Ebooks Choices and the Soul of Librarianship

Over the last few years, as a fifth of American adults have gotten ereaders, ebooks have transformed the book market and reading landscape. The library market is no exception. There’s now an array of established vendors and emerging options for libraries to choose from in order to deliver ebooks to patrons.  In my job as the librarian at one of the emerging options (Unglue.it), I’ve seen the pros and cons of various models, and thought about what those mean.

Here’s my conclusion: ebook models make us choose.  And I don’t mean choosing which catalog, or interface, or set of contract terms we want — though we do make those choices, and they matter. I mean that we choose which values to advance, and which to sacrifice.  We’re making those values choices every time we sign a contract, whether we talk about it or not.

Andromeda Yelton is one of the founding members of Unglue.It. Photo by Molly Tomlinson, http://www.photoclave.com/

Traditional vendor platforms provide access to a lot of content, while freeing libraries from the complexity of negotiating for it. But cloud hosting, restrictive terms, and uncertainty about content ownership undermine library use cases.  Emerging models often have less content, but more flexible terms.  (For a thoughtful summary, see Brett Bonfield’s recent post on In the Library with the Lead Pipe.)

What does this mean for library values?  A recent Library Journal article states, “Our primary role is to champion the rights of access for our users.”  I’d like to challenge that statement.  Access is one of the core values of librarianship, but we have others. Privacy. Sharing. Preservation. Paper books can serve all those values simultaneously.  Ebooks bring them into tension. Let’s talk about that.

Privacy

“With a printed book, there’s no such thing as an analytic. You can’t tell which pages are dog-eared,” said Scholastic editorial director David Levithan in a recent Wall Street Journal article. Historically reading has been private; there’s no way to tell which passages a reader stopped on, which books are finished or abandoned, who wrote the marginalia. Ebooks can be equally private, but they generally aren’t.  And libraries — which have a grand tradition of standing up for users’ privacy rights — can’t protect data stored with third parties. In fact, under the Electronic Communications Privacy Act of 1986, the government does not even need a warrant to seize data in the cloud.

These issues were exposed when Amazon ebooks became available through Overdrive. Although this was an important step forward for access, patrons and librarians were surprised to see emails from Amazon inviting them to buy the book as borrowing periods ended.

These strings can’t be attached to paper books. They’re anonymous, offline, and untracked. But ebooks raise a host of important questions: what data is collected on patrons and their reading habits? How, where, and for how long is it stored? Under what circumstances does your vendor’s privacy policy permit it to be disclosed?  What do your state laws and institutional policies say about your responsibility to protect patron information — and do your vendor’s policies, and national laws, even let you fulfill that responsibility?

Sharing

Second, libraries value sharing. It’s what we do: acquire resources like books, meeting space, and expertise, and share them with the community. In the case of books, we’ve been able to do this historically because of the right of first sale. This means that, when we purchase a book, we own it; among other things, we can loan it or give it away. Ebooks, however, aren’t generally purchased; they’re licensed. These licenses fall under contract law, not copyright law, and as such can impose additional restrictions.

People can’t give their used ebooks to the library, as a patron of the Douglas County public library discovered to his chagrin. Publishers and platforms can, and do, limit files to one simultaneous user, 26 total checkouts, access only while online or via specific devices. Digital rights management (DRM) software, instead of library circulation policy, imposes limits on sharing.

Preservation

One of libraries’ key roles for millennia has been cultural memory: preserving the records and ideas that made us who we are.  It’s why we feel horror and violation at the threats to Timbuktu’s cultural heritage today. Preservation matters.

But bluntly, we can’t preserve files we can’t keep. Digital preservation requires the ability to make lots of copies in lots of places, to protect against inevitable server failures and other damage.  (This is why the collaborative system for electronic journal preservation is called LOCKSS: “Lots of Copies Keep Stuff Safe.”) Preservation requires the ability to check files against each other or authoritative sources, in order to guard against bit rot or malicious change. It requires the ability to shift files to new formats as devices, standards, and needs change.

We can’t do any of this with most ebooks. The files don’t live on library servers. License terms can restrict copying, archiving, and format shifting. This is why the Amazon/Overdrive deal was news; if libraries had had the ability to shift those ebooks to .mobi format on their own, no deal would have been needed. This is why Jo Budler’s high-profile move to 3M was news: because Kansas’ contract let them keep the files during the move, but she had to fight to enforce that — and the contract language has since changed to prevent other libraries from doing the same. This is why it was news when HarperCollins imposed a 26-checkout limit, and Amazon deleted copies of 1984 from users’ Kindles.  We don’t expect that books we own can vanish.  But, of course, we don’t own ebooks; we merely license them. Their cultural memory isn’t ours to keep.

Access

Fourth, there’s the value of access to information. This is the value that usually drives discussions of ebook models: our patrons want this content, in electronic format, and we need to provide it.  Serving patrons and staying relevant matters; libraries are, in a fundamental way, about access to information.  However, the access question is complicated, because it’s not just “access to what?”, but “access for whom?”

Does your ebook platform provide content that’s compatible with all devices? If the content can be loaded on a given device in theory, can it also be in practice? Some ebooks must be downloaded to a computer before they can be added to a device, meaning they’re in-library access only for patrons who don’t own computers. Some require a complicated, multistep setup process which not all patrons can navigate. And what about five years from now, when some patrons haven’t upgraded their original devices, and others have something we can’t even imagine today — will all those devices still be compatible with the formats and DRM on your ebook platform?  Do your terms let you do anything to adapt the content you’ve licensed to the devices your patrons use?

And what about accessibility? As I noted above, tech savviness can be an accessibility hurdle. So, of course, can physical impairments.  In May, blind patrons filed suit against the Free Library of Philadelphia for its Nook lending program, because this ereader (unlike some others) is not accessible to blind users. Some publishers disable the read-aloud feature for their books, making them useless for some readers.  Ironically, this happens even when end users have full legal rights to that sort of format-shifting; Alice in Wonderland, which is in the public domain, and Lawrence Lessig’s Creative-Commons-licensed ebook Remix have both had publishers disable their read-aloud feature.

Finally, are we giving patrons of today access at the expense of patrons of tomorrow? Consider the serials crisis. In the mid 1990s, academic libraries started purchasing titles as bundles, lowering their per-title cost but also reducing their purchasing flexibility — the now-ubiquitous Big Deal. This greatly expanded journal access to patrons of the time, but locked libraries into a marketplace that wasn’t in their long-term best interests. Now, journal costs are skyrocketing and libraries are having to make hard decisions about content access for monographs as well as journals. When the bundles the must-have journals are in start to cannibalize your budget, what other content and services have to go?  In essence, libraries increased article access for patrons of the 1996 at the expense of both article and monograph access for patrons of 2012.

The same situation is happening today with ebooks. We have options which increase library ebook access for today’s patrons — but the marketplace they create doesn’t always reflect libraries’ best interests, and that’s going to affect tomorrow’s patrons. If we buy into a status quo today that can triple ebook prices and slash how often a book can circulate overnight, what will be the terms for the must-have content of 2020?  What content or services might we have to cut then to support the system we’re cultivating now?  When you have to choose whether the patron of 2012 and the patron of 2020 gets access to information, which do you choose, and why?

Tradeoffs

I’ve asked a lot of questions in this essay, and I don’t have answers. Sometimes there are no good answers. Sometimes the best answer will vary, by library and by librarian, according to your mission and your patrons and your contracts.

The one thing I know for certain about the future of ebooks in libraries is that it’s about tradeoffs among deeply held values. Right now, we have lots of options which protect ebook access via established distribution chains and publisher agreements — but they also limit it through DRM, restricted format support, and outright refusal by some publishers to sell ebooks to libraries. Negotiating preservation can be complicated or impossible; privacy questions lurk; and checkout limits put sharing at risk.

The eleven emerging models profiled in the In the Library with the Lead Pipe article, referenced above, provide different tradeoffs.  (Disclosure: one is my employer, Unglue.it.)  They vary in who hosts the files, whether libraries are free to make copies, whether DRM is applied, and what legal terms govern the use of the ebooks. This gives each of them a unique set of values tradeoffs. In general, they offer libraries more options in terms  of sharing, preservation, and privacy.  However, right now, all emerging models are limited in terms of access.  Some are theoretical or in a prototype stage, so they don’t offer any content yet. Others require libraries to negotiate themselves for content access, rather than outsourcing this to a distributor. While this puts libraries in a better position to advocate for their values, it also means more work.

I believe it’s important for libraries to do this kind of work. We need to have passionate, engaged conversations, with our eyes open, about which values we most want to defend in the ebook fray — and which we’re willing to compromise on. We need to consider which of many imperfect models offer the best tradeoffs for enacting library values. And we need to do this, not just in service to the patrons of 2012, but to the patrons of 2020 as well. How do the choices we make today affect their options for private, shared, lasting, accessible ebooks?

Vernor Vinge spoke about libraries at Midwinter 2010, characterizing them as “guardians of the past, handmaidens of the future”. Indeed, they are. But our ebook choices, and the different values tradeoffs they represent, rewrite that quote. Will we be active, thoughtful guardians of the future?  Can we remain in service to the best parts of our past?

Andromeda Yelton is a member of the founding team at Unglue.it, where she does library outreach, web development, and communications.  She writes and speaks about library technology and personal branding, is active in LITA, and is a 2011 ALA Emerging Leader.