Digital Libraries, Controlled Digital Lending & eBook DRM

CDL: Controlling ebook distribution and lending periods

How digital libraries differ from physical ones, controlled digital lending, and how ebook DRM can be used to control distribution, enforce lending periods and ensure copyright is respected.

Digital Libraries and ebook lending

Digital libraries differ from physical ones in several fundamental ways.  Shelf space is usually no longer a constraint, and searching the stack is a simple exercise.  Rare volumes can be made ‘available’ without risk of damage or mass distribution — provided you have adequate security over the digital copies and suitable licensing control(s).  However, digital libraries also have new challenges, such as spiralling license costs, lending abuse, and potential copyright issues.  This blog will discuss some of the key elements digital libraries must consider, covering:

1. The history of libraries
2. How library ebook lending works
3. The Internet Archive’s controlled digital lending lawsuit
4. How pirates defeated the Internet Archives ebook DRM
5. How to protect ebooks from misuse and enforce lending
6. Closing words

   The history of libraries

Libraries have evolved steadily since their first recorded inception more than 5000 years ago.  A brief timeline of library development helps to illustrate how rapidly they have evolved in the past century:

• 3000–1300 BC:  The first libraries are formed in present-day Iraq and Syria.  Back then, the practicalities of librarianship mainly consisted of having a good enough oven to cook the clay tablet and solid shelving to take the weight.  Details like copying were hardly a consideration, as the expense and difficulty of production made it implausible.  Not to mention the fact that the earliest libraries were likely not public and were mostly used as administrative/commercial archives.
• 1600–1100 BC:  The first libraries in China are founded under the Shang dynasty.  They include texts such as genealogies, notices and orders, records of natural events, etc.  They are generally not public but rather governmental, private, academy, or temple collections.
• 550–246 BC:  The classical antiquity period is known for its libraries, and this particular period is known to have birthed several famous libraries, including several in ancient Persia, the Persepolis Fortification Archive, the Library of Perganmum, and, most prolifically, the Library of Alexandria, which is thought to have used the first library cataloguing system.
• 27–476:  The first public libraries were established under the Roman Empire in an attempt to rival the library in Alexandria.  Successive emperors competed to open new and better libraries than their predecessors.
• 600–700:  In China, woodblock printing, which was previously used to print patterns on cloth, started being used on paper for the first time.  It quickly spread, across south-east Asia.  Suddenly, you could make exact copies, which meant you could have more books, but it wasn’t easy.  Carving the blocks used for printing was highly-skilled and time-consuming work.
• 800-1000:  China’s papermaking was acquired by the Muslim world, which developed it into an industry.  Improvements in speed and quality meant books were more easily manufactured, and public and private libraries boomed in Islamic lands.
• 1450:  Gutenberg invents movable type printing and changes everything.  By 1500, printing presses in Western Europe have produced over 20 million volumes.  For libraries, the issue quickly becomes not finding books to catalogue but having enough space to store them.  With no real copyright laws in place, publishers are forced to constantly print new works to remain profitable, as their releases are cloned shortly after release.
• 1653:  Chetham’s Library opens as the old free public reference library in the English-speaking world, with public libraries before this point typically requiring a subscription.
• 1710:  The Statute of Anne, also known as the copyright act, establishes modern concepts such as public domain and a copyright register.  This introduces new aspects for libraries to worry about.  See also IPR protection & what is copyright.
• 1850:  The true beginning of free open-access libraries begins in the UK with the Public Libraries Act, which recommends that libraries are established all over the country and introduces taxes for a public library system.  This all created the need for administrative solutions.  Books had to be catalogued, borrowers had to pay subscriptions and were limited to the number of books that could be borrowed at any one time or for a period of time, to even out demand.  Authors had to be paid.  And all without computers.
• 1947–1971:  The earliest electronic books were created, and there is still great debate over what should be considered the first “true” ebook.  Douglas Engelbart and Andries van Dam are likely the first creators of the ebook form closest to what we see today.
• 1960:  The birth of the online public access catalog (OPAC), an online database of materials held by a library or group of libraries.
• 1971:  Project Gutenberg launches in an effort to digitize, archive, and make available copyright-cleared books.
• 1976:  The Copyright Act of 1976 establishes fair use guidelines that libraries must follow when copying digital resources.
• 1998:  The US introduces the Digitial Millenium Copyright Act (DMCA), aiming to better address digital copyright while providing an exemption for non-profit libraries and archives that allow three copies to be made, one of which is digital, for internal use only.  It also allows libraries to copy works if a format has become obselete.
• 2011:  The Internet Archive’s Open Library begins its Digitize and Lend scheme, and soon after, the model of controlled digital lending was theorized by Michelle Wu.

Today’s library landscape is a difficult one.  They must manage a combination of digital and physical assets while preventing now trivial copying, keeping track of license payments, and providing sophisticated tools to search for and manage catalogues.  Most libraries employ a Library Management System, which helps them to manage all this, but even so, it is a significant challenge just to keep the lights on.

   How does library ebook lending work?

There are generally two digital library ebook lending models: licensing and controlled digital lending (CDL).  These two approaches have similar end goals but different ideas for achieving them.

Loaning library licenses for ebooks

Making a licensing deal with the publisher is the most common way to do ebook library lending.  In such a model, the library never owns the book – rather, they pay the publisher a monthly or yearly fee and can continue lending for a specified period of time as long as they meet the license terms.  Those terms can include factors such as how many digital “copies” the library can lend at once, how many borrows total, or sometimes even a set price each time a book is checked out.  Some libraries make the licensing deals themselves, but many outsource it to services like OverDrive or  BorrowBox, who make the deals and then pass the fees onto the libraries in a more simplified form.

How much do libraries pay for ebook licensing deals?

It depends.  As mentioned, many different factors go into licensing agreements, as well as different apps, licensing services, etc.  Most libraries, however, will tell you that they pay far too much.  The American Library Association (ALA) offers some examples where an ebook that costs consumers $12 can cost a library $60 on a two-year lease term, with added restrictions.  Perpetual licenses are a rarity, but ALA gave an example where a 20-year license for a single copy of an ebook cost $519.  In other words, digital book licenses usually cost several times more than physical books despite costing less to create and distribute.

  How does Controlled Digital Lending work?

With so many people now used to the instantaneous convenience of digital content consumption, libraries have been forced to adapt their offerings to a hybrid of on-premises and digital services.  However, those who have peeked at their library’s online offerings will have noticed that it generally pales in comparison to their physical stock.  Indeed, according to the Tenessee State University, only 8% of printed journals and even fewer scholarly books are available in digital format.

So came the practice of controlled digital lending (CDL).  What if libraries could lend digital copies of books?  By scanning and recreating them in digital formats, they could theoretically protect the physical condition of older volumes while increasing convenience for the customer and remaining on the right side of copyright law.

This theory was first posed by Michelle Wu in 2011, but was later built on by David R. Hansen and Kyle K. Courtney.  It’s partly based on the law of first sale, which allows owners of print books to sell them to others without infringing on the publisher’s rights.  A resulting white paper laid out a series of requirements for libraries that are lending physical copies digitally:

1. Make sure original works are acquired lawfully.
2. Apply CDL exclusively to works the library owns, not works they have licensed.
3. Limit the number of copies in circulation to the number of physical copies the library lawfully owns.
4. Each digital version should be lent to single user, similar to a physical copy.
5. Have the same lending period as physical copies.
6. Use digital rights management (DRM) to prevent users from widely copying and redistributing content.

Sounds simple.  Simple enough that various libraries have adopted and benefitted from the model over the past decade.  Just because something is theorized to be legal, however, it doesn’t necessarily mean it is.  CDL went largely untested until the case of Hachette Book Group, Inc v. Internet Archive.

   The Internet Archive’s controlled digital lending lawsuit

What if controlled digital lending could be used for more than just out-of-print or rare books?  The theory was that a digital library could bypass expensive ebook licensing by creating scans of in-print physical books.  They would pay once to own one copy of the book to lend to one person at a time – like it had always worked.

The Internet Archive was one of the first to popularize this model.  The company has run a non-profit digital library since 1996 with the noble goal of preserving culture and making it freely available.  It initially archived the web but later added digital and digitized physical books to its collection via its Open Library project.  Many of these titles were created in collaboration with over 150 libraries.

Everything ran smoothly until the pandemic when the Internet Archive began its “National Emergency Library” scheme.  This briefly removed its owned-to-loaned ratio and was a step too far for some publishers.  Hachette Book Group, Penguin Random House, HarperCollins, and John Wiley & Sons filed a copyright infringement lawsuit in the Southern District of New York.  The Association of American publishers stated that “IA displays an abandon shared only by the world’s most egregious pirate sites”.  Ebook piracy is acknowled as being a common problem.

The district court ruled in favour of the book publishers, which Internet Archive appealed to the Second Circuit Court of Appeals.  The second circuit court agreed with the district court.  The arguments for copyright infringement are as follows:

• The Internet Archive’s use of the books was non-transformative because they are available in full, and its digital copies serve the same purpose as the originals.
• Though the “first sale doctrine” does allow owners of print books to sell on their print copies, it does not allow for the reproduction of a work.
• The books were “close to the core” of intended copyright protection and available wholesale.
• The internet archive’s use was seen as a competing substitute for ebook licenses.
• It did not add meaningfully new or different features to the books.
• Though the second court considered the angle of public benefit, it said that this was outweighed by the harm to publishers and authors.

What does the Internet Archive lawsuit mean for CDL?

The vital point to note is that this ruling applies both to the Internet Archive’s National Emergency Library and its normal use of CDL.  It is clear that some of the foundational arguments for CDL were not viewed positively by the judges.  On the other hand, the rulings specifically concerned the Internet Archive’s implementation, which has some unique elements compared to traditional libraries. For example:

1. The second circuit judge took issue with the Internet Archive’s argument that it was providing an accessibility service.  This is because the IA has no system in place to verify whether somebody has accessibility issues.
2. Many of the books in question were still available as ebooks for which libraries could purchase a license.  The Internet Archive pitched itself to public libraries as a way to avoid paying license fees for ebooks.
3. The IA did not verify that partner libraries had removed their physical copies from the shelf when a digital copy was lent out.  In fact, it did not inform them at all.  Regardless, the second circuit judge considered the 1:1 ratio argument to be irrelevant as the IA was reproducing as well as redistributing.
4. The books were available in their entirety.
5. Internet Archive solicits donations on its website and online ebooks, which can be considered a form of profit/ benefit.  It also has a referral scheme for purchasing used physical copies of the ebooks.

There has been heated discourse surrounding what exactly this means for the future of CDL, with some suggesting it makes no difference and others announcing the death of libraries as we know it.  It’s important to emphasise that the only binding precedent that prohibits this practice currently exists in the second circuit covering Connecticut, New York, and Vermont.  The rulings of this case will be heavily considered in other states should similar lawsuits occur, but it’s not binding.  Naturally, it also should not impact many other countries with different implementations of fair use.  Even so, there’s no hiding that this is a blow for digital lending and a win for the publishing industry and ebook licensing.

So, what’s the path forward for CDL?  For now, at least, many libraries across the United States continue to use the model.  In the second circuit, there may be a path forward where a “watered-down” version of CDL is considered fair use.  As the ruling of this case focused primarily on lending digitized books that already have a previously-existing licensed ebook market, it can be argued that lending ones without is still fine.  Additionally, IA still has the ability to challenge this ruling further.  It can take it to the Supreme Court or ask for an en banc review by all judges in the Second Circuit.

For the time being, the Internet Archive has stopped lending books for which ebooks are on sale.  It has maintained the ability for those with print disabilities to lend books and for small portions of books to be available for those linking to them.

  How pirates defeated Internet Archive’s ebook DRM

One factor in publishers’ decision to sue Internet Archive lay in its DRM (Digital Rights Management) protocols.  In the closing words of its initial complaint, Hachette argues, “Upon information and belief, IA has not developed and does not enforce sufficiently rigorous DRM protocols and related logistical systems that aggregators must employ to ensure that books appearing on their website are not paired or unlawfully infringed by users.”

In our assessment, that certainly does not seem to be an exaggeration.  Internet Archive offers two types of DRM:

• Adobe Digital Editions DRM for downloaded PDFs/ebooks.  This is a popular solution and is used by many other libraries.
• Its web viewer protection allows users to view the digitized books in their browser.

We found widely shared methods to bypass both of them.

Adobe Digital Editions bypasses

In a previous article, we talked about why Adobe Digital Editions is not secure at length.  Removing DRM from protected ebooks is usually as simple as selecting the ebook in a DRM removal tool such as Epubor Ultimate or Calibre and then pressing convert.  This allows users to freely share, copy, and keep publisher’s ebooks without ever paying for them.

Internet Archive web viewer security

Internet Archives web viewer has similar security issues to every other web viewer out there.

It serves each page as an image, which can be easily downloaded from the “source” tab in a browser’s developer mode or by using a tool such as Chrome Cache View.

Due to the web viewer’s search tool, which is powered by optical character recognition, users can also copy and paste text.  They just have to enter a phrase to view all instances of it in the book and then copy the text from the search box.

Finally, the Internet Archive has a “Read aloud” tool that users could theoretically record to create a free audiobook version.  AI transcription tools could convert this back into text.

  How to protect ebooks from misuse & enforce lending

A successful digital rights management software needs to look out for several common misuses when a library implements CDL or ebook licensing:

1. Printing:  Often, when using CDL, libraries are sharing rare limited print editions that have good reasons for being tightly controlled.  It stands to reason that you don’t want lenders duplicating them and sharing them at will.  Ebook licenses, meanwhile, usually strictly prevent users from making copies.
2. Expiry circumvention:  Pirates can exploit library lending, removing expiry dates to create permanent copies of books that can be shared widely and freely with non-members.
3. Location and control:  Limit access to members of valid states and countries is easy enough in physical libraries, but more challenging in digital ones, where users can use tools like VPNs to spoof their location.  The same applies to academic libraries, where it’s easy enough to stop books from leaving the premises but much harder to stop ebooks from being shared or kept.

We’ll take a look at how you can prevent each of these using Locklizard Safeguard PDF DRM.

   Prevent printing of ebooks

Real books can be copied and distributed with enough time and effort.  In the physical world, you can stop allowing sensitive books out of the library, or you can choose a lending period that balances how long it may take to read/study and how long to copy.  Typically, this is a week, but it could be more.  Ebooks do not have the same luxury since it takes no time to send a document to the printer.  So, what to do?

The first option is to prevent printing entirely.  You can achieve this in Locklizard Safeguard by unticking the “Allow Printing” option.  You’ll likely want to tick “Disallow screen capture” in the Environment Controls tab also to prevent printing via a screenshot.  Unlike Adobe security controls or other ebook DRM solutions, users cannot bypass or remove these controls, helping to maintain a 1:1 lending ratio.

That said, copyright law in some jurisdictions may allow the making of a single copy for personal study reasons.  In some, this is the equivalent of a chapter or 10% of the book.  In such a use case, you have to come to terms with the fact that if anyone can make a copy for their personal use, then there is nothing to stop them from giving that copy to anyone without charge.

Restricting to printing to a single copy is not a bad idea.  However, users can still scan that copy and reprint it.  As a result, you also want to make it unattractive to the copier.  You can achieve this by watermarking prints with the lender’s name, limiting prints to black and white copies, or adding a moiré image watermark that will degrade scan quality.  For the best security, you may want some combination of the three.

Locklizard has a range of granular watermarking controls for this purpose, which can dynamically display user information in order to discourage users from taking photocopies.  You can also force printing in black and white or grayscale, log print requests, and limit the number of print copies.

  Controlling period of lending & ebook expiry

Unless they protect lent ebooks from being copied, re-distributed, and retained, libraries could have problems with publishers.  So, that means taking active steps to ensure that borrowers cannot easily make non-expiring copies and cannot just pass them on to others.

Lending periods for ebooks can be anywhere from a matter of hours to weeks.  With digital books, there is nothing delicate or rare about the digital edition unless the library decides to limit access to the documents.  More relevant is the demonstration of licensing control to stop access after a reasonable time for study without giving away ownership (see PDF expiry controls).  For the more legalistically inclined, there is the question of ‘exhaustion of rights’ – do rights transfer with the downloading of the book to the user?  Allowing the download of a copyrighted work may transfer ownership rights unless it is clear that borrowing was a limited license to use and not a sale.

Locklizard’s secure PDF document controls can help establish there was no possibility of considering a transfer of ownership.  In the expiry & validity tab of the PDF Writer application, you select “expire after n days”, or select a specific date, and the book will become inaccessible after that point.  As other controls prevent all forms of copying (save manually typing out the book), there is no way for a user to retain a non-expiring version.

It is also possible to revoke access via your admin system at any time, regardless of where a book is stored, demonstrating that you maintain control.

   Administration, location locking & licensing control

It must be possible in any system to account for the borrowing and return of books, even if they never leave the building – something that can be reproduced by limiting access to the library computer network and closing the building to stop access.  Many academic libraries do not allow anything to be taken off the premises and may lend books to each other to satisfy requests from their users (borrowers).  These need to be administered and controlled.  Locklizard enables libraries to enforce this with IP locking controls.

The fact that a book may be out of copyright may not be the end of the story.  Copyright is the right of the author to be rewarded for their efforts.  But when there is only one copy of a book and it’s owned by someone, there’s no law to say they must make it publicly available for free copying.  The exhausted rights can still allow a licensing regime that prevents the right to further distribution/copying exerted by the document owner.  If the book has been converted to digital form, document owners can rapidly lose control of any rights they may have had unless they take steps to protect the ebook.  This can be detrimental to both partners and libraries – with some libraries attracting patrons primarily through their rare collections.  Locklizard uses access control and printing control methods to help secure the rights of document owners and prevent unauthorized use.  We enable libraries to provide controlled digital lending using ebook DRM to distribute library books securely.

   Digital Libraries: Turning a new page?

The death of CDL and digital libraries is likely to be highly exaggerated.  However, irrespective of recent court cases and your lending model, it’s clear that publishers are now intensely scrutinizing how libraries handle their digital catalogs.  In recent history, libraries in the US, in particular, have been pushed towards paying ebook licenses and accepting license terms that are increasingly limited and strongly enforced.

In an ideal world, publishers want recurring revenue for each digital title a library holds.  Just as they would if they were to sell ebooks online.  Many libraries see this as unsustainable and antithetical to how public libraries have functioned for hundreds of years.  CDL represents a way for libraries to maintain aspects of the physical lending model while staying on the right side of copyright law.  However, as we have seen recently, libraries must be cautious.  For the time being, the legislative winds are blowing in publishers’ favor.  Demonstrating a clear effort to ensure that digital books are not copied, over-loaned, or misused, particularly in the context of CDL, may not be a bad idea.

Locklizard is the best way for libraries to lend ebooks securely in PDF format.  Its integration with library management systems, combined with extensive expiry, location, print and copy controls, grants libraries the peace of mind that copies won’t end up on piracy sites or be misused.  Meanwhile, a choice between fully offline, online, or mixed licensing checks ensures the DRM works for both lender and lendee.  Take a 15-day free trial of our DRM software to see if  how Locklizard can help with controlled digital lending.

  FAQs

Why can libraries lend digital copyright material?

Usually, because they have purchased a license from the publisher which allows them to do so.  Some other libraries, however, believe they can legally lend copyrighted material by following the principles of controlled digital lending.

When did libraries begin lending ebooks?

Most sources suggest public libraries began lending ebooks in the 1990s.  However, it did not fully take off until 2003, when Cleveland Public Library launched the first popular ebook and audiobook borrowing service in collaboration with OverDrive.

What is an online ebook lending library?

A service libraries provide allows users to download and use various digital books on their mobile device, computer, or e-reader for a limited time (usually 7-14 days).

Is library digital lending/controlled digital lending safe?

Yes, digital lending is safe from a consumer perspective, so long as you are borrowing from a legitimate library (not a fake site).  Libraries and publishers can safely lend ebooks, but they need to be cautious of users who try to break their license terms and redistribute borrowed books.

Does library material digitization risk copyright infringement?

Yes, if the digitization is not considered fair use.  Libraries must consider the purpose and character of the use, nature of the work, amount and substantiality of the portion used, and the effect of the use on the potential market of the copyrighted work.  As we have seen with Hachette v. Internet Archive, it is not always simple to figure out if you are on the right side of copyright law, but making an effort to prevent unauthorized sharing could go a long way.