It is one thing for a publisher to be on the Web, another thing entirely to be of the Web.
This thought was prompted by the accidental confluence of two items. First, recently I had an opportunity to talk to a journals publisher who had made a deep commitment to publishing online. But I heard a whispering in the background: “What is this Web thing anyway? We are prepared to put everything we have online, but why is doing this on the Web supposed to be different?” A legacy publisher peeks out from behind a firewall and puts a cautious toe onto the Web, but a new company jumps in head first. If the legacy company is a good one, the timidity is all the greater.
The second item was that I happened to be rereading Brian O’Leary’s excellent piece “Context First,” which argues that publishers make a big mistake when they focus on the final form (the container) their publications will take, as doing so strips the work of its context: the network of relationships with other content, with commentators and reviewers, with the world of tweets and blogs, and even from earlier or adapted versions of the same work. The importance of context is discovery, as the network of information drives up search engine rank and provides valuable clues for readers to find a work. In an information environment of superabundance, discovery is everything. Putting one toe onto the Web means you won’t get discovered, won’t get read, won’t get bought.
O’Leary’s article is astute, and grows in importance with rereading.
I was rereading the O’Leary piece because of a related piece (a contextual post, you could say) by the always-lively Eric Hellman, who is well known to academic librarians. Hellman takes issue with O’Leary, noting that the container known as “a book” has a quality of timelessness that the dynamic nature of O’Leary’s “context” does not take into account. I know both men professionally and hope never to have to compare my SAT scores with either of them, but on this particular dispute, I think Hellman has it wrong.
The whole question of books (or any content; we could say “texts”) and their containers is a vexed one. Hellman appears to be anxious about the loss of the canonical work, but embedding a work into a network of cross-references doesn’t diminish it. I live at 613 Spring Street, and my property loses no integrity by virtue of the fact that my neighbor lives at 601.
As for timelessness, well, yes — immersed in a text, time does indeed seem to stand still, but the fact that a text is connected to other texts does not mean that we focus on the entire network at every moment. Nor does the fixed nature of a text change because someone is pointing to it. Hellman appears to be confusing a publishing strategy of “context first” with the experience of reading. Publishing is not about reading or readers; it is about talking and enticing other media to increase the volume of your own voice. Do unto others such that they will do even better unto you.
I happen to be highly sympathetic to the situations of established publishers and don’t subscribe to the “tear down the walls” perspective of many critics of the publishing industry — and this because I have a businessman’s respect for cash flow — but the Web is not a newfangled printing technology. It is also a suite of enabling tools that makes some actions more probable than others, just as a box of chocolates in your home is more likely to be eaten than one left on a shelf at a shop downtown.
For someone operating a new publishing company, this may be a no-brainer, as no one today would realistically start a company with the print paradigm in mind. A new company would be born of and on the Web; its products would be conceived with all the “affordances”– the inherent qualities — of the Web in mind, from sharing to linking to published APIs. A new company would do this as a matter of course because the challenge for any publisher, old or new, is to bring attention to its products. A legacy publisher can command attention by virtue of its established base — of authors, sales accounts, access to media, its balance sheet (marketing costs money), and its brand. But a new publisher has few or none of these things. What it has is the impulse to scream for attention, and on the Web, screaming in social media is the stock in trade.
How then can a legacy publisher participate more fully in what is sometimes called the Web’s “conversation” the better to augment the visibility — and sale — of its own products? The key may be to take a tip from the entertainment industry, which has long and profitable experience in making its properties available across a network of media types.
Take a feature film — a Star Wars, say — for example. The feature film is launched, is successful. It in turn inspires two follow-on movies, which are then succeeded paradoxically by three prequels. Meanwhile, the merchandising division is going crazy with Darth Vader toys and spaceships; I think of the millions of parents in America who experienced the crunch-crunch of plastic under the Christmas tree. The brand was then extended to television (which supported more merchandise sales) and even books. If you develop the brand creatively, its extension is limited only by your imagination. Or perhaps beyond one’s imagination — it still astonishes me to realize that one such Hollywood brand was able to extend itself all the way to the Governor’s mansion in Sacramento.
Scholarly publishers have been unevenly focused on managing their brands. Or perhaps it’s more accurate to say that they have been narrowly focused on protecting their brands. A networked branding strategy is highly protective of a brand as it is extended to new — and linked — products and services, but it is perhaps not militaristically protective of the content that is an expression of the brand. There are instances where business opportunities are explored as investigations into “repurposing” such as selling individual chapters of books or collecting articles into coursepacks, but these products — all part of the network of connections — are not always conspicuously branded. As a consequence, some of the reverberating benefits of network communications are lost. And without the extension of the brand, it becomes harder to protect both the content and the context.
As scholarly communications begins to move into new areas — into author-pays open access repositories, into linked data, into ebooks with embedded video, into short-form works — the argument for thinking of each product on its own, as a stand-alone product, diminishes. The strategy for maintaining economic control of the various products and services as they participate fully in Web interactions is to assert the brand everywhere, creating a marketing network that parallels the underlying technical network of the Internet.
Discussion
20 Thoughts on "Brand, Context, and Containers: Publishing Into and Across the Digital Network"
But then “the brand” requires re-thinking too, Joe. The brand for book publishing was primarily B2B, with little distinction by subject or audience except in the broadest sense (Harvard University Press means something, even if runs across disciplines).
In the new world of the web, I think the branding should be more audience- or context-centric. The context of the U Press is perhaps necessary but certainly not sufficient in this new environment.
This is a secondary question and doesn’t change any of Brian’s core arguments, but it would change the way a publisher would tag and label things to best contextualize the content being made available.
One challenge with the concept of a “container” is that it’s easily rationalized by the forgiving mind into even a contextual and dynamic entity. I participated with Brian at a PSP debate earlier this year which touched on this topic, and people defending the status quo easily twisted “container” to be things like Wikipedia and Twitter. They are not fixed, so they are not containers in the sense Brian intends.
Perhaps we need a more evocative word, like “kettle” or “box” or “jar.”
I think publishers in general need to spend a little time learning about their new containers. The container that is poised to replace the book, EPUB3, is a profoundly capable one, and publishers that fail to focus on maximizing the value of these software+content objects are doomed to dissipation and irrelevance.
What worries me is that publishers will be lured into failure by adopting a web site business model that fails to launch the book object into the world, free of dependence on the publisher’s continued viability. Having a canonical object is nice, but having a non-expiring object is essential.
I fully agree that context is very important, but it isn’t owned by publishers. That’s why publishers need to enable context, not be fooled into thinking they have to create it.
There really is a disconnect here. EPUB3 is not a container. It is an enabling technology. A hardcover book, measuring six inches by nine, is a container. A quarterly journal, which publishes ten articles per year, printed with a paper cover, is a container. When you define EPUB3 as a container, it becomes so broad, so all-encompassing. Why not call the Web a container? How about the physical universe–or universes?
Joe Esposito
EPUB3 is a container because it carries a complete set of book-stuff inside it. A reader can interact with the self-contained object without reference to the publisher.
You don’t need a whole internet to carry all a book’s stuff around. Really, this isn’t rocket science.
Again, that’s not how Brian meant it, from what I can tell. It is tempting to rationalize non-containers as “containers,” but they aren’t, and Brian was very clear in his post what he meant. There is a difference in that a book “container” from 1896 is essentially the same as it was then, just older and more time-worn. If you look at it, you’re looking at what someone from 1896 looked at, just with your own brain and eyes. An EPUB3 book from 2011 might be a completely different experience in 100 years, presented in a different device, linking to different things, incorporated into different things, accessing new multimedia, etc. It’s not a container. It’s an enabling technology, as Joe said.
I completely agree — this isn’t rocket science.
Joe, I’ve got to disagree with you there. You define the book when you issue an ePub edition; you put some things in and you leave some things out. That means it is “contained.” An app is also “contained.” In fact, what you choose to put on a web page is ALSO “contained.”
I think Brian’s point is that publishers should stop making things for a specific contained use, whatever it might be.
One point Eric makes that seems right to me is that publishers won’t control context. But although it is right, I’m not sure that it matters. Making context clear will enable discovery and appropriate placement of the content even if somebody else controls the context.
Just to return to Brian’s definition of “container,” since, as I mentioned, there is always a temptation to overinterpret: “the physical containers we have used for centuries to transmit information. Those containers define content in two dimensions, necessarily ignoring that which cannot or does not fit. Worse, the process of filling the container strips out context – the critical admixture of tagged content, research, footnoted links, sources, audio and video background, even good old title-level metadata”
So, I think that comports with Joe’s notion that a container is not an enabling technology, since EPUB3 can carry context in the sense Brian intended. It’s not a container.
Thanks, Kent. Point taken. So “containers” are physical, since digital iterations can always contain the links, etc.
Eric, I don’t think Brian is against the presentation of content in a container. What he is arguing is that the creation workflow not be aimed at a particular contained version. And actually, except in very unusual instances, it is hard to imagine a situation where his logic wouldn’t apply. If you don’t do what Brian says, then you either trap the content in its container or create extra work and expense for yourself to exploit it some other way. What bothers you about that?
Let me try to use some less encumbered words to make the discussion more useful.
There are two sensible ways to structure a digital production process. There’s an object-oriented process, where you assemble an ur-object from which you can derive delivery objects. That’s opposed to a stream-oriented process, where you basically wire up data flows.
Brian rightly points out that book publishers need to add more “stream” to their object-oriented process. But I think that most trade books are best produced with a mostly object-oriented process, and not go to the stream-oriented process. Stream processes ARE appropriate to more dynamic, data oriented products like newspapers, cookboooks, travel guides, etc.
O’Leary can speak for himself, but my understanding is that he is only incidentally talking about a digital *production* process. He is talking about a *publishing* process, which happens to be digital. It’s now clear to me that this entire conflict stems from the fact that you did not read Brian’s piece carefully and that you imposed meanings on his terms that are yours and yours alone.
One more try.
“physical container”= codex.
“container” without “physical” = could be some sort of standard digital format.
“context”= “the critical admixture of tagged content, research, footnoted links, sources, audio and video background, good old title-level metadata”.
“publishing is unduly governed by the physical containers”. AGREED.
“we must start with context and preserve its connection to content.” I’m not sure how you start with title-level metadata, for example, unless you’re Demand Media.
“Containers limit how we think about our audiences.” OK, I’m fuzzy on how that works.
“Think Craiglist. Think Monster. Think Cookstr, a born-digital food site that started with and continues to evolve its taxonomy. Context first. ” Those are built in a what I call a stream-oriented process. They aren’t “published” in the publishing-industry sense.
etc.
Apologies for weighing in late; I was traveling for work the last two days.
I appreciate Joe’s initial post and the discussion that followed. In the comments, Joe and Kent have made arguments that I agree with and support.
“Container”, for me, is a shorthand for the vehicle used to transmit content to a reader. When I wrote “for centuries”, as Kent picked up, I was also trying to set the stage that publishing workflows treat digital uses as a derived or secondary use.
In my essay/presentation, I try to say that creating content for specific containers limits the ability of publishers to maintain that content in a way that supports both anticipated digital and unanticipated downstream uses. It’s part of a call to change workflows, so that authorial and editorial insight about content is captured and maintained internally, even if in the near term the primary outputs remain physical containers.
I think all periodicals, including journals, and most books would benefit from such a change. Today, most publishers fail to link even title-level metadata to the content it describes (creating back-office challenges of some note). That means that discovery of these long-form works is almost entirely left to channel partners.
Today, book publishing in the United States amounts to a $40 billion industry, of which perhaps $7 or $8 billion is trade fiction. Even if we agree to exclude all trade fiction, there are immediate and important opportunities to move traditional book publishing to more agile workflows, and every product we create using traditional methods adds to the backlist of things we’ll need to fix later or abandon.
In his most recent comment, Eric Hellman defined a number of terms, three of which I might filter differently.
– The “physical container” is not synonymous with “codex”; it is in use and applicable to periodicals as well as books.
– Delivery of a digital format is still a container, albeit a more flexible one, but it helps only if workflows are changed to create and maintain context that may or may not be used by the digital contained.
– “Containers limits how we think about our audiences” is drawn from my belief that publishers assume that the (mostly physical) format defines what readers need. We don’t know that; we just know that we sell a lot of these containers now. The growth of short-form content, interstitial reading and media “lockers” in the cloud suggest that there are untapped opportunities.
I do agree with Eric’s assessment that the evolution of workflows could be segmented into “stream” versus “object-oriented” approaches. Using different language (frequency or revision and likelihood of reuse), this is something the presentation addresses early on. Publishers can decide if context is a front-end or a back-end priority, but I don’t think they can avoid it in an environment in which search increasingly drives awareness, trial and purchase.
Thanks, Brian, for the explanation of the audience comment. I thought you meant the container limited our thinking of who the audience was, not how the audience interacted with the content. Thus my puzzlement. Indeed, with digital transmission vehicles, even the word “audience” is obsolete, as it implies a one-way interaction.