By Noz Urbina, special to The Content Wrangler
I just finished two lengthy interviews with Bob Glushko, adjunct professor at the University of California Berkeley School of Information. Bob is the author of multiple books, a pioneer in the world of structured information in business, and (my favorite), an honorary member of the Cognitive Science Society. I also just finished teaching a course at the University of Graz on adaptive content modeling and I am working away on my first drafts of a book on the same topic.
As it turns out Bob and I are both pushing the envelope on the definition of a book – a lot. Hopefully, we, and other authors who pile on, will push it to the point of breaking and change the future of books forever. But there are major challenges along the way.
From Authoring a Book to Modeling Content
This is not a book review, but to understand why I think “books must die,” it’s important to understand both the nature and the history of Bob’s book, The Discipline of Organizing, (TDO) — originally published in May 2013 by MIT Press, with the second and third editions published by O’Reilly Media. As far as I am aware, it’s the first professionally published book of its kind.
TDO is a textbook applicable to a range of industries including:
- web and commercial business
- library sciences and information sciences
- philosophy and linguistics
- and more – with a total of 11 disciplines so far
In fact, the back cover of the book includes a bar chart of how much content is relevant to each field. It’s the result of a collaborative authoring process with Bob as owner of the table of contents and the editor for some twenty contributors.
The book was originally written as a more traditional manuscript. As it progressed, Bob realized he wasn’t practicing what he preached. Consider the stakeholders – users, if you will – of a university-level textbook around the discipline of organizing. There are:
- Graduate level
- Undergraduate level
- The various authors
- Professional readers in a variety of fields
Then multiply that audience complexity by the usual channels and potential delivery formats available to the modern author. There’s print and eBook. With a textbook, there is potential for interactivity such as quizzes.
What appear as key points listed in a print book, could become interactive question-and-answer pairs in the electronic form. However, not all eBook readers support the same level of interactivity, therefore several electronic versions are required.
Then, multiply the complexity again because all the users will have specialist interests. For example, a librarian, a content strategist, and a computer programmer will have very different interests, and find different examples and references authoritative and relevant. A content strategist and modeler like myself, might be most interested in examples from the web, business, and some from information and cognitive science. If certain references or examples aren’t given, it would be hard for me to take it seriously as a relevant work for my industry.
Some quick math gives us:
Eleven disciplines of interest x 6 audience profiles x 5 output types, or a total of 330 different perspectives.
That’s a lot of different perspectives to take into account. Then consider that any one person might want any blend of the discipline-specific material. Bob calculates there could actually be 2048 variants of his book, depending on how you slice and dice the content.
The challenge stops sounding like “authoring a book” and becomes very much the challenge of a digital professional creating, in the terminology I use, a body of adaptive content that works across multiple dimensions. To satisfy all the user requirements, this “book” didn’t just need to be rewritten, it needed to be fully redesigned and remodeled to work differently for different people, at different levels of depth, in different contexts across time.
From Books to Multiple Dimensions of Adaptive Content
To optimize the experience of all his users, Bob updated the content design, restructured, and did a deep rewrite, moving all the various discipline-specific materials – 26% of the book –into specifically tagged independent units. Things like case studies, examples, tables, notes and references were each named and tagged, so that they could be clearly distilled out of the “core” of the book. Each component of content was recognizable as to what it was and who it was for. In theory, the supplementary material could now be read, or not read, depending on user interests or situations.
All discipline-based content could be automatically highlighted using colors, icons, box-outs or whatever combination worked best. You could read straight through the core, and only circle back to the examples when studying for or preparing for a test, or vice versa. Suitably, just like software or a website, they did extensive user testing of the layout (use of colors, icons, spacing, etc.) after having refined the input model.
A key take-away: all reviewers and co-authors thought that this massive structured authoring effort actually improved the quality and readability of the material.
However, once you’ve created nicely modeled content like this, now you have a challenge: your content has too much potential!
In my classes, I teach students how many content deliverables — especially books — are two-dimensional things. Each page is generally its own static content space with only page length and width. They aren’t optimized to work in 4 dimensions, with layers (depth) of additional content, audience-specific content, on top of the core material. They also can’t display differently depending on what you’re wanting to get out of them at that moment in time. Even in eReaders, jumping back and forth between references and the endnotes is tedious. Popping up windows with extra layers of information is fraught with technical difficulties and a complete lack of standard features to rely on. And there’s simply no concept of temporarily turning off content that you might want to read later, or that just isn’t of interest to you personally.
And in print? Forget about it.
Adaptive content is content that is ready to bend and flex to the particular situation and need of the audience and publishing channel, but the content being ready doesn’t mean that the world is ready.
From Chickens to Eggs, and Breaking the Book Model
So Bob has a book that currently no one, anywhere, can realistically publish in all its glory. We can assume that 2048 editions would be pointless as there wouldn’t be enough differences between many of them. However, Bob got it down to four very useful and meaningful editions by grouping the material together for related fields like Cognitive Science and Linguistics in a “Sensemaking Edition,” and Information Architecture, Linguistics and Web into an “Information Architecture Edition,” and so on. But four parallel editions are a lot to release at once.
Compounding the challenge, Bob found that once this clean, clear structure was created, it suddenly became so much easier to contribute to it. Now case studies and examples are coming in from far and wide — with over 50 and growing steadily. A new edition or set of editions could be warranted every year (or less). And, of course, you can imagine this becoming a digital community, with local, customized versions — versions that adapt on-the-fly depending on user selections and behaviors and so on.
Although Bob’s publisher is one of the most sophisticated in the world, their whole book supply chain would need to change to be able to accommodate such a pace — even of static editions. As a smart company, they are not going to do that to publish this “one book” of Bob’s. Here we encounter several chickens and a few eggs:
- A publisher isn’t going to gear up to handle such dynamic, adaptable content when there aren’t other authors writing books to publish that take advantage of it
- eReader creators aren’t going to establish standards or improve their range of features when there’s no books and no explicit demand
- Authors aren’t going to try to write such books because no one can publish them or read them in eReaders, nor will they deal with the complexities of most structured authoring tools
- Structured authoring tool vendors won’t package solutions or refine their tools as needed because of all of the above
- Audiences are not going to demand such things from books as that’s what they expect from the internet and apps, but not books. So demand will not increase.
And round and round we go.
Killing The Book As We Know It
As it is, TDO has gone through 3 editions. The 3rd editions – two, not one – came out together in August 2015. The Professional Edition contains all the supplementary content; the Core Concepts contains none of it. This makes four editions of the “book” in two years, 3 months. This is, in the “book world,” an absolutely break-neck pace. But for me, and Bob, and hopefully many other authors who don’t want to publish books for a single reading and a single, grossly generic “reader persona,” it’s still too slow and limiting.
A vast number of books are not created for linear consumption; they’re created for communicating knowledge. The benefits for readers, and dare I say, human culture, are enormous. Being able to consume a book in such powerful ways aids transfer of knowledge, and that drives the world forward — today more than ever.
So why do I think books must die? So that they can be reborn.
Bob and I don’t see ourselves just as authors, but (also) curators of a structure, which we and others can contribute to, increasing the value for all. I think it’s the only way we’re going to get out of those chicken-egg loops is to have more authors step out of the book model and bring out valuable adaptive content.
We need to kill books as they’re envisioned today and move to something like “content collections” or “author-maintained IAs” or “Bodies of Organized Knowledge.” I like that last one… we could call them BoOKs for short.
Until we decide on a new name, do yourself a favor and buy the book!