The Ebook Cargo Cult
Libraries created the present crisis in scholarly publishing, and we are creating a similar crisis now with our approach to ebooks.
We created the crisis in scholarly publishing by ceding control of an intrinsic library function, abstracting and indexing, a decision with inevitable consequences. Consequences like the present need to boycott Elsevier for its predatory practices. Consequences like libraries spending as much money as we can muster on only just minimally justifiable user experiences: bundled interfaces that are confusing individually and often unusable collectively, which is why many libraries spend even more on federating services like Summon in order to offer search to our users in a way that makes sense to them.
Hiring abstracting and indexing firms at the beginning of the 20th century was an understandable decision. When abstracting and indexing services were first established, libraries’ reach exceeded their grasp. Late-19th and early-20th century libraries knew they wanted to collect everything they could, and to make those collections accessible to as many people as possible, but there was not enough funding or infrastructure for them to do it collaboratively: there were fewer libraries and librarians, less specialization, and library training was still in its infancy. In addition, libraries had not yet reached the kinds of cooperative agreements that would have made it possible to divide the abstracting, indexing, and archiving of serials into manageable tasks and then share the results. What we had was an inefficient system, one in which a lot of libraries were putting redundant effort into the same core works, and few or no libraries were able to adequately cover that era’s smaller or more niche publications.
When there are inefficiencies in a system, entrepreneurship and private enterprise are generally the best ways to create efficiencies. Informed librarians, acting individually but uniformly, made a calculated risk, choosing to select and store serials themselves, and hire abstracting and indexing companies to catalog this material. Within our hierarchy of values, we placed immediacy above ownership, and convenience above preservation. And so, when it comes to serials, the library’s inherent character is compromised: the core values we apply in our other activities, most notably our work with books, are not applied to serials.
To date, the decision to hire firms to provide abstracts and indexes for our serials is the largest mistake libraries have ever made, leading inexorably toward the indexing and eventually the archiving of newspapers, magazines, and journals being controlled by a small group of commercial enterprises. Creating the same sort of problem for ourselves with ebooks cannot be justified. While there were a few wealthy, sophisticated libraries at the time, collectively we did not have the funds, expertise, or consortial structures in place to handle cooperative abstracting and indexing 125 years ago. Now, with ebooks, we have the resources we need to avoid repeating our greatest error. The question is whether we have the will to maintain our values rather than simply preserve our role, or at least what we have come to think of as our role.
Cargo Cult (Library) Science
In a now famous speech, his 1974 commencement address for the California Institute of Technology, physicist Richard Feynman popularized the idea of “cargo cult science”:
In the South Seas there is a cargo cult of people. During the war they saw airplanes land with lots of good materials, and they want the same thing to happen now. So they’ve arranged to imitate things like runways, to put fires along the sides of the runways, to make a wooden hut for a man to sit in, with two wooden pieces on his head like headphones and bars of bamboo sticking out like antennas—he’s the controller—and they wait for the airplanes to land. They’re doing everything right. The form is perfect. It looks exactly the way it looked before. But it doesn’t work. No airplanes land. So I call these things cargo cult science, because they follow all the apparent precepts and forms of scientific investigation, but they’re missing something essential, because the planes don’t land.
This is how I characterize the cargo cult thinking I keep hearing around ebooks: “We’re librarians. It’s our job to pay for access to books and then share them with our community.” These are our apparent forms and precepts. For instance, at the 2012 Public Library Association conference in Philadelphia this past spring, I attended a discussion on ebooks that was so popular we ran out of chairs and had to turn people away to keep from violating the Convention Center’s fire codes. During a small group discussion that took place during the session, one of the presenters told me that by boycotting HarperCollins, and by making plans to leave Overdrive, the only vendor that supports Amazon Kindles, I am making the library irrelevant for my neighbors—that I was in danger of losing a whole generation of users. When we reconvened as a full group, another participant, to general agreement within the room, said that she was looking forward to being able to offer access to ebooks the way we offer access to journal articles. The presenters even suggested that HarperCollins may be “one of the good guys” because they are among the few big publishers currently willing to sell their ebooks to libraries (despite the fact that they self-destruct after 26 uses). I have begun hearing this phrase in connection with HarperCollins more and more since that session, including a recent interview with a 2012 Library Journal Mover & Shaker on the Circulating Ideas podcast.
When we think this way, we are still “waiting for the airplanes to land,” going through the same motions as though nothing has changed while undermining the intent of those activities and compromising the core values of our profession. What we have done throughout the modern history of libraries is rely on and uphold fair use and first sale. We have fostered and protected intellectual freedom. We have helped to preserve our culture. If there is a science unique to library science, these are its tenets and its bulwarks.
When we throw away privacy and ownership in order to distribute ebooks onto Kindles via Overdrive, when we abandon first sale and fair use in order to purchase ebooks from HarperCollins, when we tacitly endorse our partners’ use of digital rights management (DRM), we are abandoning the principles that have made libraries valuable to our communities. When we sign contracts that ignore our first principles, we are abandoning our moral and financial obligations to the people whose library collections and intellectual opportunities we are supposed to be stewarding. We are “paying for stuff” and we are “sharing it with our community,” but unlike before, we are not actually buying anything.
When we stop buying authorized copies, when the people who have hired us no longer own the material we are purchasing with their money, we are ceding control. And when we cede control, someone else seizes it away from our neighbors, colleagues, and students—the people who entrust us not to let that happen. This not only drives up costs for libraries, but puts access to knowledge in a tenuous position: by relying solely on commercial enterprises, such as Amazon or Netflix or Spotify, to preserve works and make metadata about those works accessible on an ongoing basis, we increase the likelihood that materials could become more difficult to locate and analyze, and create the possibility of that information becoming permanently inaccessible. If these works or their attendant metadata are not seen as sufficiently profitable, or if the company that controls that information is unable or unwilling to share it, this material is likely to vanish.
Individuals who choose Amazon or Netflix or Spotify for their personal use are taking an understandable risk, since the cost-benefit ratio may well work in their favor, even if their access is ephemeral or otherwise limited. But if libraries, collectively, elect to ignore society’s long term needs for preservation, and immediate need for intellectual freedom, we are making a decision whose consequences seem likely to work against everyone’s interests. Information tends to spread: if even one entity is preserving and sharing it, it tends to make itself widely available. This lowers the barrier to entry for businesses (e.g. Amazon and LibraryThing, as discussed in “A Useful Amplification of Records That Are Unavoidably Needed Anyway“) and lessens the need for individuals to maintain their own archives, as well as bolstering libraries’ case that our core values remain relevant in an increasingly digital world.
Implementing a sensible, long-term plan for acquiring ebooks may be the most pressing issue in American libraries today. According to the often quoted January 23, 2012 report by the Pew Internet and American Life Project, “The number of Americans owning at least one of these digital reading devices jumped from 18% in December to 29% in January.” And serving smartphone users is just as important, or soon will be, given that the reading and listening experience on a phone is likely to continue to improve until the difference between reading on a phone and reading on an ereader is akin to the difference between reading a mass market paperback and a hardback. If we as librarians wish to provide the greatest possible access and highest level of service to people who own these devices while also upholding the core values that best serve readers, we need to develop ebook practices and software that complement one another. To do that, we need to ensure that the negotiations with our vendors maintain the balance of control that has traditionally served everyone’s best interests, helping publishers and other vendors maintain their profitability and promulgate reading, supporting libraries’ mission (e.g., intellectual freedom and preservation), and giving readers the greatest possible variety of choices in what to read and how to access it, while also protecting their privacy and confidentiality.
Learning from the Crisis in Scholarly Publishing
The present crisis in scholarly publishing is the inevitable consequence of the decision to privatize serials cataloging and archiving. Together, libraries and vendors created an oligopoly, a hydra with multiple heads and an unappeasable craving for library budgets. Resources that not only have to be bought, but have to be bought from a single source (or just a few sources acting more or less in concert with one another), are generally going to be produce steady, predictable profits for the companies that own these resources. Warren Buffett refers to this as a moat that protects companies from their competitors, and he seeks out companies that have this kind of moat, such as recent purchases Burlington Northern railroad, which has few or no competitors along many of its rail lines, and newspapers that are the only paper in the towns and cities they serve. Examples within libraries include the most prestigious journals in a field of research and the latest bestsellers.
Oligopoly pricing, especially unregulated oligopoly pricing, leads to unbalanced negotiations, which in turn leads to significant, ongoing profits. And those profits are often best spent expanding the oligopoly into complementary markets. This is why the abstracting and indexing companies bought up journal distribution, and eventually the journals themselves. Think of Google branching into hosting content (by purchasing Youtube, Blogger, and the Usenet archive through its Deja News acquisition, as well as partnering with libraries to scan books) and providing internet access (through the Fiber Project in Kansas City, as well as its open spectrum purchase). Amazon regularly purchases complementary businesses as well. When you have a monopoly, or you are part of an oligopoly, you have a lot more control over what you charge than you do when you compete with other enterprises in a more free-flowing marketplace.
This is the reason that the Random House executive who authorized higher ebook prices for libraries should probably be fired. Not for tripling ebook prices, but for not doing it sooner. Or for not quintupling them instead. It seems safe to assume that libraries have continued to purchase enough Random House titles to increase the publisher’s profits, because Random House has not reverted back to its previous prices. Which means that Random House was giving away money to which it had an obvious claim, and is likely doing so even now. In addition, despite what would seem to be a great opportunity for libraries to demand more in return for higher ebook prices, I am not aware of any libraries having negotiated features like greater privacy or accessibility on behalf of our users, or any other features that represent libraries’ core values.
Do We Have Any Good Options?
The only ebook options that uphold libraries’ core values either provide libraries with ownership rights for the authorized copies we purchase and circulate or are free of DRM software, ideally both.
Libraries need to retain ownership of authorized copies in order to perform key library operations, including acquisition, organization, dissemination, and preservation.1 In the United States, the legal basis for libraries has been the combined effect of two copyright exemptions, fair use and first sale (which was recognized as a doctrine in 1908 and became law in 1976). Fair use allows library users to make use of library materials for research and education, and allows libraries to archive these materials. First sale gives libraries the privileges of ownership, including the right to lend materials to library members, and even the right to sell materials from their collections should we choose. The Digital Millennium Copyright Act (DMCA) compromises libraries’ ability to serve our communities, in part by weakening first sale, but the law went into effect in 2000 and libraries have managed to continue to serve the public in spite of its limitations.2 From the perspective of intellectual freedom, DMCA was a terrible law and the American Library Association has worked hard to fight against it. But it took HarperCollins, and complicit libraries, to demonstrate our apparent willingness to move to a post-first sale world in which libraries license the books we circulate rather than own them.
As ALA reports on its website under the heading, Digital Rights Management (DRM) & Libraries, “The purpose of DRM technology is to control access to, track and limit uses of digital works.” This hinders our ability to disseminate materials, hurting usability, and also increases the likelihood of obsolescence, hindering our ability to preserve materials in a way that guarantees our ability to access them in the future. More chillingly, DRM undermines intellectual freedom, as described in “The Right to Read” (1997) and “The Digital Imprimatur” (2003).
Though DRM is an inherently flawed technology, there is a possibility that it could be replaced by creating better software. At the 2012 ALA Annual in Anaheim, Peter Brantley referred to the possibility of replacing our existing DRM technologies with “tenable protection measures,” and the International Digital Publishing Forum has released an “initial statement of requirements for a potential ‘lightweight content protection’ technology.” I think we are more likely to replace DRM by developing business models and agreements that enable libraries to acquire ebooks while also helping to ensure that authors, editors, and others involved in publishing get paid for their work.
Rather than focus on the materials that are not presently available with ownership privileges or without DRM, it is more useful to focus on the many options that are. There are currently eleven viable ways for libraries to offer ebooks to our communities while still maintaining our professional commitment to our core values.3
As summarized in the table that accompanies this article, the available options are:
- Open Library
- Open Source/DRM Hybrid Model
- Unlimited Content License Model
- Portability Model
- Publisher Hosting Model
- Library License
- Sneakernet Model
- DIY Model
- Steampunk Model
- State Redistribution Model
By analyzing every existing model that either involves ownership or is DRM-free, libraries and people who are interested in libraries can develop a consensus around realistic and efficient models, and can create and adapt software, licensing, and other practices to support the models that are described below. Some of these models are intended to be self-contained solutions to the ebook situation, most notably the State Redistribution Model. Other models can be combined with one another, as noted in their descriptions.
Open Library, which has been endorsed by all fifty state librarians in the Chief Officers of State Library Agencies (COSLA), partners with libraries to create a nationwide, shared ebook collection. Participating libraries contribute at least one print-based book to the shared collection, which Open Library digitizes and then locks away in long-term storage, so the print-based copy of the book cannot circulate. Instead, Open Library circulates the digital copy of the item.
In return for contributing a book to the Open Library collection, participating libraries’ cardholders are given access to all of the ebooks in that collection (Open Library retains ownership of the authorized copies of ebooks on behalf of the participating libraries.). Access is much like it is in many consortia, in which a single item circulates to one person at a time within the consortium. One-borrower-at-a-time is enforced by a DRM-enabled server that uses the same underlying technology as Overdrive, 3M, Ingram, and Baker & Taylor—Adobe Content Server. Open Library has also experimented with purchasing new, original content, though with limited success, and this material is currently a tiny portion of the overall collection.
In addition, Open Library mimics a feature that is familiar to Nook owners who have taken their Nooks to a Barnes & Noble store and taken advantage of the bookstore’s “Read in Store” service, which provides free, unfettered access to the Nook catalog of ebooks and digital magazines for up to one hour per day. Like Barnes & Noble, Open Library partner libraries offer greater access to people who are using Open Library from within the physical structure of a member library, though they do not limit access to one hour per day.
Open Source/DRM Hybrid Model
Perhaps the most promising and fastest growing model is the Open Source/DRM Hybrid Model pioneered at Douglas County Libraries in Colorado. The model is for libraries to deal directly with copyright holders, have them sign a simple Common Understanding license, and host the ebooks either on an open source server if the rightsholder is willing to allow for more liberal circulation of their ebooks or, if the rightsholder wants more restrictions, Douglas County also hosts ebooks on its own Adobe Content Server.
The Open Source/DRM Hybrid Model’s most visible advocate is the library’s director, Jamie LaRue, who writes about it frequently, including in the recent “All Hat, No Cattle: A Call for Libraries to Transform Before It’s Too Late.” One sign that this model is viable: in addition to being copied by other libraries, the system is also being adapted by a commercial entity, Bibliotheca, which three weeks ago announced that it had hired Douglas County Libraries’ IT Director, Monique Sendze.
Unlimited Content License Model
In the Unlimited Content License Model, which was created by Ann Arbor District Library, publishers agree to allow the library to distribute unlimited, unencrypted copies to their cardholders, and libraries agree to pay the publishers a flat annual fee. If the library chooses not to renew its payment, it must remove the files from its servers when the term is up. Additionally, the publisher makes explicit its “reps and warranties” that the content is theirs to license, and indemnifies the library against infringement claims if they made a mistake. Finally, the publisher agrees that the library’s users can make copies and derivative works for their personal use. The most notable example of this model is Ann Arbor’s agreement with Magnatune to license the music label’s audio file collection, though there is no reason this model could not work for ebooks as well.
This model is perhaps best associated with the State Librarian in Kansas, Jo Budler. When Kansas signed its first contract with Overdrive, it negotiated the right to transfer the titles it bought through Overdrive to another vendor should it so choose, a right that Overdrive sought to eliminate after the first contract expired. Budler refused to sign away that right and Overdrive refused to include a portability option in the second contract: all titles purchased through Overdrive would not be allowed to leave the Overdrive platform, including titles that had been purchased during the first contract.
Budler had to find a vendor that would not insist on locking her into its platform, and she and her colleagues also had to write to each of the publishers whose books they had purchased through Overdrive to affirm that the publishers would not seek to block Kansas from transferring its authorized copies to another host. All but two or three publishers agreed to the transfer, and Kansas eventually signed a contract with 3M. It is my understanding that King County, which is working with Baker & Taylor’s Axis360, has similar rights to portability for the authorized copies it purchases.
Publisher Hosting Model
This model is most prevalent in contracts between academic libraries and scholarly publishers, though this is more a matter of convention, and of large libraries’ buying power, than a reflection of anything intrinsic to the ebooks being sold. For instance, Cornell University Library, which will not enter into vendor contracts that require nondisclosure of pricing information or other information that does not constitute a trade secret, focuses its attention on contracts that call for either perpetual or archival access (the language for these contracts is based on the NorthEast Research Libraries Licensing Guidelines and Model License). Perpetual Access is when a library’s cardholders have access to the ebooks a library has purchased even if the library does not have a subscription to the vendor’s ebook platform. Archival Access, which is similar to the Portability Model, is when a library retains the right to transfer an authorized copy to its own servers or to another vendor.
Unglue.it uses a Kickstarter-like crowdfunding model, but focuses solely on working with rightsholders to set prices for “ungluing” their books: that is, releasing ebook versions of their work under a Creative Commons license. The rightsholders set a price, and Unglue.it helps them get it. Ownership of authorized copies of ebooks in the Unglue.it model is universal, so libraries do not have any special rights or privileges, but there is also no reason for libraries not to add successfully relicensed ebooks to our collections, and we are free to contribute whatever we can afford toward the campaigns to make currently inaccessible works universally available. Unglue.it has also built a large catalog of books that are candidates to be relicensed; for libraries that are interested in gaining unrestricted access to out-of-print material, or who have a strong interest in having a particular title relicensed under Creative Commons, the Unglue.it community is a useful place to let that be known and to find others who share that interest.
Library License, proposed by Jeff Goldenson of The Harvard Library Innovation Laboratory, is similar to Creative Commons, but it is directed solely at authors who want their work to have a special, companion license that allows libraries to own and circulate their work. The initial draft calls for three possibilities:
- An embargo. For instance, libraries could have the work for free, but only after it has been commercially available for, say, five years;
- An institutional “carve-out” in which, for example, Harvard Library can have this work for free the moment is it published, and circulate the work to everyone who has a Harvard Library card, but everyone else has to acquire it through standard, commercial means;
- A performance-based clause, sort of like Kickstarter or Unglue.it: once the publication makes a certain amount of money in sales, libraries will be able to add it to their collections for free.
Unfortunately, any progress that has been happening since this proposal was introduced has been pretty quiet, though Harvard was involved in the creation of Creative Commons, so it was proposed by someone at the right institution. As Goldenson writes, “This is a young project, working itself out as we speak. If you are interested in exploring it further, visit librarylicense.org and reach out.”
A seeming step back, but perhaps the most practical way to wait out the technological, financial, and legal changes taking place within the ebook market. Using the Sneakernet Model, libraries would buy ebooks just like any other purchaser and place our authorized copy on a read-only flash drive. We could then circulate the drive like any other physical object we circulate, such as books, DVDs, and CDs, using our existing software (we could even use bar codes). Making use of a flash drive with an ereader or phone would require that borrowers purchase an extra adapter or dongle, but the added price is less than the cost of two or three ebooks, and the inconvenience would be fairly minimal, if clearly less than ideal. When a new server technology emerges (or an existing model, such a the Open Source/DRM Hybrid Model, becomes simple and inexpensive enough for even the smallest libraries to implement), we could transfer our ebook files to a server and erase them from our read-only flash drives.
DIY, which stands for “Do It Yourself” is an homage to the members of the indie/punk music scene of the 80’s and 90’s, who believed bands could and should avoid major record labels, and even published guides/zines describing how to put out your own records. Specifically, the DIY Model uses Open Library’s model as a precedent and gives individual libraries control of their own authorized copies. Open Library’s interpretation of copyright is that libraries, when they purchase an item, also purchase the right to circulate one copy of that item at a time, even if the library elects to transform that item into another format.
So far, most of the items in the Open Library collection are either public domain or are unlikely to be challenged by a presumptive copyright holder. However, it is not clear that the model would only work in these instances. What if a library were to digitize a copy of a print-based book by a publisher that does not sell to libraries, then lock away the print version and circulate only the ebook it had created? Or what if the library did the same thing with a still-in-copyright item that is not available digitally? The library would own the item, it would only be circulating one copy at a time, and it could be argued that the library is circulating the item in the only way it can, and that it is causing no harm, or de minimis harm, to the copyright holder. Options for circulating DIY ebooks include the Open Source/DRM Hybrid Model and the Sneakernet Model.
Like the Sneakernet Model, the Steampunk Model is another technological step back (in this case from digital back to print), but potentially worthwhile because there are many solely digital works that some libraries might want to circulate in print. One example: Download the Universe reviews science ebooks, most of which are not available in print. Given the limitation of print, authors may agree to “gluing” their books for a modest fee, or to the print-only equivalent of a Common Understanding license or a Library License. Don’t you want your grandfather to be able to read your book? Or the person on the other side of town who doesn’t have internet access at home? And wouldn’t it be cool to see your local library add a print copy of your book to its circulating collection? Libraries know what to do with print-based books, and our members know to look for them when they visit us. The key is to work directly with rightsholders who like libraries or who see us as a useful market for increasing their sales and promoting their work, especially if they would not otherwise consider printing their books at all.
The State Redistribution Model
This is the least viable plan, in that it involves significant changes to current laws. Richard Stallman of the Free Software Foundation proposed the idea, originally in a letter to the Brazilian Government. Under Stallman’s proposal, the author or publisher would retain ownership of the
authorized copies copyright, but sharing them noncommercially copying and redistributing authorized copies would be legal and authors would be reimbursed based on the popularity of their work. Popularity would be measured not by counting downloads or monitoring use of the work in other ways, but by responses to voluntary surveys in which readers would report what they have read. Rightsholders Authors and artists would be paid from a pool of funds set aside by the government to pay authors for their work. (Note: thanks to Richard Stallman for correcting a couple of errors in my original summary of his proposal. In addition, though sharing would be legal in the State Redistribution Model, people would remain free to purchase the books directly. In that case, the purchaser would own the authorized copy. As Stallman wrote, “I think that if you buy a book, you should own what you bought.”)
What Happens Next?
I wrote about ebooks in a previous Lead Pipe article, “Tangoing All the Way: Is Everything Negotiable?” What was different when that article was posted, just over a year ago, is how many fewer options I could identify. At the time, instead of buying HarperCollins books through Overdrive, I was in favor of buying books through Overdrive from other publishers. Now that Overdrive has revealed its insistence on vendor lock-in and its willingness to allow Amazon to violate library members’ privacy, I am in favor of writing off our Overdrive purchases as a sunk cost and of moving on to other options. Constructive engagement has its place, as evidenced by the ReadersFirst movement (though I prefer the North Carolina State University Libraries “Value Statement for the Scholarly Ebook Marketplace,” which is currently hosted on the Triangle Research Libraries: Beyond Print website). But often the most sensible response to a bad option is to ignore it, cut it out of our budgets as quickly as possible, and focus instead on those options that offer greater promise.
While the eleven viable ebook options described above are promising, they are not yet ideal. They will take work. The open question is our willingness to appreciate the important features these options make available rather than the features we wish they had. Are we willing to do what is best for the people we serve in the long term, while perhaps disappointing them in the short term?
Unfortunately, we do not have much choice: even if you work at one of the two New York libraries that have access to Penguin’s ebooks, you still can only offer popular materials from three of the “big six” publishers; for the rest of us in public libraries, we can only offer ebooks from two, and even those include the significant issues I have discussed above regarding HarperCollins and Random House. For academic librarians, I can only imagine the pain involved in watching a student trying to figure out how to make use of some of the ebooks in the university collection before they give up in frustration. I am currently a student myself, and I know that frustration well.
When we remind ourselves about our core values, the answer is obvious, at least the answer about what not to do. If we keep pretending that all of our vendor relationships are serving our cardholders’ best interests, the airplanes will never land. When we cede power, we know where that leads.
We have also seen what happens when libraries meet adversity with innovation. The State Library of Kansas helped establish the Portability Model. Ann Arbor District Library created the Unlimited Content License Model. Douglas County Libraries created the Open Source/DRM Hybrid Model. Previously, the Georgia Public Library Service created Evergreen and North Carolina State University introduced faceting to library catalogs with its Endeca project (which Villanova University helped to perpetuate with its open source VuFind project).
Single agencies, and even statewide consortia, can be nimble, creative, and powerful, though it seems clear that even broader collaboration will be necessary in order to bring our handling of ebooks into line with our core values. Right now, it looks as though the Open Source/DRM Hybrid Model, like Evergreen and open source faceted catalog interfaces before it, is following the growth pattern established by some of libraries’ most successful cooperative projects, such as the MARC bibliographic standard, interlibrary loan technologies and agreements, the WorldCat union catalog, the Z39.50 protocol, and the Library Bill of Rights. But even if none of the presently viable ebook models ultimately enjoy widespread adoption, it is heartening to see how much progress we have made in just one year. Not all that long ago, we had no options that honored libraries’ core values. Now we have several. The next step is to work together, to vet and develop the current options, and to help create new ones if necessary.
- The Publication Standards Project (along with a two-part issue of A List Apart on digital publication standards)
- Andromeda Yelton’s talk at the 2012 Computers in Libraries conference discusses ebooks within the context of libraries’ core values
- “The eBook User’s Bill of Rights,” which was co-created by Sarah Houghton and Andy Woodworth
- “Purchasing E-books for your Library,” a presentation by No Shelf Required’s Sue Polanka, particularly her “eBook Access Levels” graphic
- Pew Internet & American Life Project’s “Libraries, patrons, and e-books”
- Michael Buckland’s Redesigning Library Services: A Manifesto (1992)
Thanks to Michael Bills, Jo Budler, Tim Coates, Shea Crow, Marin Foster, Jeff Goldenson, Amber Kiepe, Jesse Koennecke, Jamie LaRue, Robert Miller, Mary Minow, Eli Neiburger, Michael Porter, John Saylor, Karen Schneider, Monique Sendze, Ari Shanok, Kate Sheehan, Richard Stallman, Katy White, and Andromeda Yelton for their help with my research. I have done my best to include information as it was presented to me, though any opinions or mistakes included in this article are solely my own. In addition, thanks to Emily Clasper, John Jackson, and my Lead Pipe colleague, Kim Leeder, for their helpful comments on an earlier draft of this article.
- As Jesse Shera wrote in The Foundations of Education for Librarianship:
“The processes of selecting, acquiring, and making available for use the contents of graphic records comprise the operational aspects of librarianship. Librarianship is a trinity of acquisition, organization, and dissemination, in which acquisition relates to the selection and accumulation of materials, organization to their preparation for efficient use, and dissemination to the processes of making the contents of graphic records available to the user” (Shera, 1972, p. 193).
Libraries have since greatly expanded their work into non-graphic records (such as audio, video, and games), but acquisition, organization, and dissemination remain a good characterization of what we do. Along with Shera’s list, it seems sensible to include preservation, privacy, and, increasingly, production:
- Libraries play an important role in helping to identify the books that people can trust, or are especially likely to enjoy in their leisure time.
- Metadata created and preserved by libraries is a public resource, as are the standards libraries develop to organize this metadata.
- Making materials available remains relevant, even when these materials are digital: just because people can afford ereaders does not mean they can afford ebooks. In many libraries, including the one where I work, video games and DVDs have the highest per-copy circulation even though the equipment required to make use of video games and DVDs is far more expensive than an ereader.
- HathiTrust is an important component of the digitization efforts that have taken place to date, as is the Internet Archive, but these efforts apply primarily to preservation and access for works that were initially released in print. Libraries need to make different arrangements for work that exists only in electronic form. As David M. Levy wrote (about the early history of the Web) in “Digital Libraries and the Problem of Purpose” (2000), “…we have been rushing to put materials online with no thought to (and no idea of how to) preserve them. One consequence is that, almost certainly, there will be a hole in history….” Given that libraries’ relationship with serials, especially electronic journals, can be seen as a lesson in how not to work with commercial interests, it is worth noting the recent report by the university libraries at Columbia and Cornell which finds that only “15-20% of the e-journal titles in the libraries’ collections are currently preserved” by libraries’ two leading digital preservation initiatives, LOCKSS and Portico. One way for libraries to avoid these kinds of difficulties is to find ways now, while ebooks are still relatively new, to begin preserving copies of ebooks in ways that will ensure their continued availability.
- This also includes confidentiality and intellectual freedom. From the ALA Core Values of Librarianship, “Protecting user privacy and confidentiality is necessary for intellectual freedom and fundamental to the ethics and practice of librarianship…. We uphold the principles of intellectual freedom and resist all efforts to censor library resources.”
- Libraries can help authors publish, and not just in the traditional sense of library-as-place, e.g., Marx in the Reading Room of the British Museum, surrounded by documents and given a quiet place to work (for a more contemporary example, listen for the Chicago Public Library’s role in “What Gave You That Idea?,” a recent episode of the 99% Invisible podcast). Though it remains an important library function, library-as-place is less central to users’ interactions with ebooks, which are generally downloaded and read outside of libraries. However, with relatively modest cooperative investments in technology and training, the connection between libraries and the publication of ebooks could be far more practical: helping authors turn their manuscripts into ebooks, and helping them find their audience (see Nate Hill’s “A two part plan to make your library a local publisher”).
- In part because every three years the Library of Congress can determine exemptions for certain activities. [↩]
- The following companies and models are not considered viable ebook options either because they do not provide for the sale of authorized copies to libraries and because the ebooks they license to libraries are DRM-encoded, or because there is not yet enough known about them to describe their business model or how libraries would make use of their services:
- Bilbary, an ebook store that went live in March 2012 and plans to eventually feature titles from almost every major publisher, has been working with Chief Officers of State Library Agencies (COSLA) to find a way to support libraries. The most likely arrangement would allow for libraries to benefit from an affiliate program for sales referrals, one that could eventually include co-branding. Another possibility would be for Bilbary to facilitate short-term rentals, sort of like a cloud-based, ebook version of McNaughton Subscription Service, though libraries would be charged per transaction rather than paying a monthly access charge. All titles would include DRM and no authorized copies would be owned directly by the library; instead, libraries would subsidize or underwrite cardholders’ rental costs.
- Freading (by the company that offers Freegal) provides pay-per-download, DRM-encoded ebooks.
- Ingram’s MyiLibrary allows publisher to decide how they want to handle DRM, and at least one MyiLibrary customer, Rosen Publishing, says that its ebooks are sold without DRM. It is not clear if purchases are transferable to another vendor or server.
- Library Renewal, a project led by Michael Porter, is currently planning a groundbreaking system that would eliminate inefficiencies and unnecessary expenses in the ebook market, and offer transparent pricing for publishers, rightsholders, and libraries.
- OverDrive does not provide DRM-free downloads and changed the provision in its contracts that supported library ownership of the copies that libraries purchase through OverDrive.
- Safari has a consumer product that allows for DRM-free downloads, but access to the titles in Safari’s library product is online-only. Safari was started by O’Reilly Media, whose founder, Tim O’Reilly, proudly states that his company’s books are sold without DRM. Except, it seems, to libraries.
There is another very important role that the library once played, in the market for academic periodicals, especially in the sciences. Libraries once helped limit the impact of the “moat” on the journal publishing market, by offering an alternative.
It was once quite common for science professors with a research focus to subscribe to paper journals. Office walls were often lined with them. However, the library offered the same journals. The scientists just found that the benefit of having their own copies, in front of them at all times, outweighed the cost, at least for some journals. Publishers had a clear market incentive to keep prices down to the point that they could sell multiple subscriptions. There was good money in selling more subscriptions, but if they raised the price too high, individuals or labs with subscriptions would, en masse, just say, “Screw it! We’ll just look at the copy in the library, for that money.”
With the shift to electronic journals, these incentives were turned around, completely. Instead of acting as an alternative source for information that would encourage price consciousness in a journal’s many customers, the library began to act as a layer of abstraction between the consumer of the good and the supplier, which means the consumer is insulated from the price, and just wants the good. The supplier loves this, because it’s a lot easier to sell something when price is not even discussed.
This is the classic “third-party payer” problem in today’s market for medical services — the patient receives the services from a doctor or other provider, but the insurance company pays. The patient has no personal interest in the price, and my experience has been that hospitals won’t even quote prices to a patient (something that would be illegal in most businesses, but I digress).
As with medicine, prices of journals in the sciences have gone up at multiples of the rate of inflation. The scientists need them, accreditation might even require them, but the library foots the bill. Where once the presence of a copy in the library helped keep costs down, today, a library subscription helps vendors charge far more than the market would bear, if actual consumers had to weigh their individual costs and benefits. In a tragedy of the commons scenario, universities and, if they are public, taxpayers, ended up getting screwed by the unintended consequences of libraries’ direction. In the rush to embrace an obviously superior new technology, I don’t think anyone really considered that this would be the result.
There is a real danger that e-books will follow the same path. Libraries are in a position, right now, to take a stand to see that they don’t make the same mistake, again.
Great thoughts on the ebook issue, but I’d like to comment on your statement “To date, the decision to hire firms to provide abstracts and indexes for our serials is the largest mistake libraries have ever made, leading inexorably toward the indexing and eventually the archiving of newspapers, magazines, and journals being controlled by a small group of commercial enterprises.”
Perhaps it’s not too late to begin, if not to completely correct the mistake, to amend it. As libraries move into a greater involvement with university presses, institutional repositories, and other forms of scholarly communication, libraries should insist on taking control of the metadata (seizing it by force, if necessary). We know they can do a fine job with it, but they also need to get the metadata out there so the publications can be found without the aid of an A&I service. Get it into the catalogs and on the web. Make it easier to find than the other guy’s stuff. Maybe this kind of action by libraries could generate enough momentum to pay for itself in saved subscription costs to indexing services. Or maybe there’s something glaringly obvious I’m missing as I try to write this on what’s left of my lunch hour. Either way I enjoy reading your blog. Congratulations on your award!
I agree that it’s not too late to correct our mistake. Thanks for pointing that out, and for discussing some really encouraging ways that’s starting to happen.
What if librarians started the take-back of indexing and abstracting by taking back our own literature?
Imagine a cooperative effort to index and abstract LIS literature. Maintain it in a wiki, with each article having its own page. A template would guide reviewers in entering structured information about the article. Non-reviewers could comment on and ask questions about the article and vote on whether they would recommend it. The wiki could have author pages, with each page having links to all of the author’s abstracted articles, and registered user pages, with each page having links to all of the user’s contributions to the wiki.
If the idea works–if a critical mass of library practitioners read and contribute to the wiki–then it could eventually evolve into something more: A place for authors to publish their work, bypassing the journal process altogether. A place for a community of readers and writers of LIS research to form, connect, and grow. And a training and testing ground for the larger project of re-taking indexing and abstracting.
I love this idea. A few years ago, Gabriel Sean Farrell and I played around with a project we called LISinfo, and we hoped it might some day include the information you’ve described above. Unfortunately, we never got the project even to an initial release. I hope you get farther with it than we did. Please let me know if there’s anything I can do to help.
Reply to both Brett and Steve:
Well, thanks to you Brett for bringing up the indexing issue in the first place. Ever since, I’ve been thinking a lot about this and have even more ideas. I’m so excited I may even write a blog post about it. Too bad your project didn’t get off the ground but timing is, as they say, everything. And I think now may be the time.
Steve, that’s an interesting idea, using a wiki. To tell the truth, that never entered my mind, but all possibilities should be examined!
Brilliant. You had me at “To date, the decision to hire firms to provide abstracts and indexes for our serials is the largest mistake libraries have ever made, leading inexorably toward the indexing and eventually the archiving of newspapers, magazines, and journals being controlled by a small group of commercial enterprises. Creating the same sort of problem for ourselves with ebooks cannot be justified.”
I agree completely with your assessment and it makes me sick to see libraries continue to play the fools in this eBooks game. I also appreciated your reference to our core values. I think it is high time we all step back and look at those core values and rethink what it means to stand up for those core values in the 21st Century.
Pingback : truthberry pickings « desperately seeking truthbrarian
“The only ebook options that uphold libraries’ core values either provide libraries with ownership rights for the authorized copies we purchase and circulate or are free of DRM software, ideally both.”
Ah, so in other words: in order to be considered a librarian with values, one must accept your core assumptions prima facia. This is a bit of rhetorical grandstanding that neatly places all dissenters into two camps: people who are not librarians or librarians with faulty values.
This moral grandstanding and unnecessary scolding tone greatly reduces the impact of an otherwise astute and well-researched argument.
I consider myself a librarian with a solid core of professional values, but I reject the unstated assumption that a library must always be, by definition, an organization that purchases content from publishers and distributes that content to our patrons.
That is print thinking, pure and simple. If libraries insist on defining our services on print terms (see also: first sale doctrine), we will become irrelevant sooner rather than later. Digital media are different from physical media. There are, of course, better and worse deals we can make with Big Content, but a shift from an ownership model of librarianship to an access model is taking place. If our profession, by sheer force of will, insists on holding to a service model based on the affordances of printed materials, there will be no role for us in a digital world.
While in general I agree that concerns with ownership is “print thinking,” as long as publishers (and Amazon) continue to treat their digital content as if it were print, with DRM and the like, trying we will need to engage in print thinking to give our patrons access to the materials they desire. At some point, the publishers’/Amazon’s ability to treat digital content like print content will be undermined. Yet until that moment comes (and at the moment is seems far off, though one can never tell), choosing access over ownership frankly just seems to be giving publishers the right to toy with us and prevent us from providing sustainable access (or even any access) for our patrons. Library’s reliance on the first sale doctrine will only be irrelevant when there is another way to ensure that we are able to give out patrons access to material.
Yes, from a context of providing services in the here and now, this (and the original post) is clear practical thinking.
My objection is that I cannot imagine a future of libraries where circulation is still our primary task. Thus, if the future is providing access without ownership, efforts to enforce the ownership/distribution model appear to me as conservative actions intent on reinforcing the status quo.
Reading the OP again, it reads less like grandstanding, but it still insists that library values require an ownership/distribution model. I can’t seem room for that in any future, other than as a niche role and not in the mainstream.
eBooks are not books. That is to say, electronic texts are not codices. This post is a highly developed argument on the ethics of distributing codices.
Nicholas, thanks for the critique.
I’m working on a couple of essays, one about libraries in 2025 and another about libraries in 2112, so I’ve been thinking a lot about the future of libraries. What do you see as our future?
In response to Nicholas, I think that the concern for ownership is not primarily an attachment to “print thinking” but something that stems from our sense of responsibility for the preservation of the cultural record. It doesn’t seem that any other societal institution has a sense of that responsibility, so it’s natural for us to want control of usable files (ownership amounts to control) in order to preserve them. I don’t pretend that the library or the archival profession has digital preservation all sewn up, but it does seem to be our responsibility.
In response to the article, I would only take issue with Brett’s claim for the function of fair use exemptions in library lending. I don’t think that is the part of copyright that we actually rely on in lending books or in making copies for preservation purposes. Section 108 and other parts of the copyright act are more important to what libraries do. I think our interest in fair use has more to do with occasional needs of library users.
Rory, thanks for clarifying this for me. Every time I think I have an understanding of how copyright works in practice….
Also for Nicholas… Another reason for libraries’ interest in ownership is also not based on an attachment to print, but rather on the idea of universal access. If vendors maintain ownership of the files they sell, it is more complicated and not necessarily possible to provide universal access. Rights holders may succeed in charging end users a per-use fee, which sacrifices the thing that makes libraries different from bookstores. I think that in the end, librarians don’t have as much power to shape the next stable situation of publishing and distribution of texts as Brett thinks we do. But I think we will manage to negotiate some arrangement that provides limited free access to ebooks to members of communities, paid for by communities on a subscription basis. I think publishers will end up going for that if it is workable for them financially. DRM of some kind will be a part of that picture, unavoidably, IMO.
Pingback : To Own or to License? | EdReach
FWIW (just found this out) Safari doesn’t run its library platform itself — contracts out for that. Which may or may not change your calculus with respect to footnote 6 :) But I totally respect if their cost/benefit analysis or development resources don’t make developing an in-house library platform feasible (I’m sort of keenly aware these days of how one cannot Develop All The Desirable Things), and it may be that third-party options don’t provide DRM-free library distribution platforms (& they don’t have the negotiating leverage to make that happen).
I really like O’Reilly media and want to use library funds to buy their books, so when I first started researching this article I was excited to learn that Safari offers DRM-free downloads. When I emailed Safari to ask for information about DRM, ownership of authorized copies, and pricing, I was informed that libraries can only buy O’Reilly publications with the restrictions I mentioned. The footnote itself is meant to be a brief summary of that email exchange, along with a bit of context.
Pingback : Curiouser and Curiouser: What caught our eyes online this week | Chasing Reference
Pingback : Ebooks Choices and the Soul of Librarianship — The Digital Shift
Pingback : Libraries and the cargo cult mentality « St. Kate's MLIS Reorganization Discussion
Pingback : Digital Public Library of America » Blog Archive » Digital Library Digest: August 3, 2012
Pingback : Open Access … Open A&I? Thoughts and Questions | Librarian Squared
Pingback : O que os bibliotecários precisam saber sobre ebooks? | Moreno Barros