work, perhaps identifying a release, identifying the specific product) and not less of them. And I want to add that ISBN can be enhanced, for instance it is possible to turn an ISBN into an ISBN-A, which, like a hybrid, provides the best features of ISBN and enhances it with the resolvability of the DOI system.
Ronald: I think that we shouldn’t underestimate the value of a unique identifier. The supply chain for producing books is very different in different countries. But all the different supply chain management systems have in common that they accept the ISBN as an identifier.
Do we all agree that ISBN is the best identifier for books, while other identifiers might be the better choice if different types of products or media are produced?
Steve: Yes, it’s got a fairly good point to it. But I think the problem we’re experiencing is that many products, even traditional book products, are starting to incorporate multimedia variants. So it is getting increasingly difficult to differentiate between ‘books’ and ‘other media’.
Ronald: I don’t think that this contradicts what I’ve just said. If you are a trade publisher and publish a book which incorporates a video, you might want to choose to use a different identifier to identify the video as such. If it is an audio file, it might be another identifier which is the best choice for this type of file. But the main product – if it is a book – can and should be identified by using an ISBN. Even a chapter of this book, which is to be sold with a video and an audio file, could have an ISBN. Again, we really shouldn’t underestimate the value of a standard identifier.
Steve: Yes, I think we all agree with that.
Stella: Talking about this, an important question which comes up again and again is: what is a book? I have to say that we don’t identify a book in the international standard, and there’s a good reason for that: we don’t want to define what a book is. Tomorrow someone might invent a marvellous new product which we can’t envision today, but when we see it we will recognize it as being a book.
There is an alternative way to find out whether something qualifies or not for getting an ISBN, which is to ask three simple questions: Is the product available to the public? Is it something finished, in other words: not a serial like a newspaper or a journal? And is it text-based? It can, of course, incorporate other elements. If the answer is yes to all these questions, then the product can be identified in the supply chain by an ISBN. To some degree, ISBN has quite few and low barriers about whether or not something qualifies to be identified by it. This is also a reason why ISBN has become so successful, in my opinion.
The Klopotek system is not based on the ISBN as a driver, so you don’t have to start with an ISBN to create a record but you can start with a private number. So if you choose to sell something you can decide, what is the most appropriate identifier for this product? In many cases it will be an ISBN but in other cases it might be an ISMN [International Standard Music Number] or whatever it might be. There are opportunities for publishers to have a lot more flexibility if they base their system initially on a proprietary number.
I’d like to stress again that the main difference when it comes to identifiers is the need for something public versus the need for something private. For identifying objects in the public supply chain, ISBN helps publishers to maximize their opportunities because it enables multi-channel sales. This is something which proprietary identifiers cannot do in the public world.
Michael: I agree.
Steve: I think I will approach some of our customers, larger ones and smaller ones, and ask them how they think identifiers should be used and write an article about that. I think it would be important to get them involved in this discussion.
Ronald: This sounds excellent. I look forward to reading this and I’d like to thank you all for participating in this talk.