blogs by Gio

CDL: Publishers Against Books

Combining lending with digital technology is tricky to do within the constraints of copyright. But it’s important to still be able to lend, especially for libraries. With a system called Controlled Digital Lending, libraries like the Internet Archive (IA) made digital booklending work within the constraints of copyright, but publishers still want to shut it down. It’s a particularly ghoulish example of companies rejecting copyright and instead pursuing their endless appetite for profit at the expense of everything worthwhile about the industry.

Controlled Digital Lending🔗

“Controlled Digital Lending” is a system that enables libraries to lend their texts digitally. It is exceedingly careful to stay within the legal and philosophical confines of copyright law, while still allowing libraries to serve their primary purpose and lend their books. CDL is in active use by many major libraries, including the Internet Archive’s Open Library. With CDL, libraries lend DRM-protected, temporarily-provisioned digital scans of real, physical books the library owns, which are taken out of circulation for as long as the corresponding digital copy is still usable.

So how does it work? Quoting from the 2018 Controlled Digital Lending whitepaper:

David R. Hansen & Kyle K. Courtney, A White Paper On Controlled Digital Lending Of Library Books This paper is about how libraries can legally lend digital copies of books.

CDL enables a library to circulate a digitized title in place of a physical one in a controlled manner. Under this approach, a library may only loan simultaneously the number of copies that it has legitimately acquired, usually through purchase or donation.

Circulation in any format is controlled so that only one user can use any given copy at a time, for a limited time. Further, CDL systems generally employ appropriate technical measures to prevent users from retaining a permanent copy or distributing additional copies.

The system we propose maintains the market balance long-recognized by the courts and Congress as between rightsholders and libraries, and makes it possible for libraries to fulfill their “vital function in society” by enabling the lending of books to benefit the general learning, research, and intellectual enrichment of readers by allowing them limited and controlled digital access to materials online.

According to the whitepaper, for a book to be usable for CDL it must be owned outright by the library, not simply licensed or borrowed. This is a strict standard (even stricter than physical books, which you can sub-lend), but a fair one, especially given the desire for CDL to cleanly fall outside any right copyright reserves to the publisher.

The Internet Archive (IA) was a pioneer of CDL, but after the framework was published libraries everywhere were eager to adopt it. More than that, libraries want CDL to be a standard practice worldwide:

Position Statement on Controlled Digital Lending by Libraries endorsed by COSLA The Board of the Chief Officers of State Library Agencies (COSLA) recently endorsed the Position Statement on Controlled Digital Lending by Libraries as a thoughtful analysis that presents a unique model for leveraging print collections in the digital world. Originating conceptually from the copyright community and pioneered by the Internet Archive through their Open Libraries program, Controlled Digital Lending (CDL) allows libraries to digitize older in-copyright print books.
… A broad library-based community of practice is emerging around controlled digital lending. To date, 371 libraries have implemented CDL. COSLA’s endorsement of the CDL practice serves as encouragement to libraries throughout the United States to explore Open Libraries as a service partner.

The National Information Standards Organization (NISO) is currently in the process of developing technical standards for implementing the whitepaper, although libraries are already using the techniques it lays out.

Preventing Reproduction to Enable Lending🔗

Normally the reason owners don’t have a simple right to “lend” things digitally is it infringes on the reproduction right. In the real world, the “natural”, default operation is movement — denying the lender use of the item — while with digital media the default operation is copying. Since data is infinitely copyable and digital storage is read/write, you can’t “move” a file without making two equally-usable copies of it, both of which are themselves copy-factories for unlimited more copies. While you can effectively move a file, you can’t “do it in one”, you have to compose the “move” operation out of copy and delete operations.

This is the essence of the core “reproduction” right of copyright: while a work is under copyright, only the owner of the monopoly can reproduce and publish new copies. Real-world items can easily be transferred without violating the reproduction right, but this isn’t so for digital media, as files aren’t actually “moved” between devices as an atomic operation, only read and copied. Unlike analog distribution, digital distribution typically necessitates a “reproduction”, which — in the absence of special corrective measures — infringes on the reproduction right.

This means as soon as something is digitized, it becomes impossible to share without a “license” from the copyright holder because doing anything involves reproducing the work. So with digital files, traditional property rights are suddenly very, very limited. Most incidental reproductions are covered under personal use exemptions, but providing media to another party is not. This is why, by default, you can’t just “lend” a file as a natural extension of your rights over it. Without a system in place to ensure the data is only available one place at a time, copying a file to lend it is duplicating the media for non-personal use, which is not allowed.

But, as much as I despise it in general, DRM fixes this problem and is ultimately the glue that makes current implementations of CDL work. The DRM-enforced scans2 patrons borrow with CDL are not reproducible digital copies. Because CDL copies are only usable within the library’s lending period, they’re also not substitutions for owning a book. They’re not even substitutions for owning a scan of the book, let alone a copy of the book itself. They are — as enforced by the law, code, and cryptography — ephemeral, non-reproducible, unsharable images to allow temporary remote use of a physical book. Copyright grants the rightsholder (temporary) control over the number of copies in circulation, and CDL is carefully designed to preserve that right, and enable remote lending of physical books without infringing on the reproduction right.

Format Shifting🔗

The process CDL describes in their whitepaper is a kind of “format shifting”: converting existing property to a new form or temporarily expressing one thing in a different format, as with “ephemeral copies”. The digital version of the book is an intermediate artifact: that form only temporarily exists as a necessary step to facilitate the lawful (and necessary!) behavior of lending an owned book.

Fundamentally, the scans the Internet Archive loans aren’t ebooks, but are instead representations of physical books in a temporary digital form. Loaned scans have some of the benefits of ebooks (more on this in a moment), but ultimately what patrons “get” are only shadows of the physical books themselves. So it’s important not to confuse CDL with “ebook lending”, as lending officially published made-for-digital publications is its own separate industry (more on this in a moment also). CDL lends scans of physical books, so other than both just “using computers”, they’re not directly related.

Some of the books available via CDL do have separately published ebook releases which CDL is unconcerned with. CDL does not describe any way to lend purchased digital ebook files — while the right to lend digital property is an important problem, that’s not part of this conversation at all. CDL only facilitates digital lending of physical texts.

CDL has a deeply ingrained dedication to working within the law, in both letter and spirit, both in its purpose and throughout its inner workings. This is not describing a system of mass piracy, but rather solving the challenge of how to reasonably express owners’ right-to-lend using digital means.

Any book used in CDL must be bought and paid for by the library already, meaning authors and publishers are fully compensated. Publishers, in the act of selling the copy, authorized that particular copy to be in circulation. (That’s what “sale” is.) Hypothetically, the Internet Archive could instead be mailing the books to each patron, like Netflix used to do. CDL lets libraries provide this same functional service without violating anyone’s rights by simply improving delivery to people already entitled to receive the content.

In fact, in addition to all the other limits, the Internet Archive has a specific policy for their Open Library to not scan or lend books within the first five years of their publication — the first year of which, statistically, contains the bulk of sales — as a way of ensuring authors are compensated, and its library doesn’t impact book revenue. This olive branch gesture means IA is actually giving even more deference to publishers than brick-and-mortar libraries do.

All this ensures the process of CDL is functionally identical to the existing conventions of book lending. The only thing getting updated is the use of modern technologies to make the practice of remote lending fast, practical, and efficient for both libraries and their patrons. CDL is game-changing not because it represents a new paradigm, but because it’s a practical, resource-efficient mechanism to make the established lending process viable for a non-profit.

Ironically, the Internet Archive’s Open Library isn’t actually all that useful for any of my personal needs. The stress of having limited time to read a book and having to orchestrate the logistics of pickup and return means I exclusively purchase-to-own ebooks and physical media. The library use case is a very specific one.

Because CDL intentionally applies the limitations of physical lending to its digital copies, IA’s lent copies are equally undesirable to me because they have all the same limits. I didn’t like CDL because the DRM was very restrictive. I like collecting books and saving them for later reference, and when it comes to its scans of commercial books, IA doesn’t let you do that. CDL is very clearly not a substitute for the commercial product because it failed to substitute for the commercial product!

Great for Libraries🔗

But when it comes to what CDL is for — lending-library access to knowledge for the public, not book piracy (or enabling a hoarder) — it does its job beautifully, and everyone who matters loves it.

I’ve already talked about the powerful testimonials from COSLA, but there are many more.

The Board of Supervisors of San Francisco (the Internet Archive’s hometown) passed an incredibly strong resolution in support of controlled digital lending, the Internet Archive, and general power for libraries to operate independently of publishing companies:

FILE NO. 230418 Resolution recognizing the irreplaceable public value of libraries, including online libraries like the Internet Archive, and the essential rights of all libraries to own, preserve, and lend both digital and print books to the residents of San Francisco and the wider public; supporting the Internet Archive and its public service mission; and urging the California State Legislature and the United States Congress to support digital rights for libraries, including controlled digital lending and the option for libraries to own their digital collections.

WHEREAS, The Internet Archive provides digital books through its controlled digital lending program where a library owns a book, scans it digitally, and loans the digital copy one user at a time; and
WHEREAS, Authors, researchers, journalists, and other creators benefit from digital access to library collections, including the Internet Archive’s books, because libraries keep books available beyond their commercial life and provide access to older, difficult to find materials that provide the building blocks for new creative works; and
WHEREAS, The Internet Archive provides access to and preserves books that the public might not otherwise be able to access, either due to book bans, physical obstacles, or geographic or licensing unavailability; and
WHEREAS, The Internet Archive contains many books that have been banned from libraries across the country, and attempts to ban books has nearly doubled across the nation, reaching the highest point ever recorded in 2022; and
WHEREAS, Digital libraries are necessary to protect democratized access to information, knowledge, and truth; and
WHEREAS, The Internet Archive has a critical role in providing open, nondiscriminatory access to books, music, videos, and the history of the web; now, therefore, be it
RESOLVED, That the Board of Supervisors of the City and County of San Francisco recognizes the irreplaceable public value of libraries, including online libraries like the Internet Archive, and the essential rights of all libraries to own, preserve, and lend both digital and print books to the residents of San Francisco and the wider public; and, be it
FURTHER RESOLVED, That the Board of Supervisors commits to supporting the Internet Archive and its public service mission; and, be it
FURTHER RESOLVED, That the Board of Supervisors calls on the Legislature of the State of California and the United States Congress to support digital rights for libraries, including controlled digital lending and the option for libraries to own their digital collections; and, be it
FURTHER RESOLVED, That the Clerk of the Board send a copy of this Resolution to Senator Scott Wiener, Assembly Member Matt Haney, Assembly Member Phil Ting, Senator Alex Padilla, Senator Diane Feinstein and Speaker Emerita Nancy Pelosi.

(I’m sure Feinstein got right on that.)

Brewster Kahle and supporters at SF city hall Brewster Kahle and supporters at SF city hall

EFF Press Release, “Internet Archive Seeks Summary Judgment in Federal Lawsuit Filed By Publishing Companies” “Should we stop libraries from owning and lending books? No,” said Brewster Kahle, the Internet Archive’s founder and digital librarian. “We need libraries to be independent and strong, now more than ever, in a time of misinformation and challenges to democracy. That’s why we are defending the rights of libraries to serve our patrons where they are, online.”

Ebooks are broken🔗

But here’s a question: what’s the point?

We live in a world where, for most books still under copyright, specially designed ebooks already exist. Public domain books can be scanned and shared freely, without the logistical headache of controlling lent copies. And for copyrighted books, libraries should be buying ebooks from publishers and lending out their collection to patrons.3

To a librarian, digitizing existing books might sound like a way to make them more useful, but the scanning process is an enormous time commitment, and they’d have to keep the book off the shelves whenever it was checked out online.

So what could possibly make investing so much time and resources into this system worth it to so many libraries? Why are so many resources being thrown at a problem that looks like it’s been solved? What’s broken?

Well, it’s ebooks. The sensible practice — buying ebooks and lending them out just like you do with regular books — isn’t an option. And it’s not a technical issue, it’s because publishers have decided to be bastards.

Media Ownership🔗

In the US, libraries don’t have to “license” the physical books they loan out, and they never have. When it comes to library loaning, a book is a book is a book. A library can buy a book from a publisher, or a bookstore, or a used book shop, or they can even get a used book as a donation. It doesn’t matter; if it’s their book they own outright, they can loan it out one-patron-at-a-time, just like any book owner can do with any book. There’s no special permission needed from the publisher or author or printer or paper company or the maker of the font; the only rights you need to loan a book come with — and are inseparable from — owning the book. That’s what it means to own something.

People have a right to resell, trade, give away, or lend their own property — regardless of the wishes of the copyright holder — as per the principle of the exhaustion of rights, aka the first sale doctrine. I’ve already discussed first sale in this series, but to summarize: once a copy has first been sold, that item is like any other piece of physical property. The copyright holder’s right to control distribution is exhausted after the first sale. As I’ve described and as §109 makes clear, if you own a copy of media, the copyright holder is irrelevant to future distribution. You still can’t make new reproductions — that specific right is retained by the copyright holder — but if you own it, you can lend it.

In the words of the law, quoting directly from 17 U.S. Code § 109 - Limitations on exclusive rights, “the owner of a particular [copy] lawfully made under this title, or any person authorized by such owner, is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that [copy].” So in order to lend books without needing any special license from the copyright owner, libraries need to own the books outright, not simply have access to licensed copies. The trade-off for this is well worth it: once one owns a copy, they do not need special license to sell, donate, or lend it. Under these special conditions, the copyright holder is made irrelevant to the transaction; in fact, it’s inappropriate for them to be involved at all.

The first sale doctrine — the notion that you own the things you buy — is utterly foundational to virtually everything in a capitalist society, in a way that’s difficult to overstate. It’s foundational to the good kind of copyright: remember, the exhaustion of rights is not an exception to copyright, it is a foundational part of it.

But first sale is also foundational to the good parts of capitalism: people can own capital and have meaningful rights over their own property, instead of being subject to the discretionary authority of a tyrannical government or a feudal lord. Feudalism is, quite literally, the thing that makes capitalism look good in comparison, and to advocate against the exhaustion of rights is to advocate for feudalism.

Owning a book means you have rights to use that book by right of ownership, and which aren’t reserved by the corporation that manufactured it. Corporations automatically object to this because it’s theoretical control they could keep and use to squeeze more money out of people. To use Internet Archive founder Brewster Khale’s phrase, what publishers want is ultimate control over every reading event: who’s allowed to read, when they’re allowed to read, and if they’re allowed to read. This is a level of control they don’t, can’t, mustn’t have. But it’s what they want and why they do what they do.

That sounds like a bold claim, but it’s not hard to spot. Audaciously, publishers have tried to force through legislation to sabotage libraries by charging fees on books they already own. Like H.R.4850, to “compensate authors for uses of their books by public libraries, college and university libraries, and similar institutions” and S.658 to establish “a compensation system of payments… to authors for the lending of their works” were both such attempts, and both were summarily rejected.

In fact, the codification of the first sale doctrine itself in law was prompted by Bobbs-Merrill Co. v. Straus, in 1908, when a book publisher tried this same kind of power grab. The Bobbs-Merrill company (now Simon & Schuster LLC) attempted to abuse copyright to sabotage the used book market by printing this demand on their books:

The price of this book at retail is $1 net. No dealer is licensed to sell it at a less price, and a sale at a less price will be treated as an infringement of the copyright.

The Supreme Court found that, indeed, “one who has sold a copyrighted article, without restriction, has parted with all right to control the sale of it. The purchaser of a book … may sell it again…”. There’s nothing novel or thought-provoking about attempts to sabotage the secondary market; it’s a tale as old as greed.

While not novel, all such attempts by publishers to attack the concept of consumer-owned property are wrong, and have (mostly) been successfully struck down by the adults in the room. Anyone, including libraries, has the right to gift, lend, or dispose of books they own, even if publishers wish they could charge them for the privilege. Greedy publishers lost that battle and must always lose that battle.

But digital media has introduced myriad ways for corporations to withhold ownership rights from customers, not because the law entitles them to do so, but through technical measures they impose themselves. This is through the insidious rot of the DMCA “anti-tamper” provision, which makes it illegal for customers to modify their property to remove a restriction, even if they’re being restrained from doing something they have a legal right to do. This lets “DRM” create automatically policed business models out of whole cloth by corporations simply imagining how customers should have to behave, and makes “contempt of business model” a felony offense in the digital world. In effect, we’ve regressed all the way back to 1908, except without the sound court decision; the adults in the room are gone, the systems are compromised, and any absurd, arbitrary demand made about a digital work is automatically legally binding.

This was a well-known problem with the text of the DMCA. In fact, its passing was conditional on a follow-up report to study the effects on property rights and adjust the law as necessary. The report showed significant issues, specifically with how libraries depend on first sale, but publishers fought so hard against the necessary policy changes that congress ultimately neglected their duty — the very condition the DMCA was predicated on — and did nothing about it.

A surprisingly relevant example of this is region coding. Due to [out-of-scope economics involving how much people can afford to pay for entertainment], media companies can make marginally more money by selling physical media (games, blu-rays, etc.) at different prices in different regions. For example, Starfield on Steam is $69.99 in the US, but ~$59 if you live in Brazil.5 Even though the product — a digital game — is exactly the same, and digital distribution means there is no associated discrepancy in costs for manufacturing or distribution. This is a business practice known as regional pricing or price discrimination, with the ultimate goal being to skip “market pricing” and instead charge everyone the maximum price they would personally be willing to pay, like a hell auction4.

In a “free market”, regional pricing should be a non-starter because the first sale doctrine gives a secondary used media market influence over the maximum sale price of a work. For example, if companies sold DVDs for significantly less in Brazil than the US, a company could purchase copies there and re-sell them in the US in bulk, undercutting the US MSRP but still making a profit. This is categorically legal: you’re selling individual, purchased copies of the media, items you own outright. What the corporations use to prevent this practice isn’t the law, but region-coding: DRM baked into the discs and players that prevent people from playing imported media.

Anti-tamper laws let corporations squeeze more money out of people by unilaterally creating trade embargoes and artificially segregating the market. Digital dissemination, whether through transmission-streaming or DRM-locked copies, dramatically empowers a copyright holder to price discriminate by obliterating the property quality of the media.

But I’ve talked about this subject extensively, so I’ll try not to repeat myself too much here. For this essay, let’s focus on books.

When it comes to books, publishers can’t put DRM between paper and eyes — as much as they’d like to — but they sure can DRM themselves up some ebooks!

Feudalism on Overdrive🔗

Given the enormous potential of exploitative profit, it should not be a surprise to anyone that publishers are eager to shift to a rentier system, given an opportunity to do so.

For publishers, the value of ebooks is not in the incredible, automatic benefits digital formats give them to distribution, accessibility, and preservation. For them, the value comes from the excuse to shift to an unjust economic model that enriches themselves at the expense of the public, using the technological difference as a justification.

Publishers have built technical restrictions into ebooks that prevent broad categories of “unauthorized uses”. The use of DRM and technical protection measures lets publishers circumvent property law and the due process of evaluating contract enforceability, and instead, jump directly to having arbitrary, discretionary power. This has allowed them to invent their long-coveted “lending right”, which they can reserve to themselves and refuse to sell to consumers, or charge libraries (read: taxpayers) astronomical prices for.

Drop in a token, look at a book.

Right now, the single, monopolist provider of library ebooks is Overdrive6. Library patrons make an account using the library credentials and can check out ebooks to temporarily borrow through the Overdrive app.

For users it can be alright, but for libraries it’s a nightmare. Unlike with physical books, libraries cannot purchase ebooks to own outright and lend out — like they do physical books and media — because the publishers won’t allow it. They refuse to sell books to libraries outright, and they’ve completely prevented the existence of a secondary market. So instead, libraries are forced to go through Overdrive and “license” virtual copies — which the library never even handles! — for their constituents to have access to.

Abusive Licensing Terms🔗

Ebook licensing for libraries is a predatory model designed to siphon public funding away from library services and force scarcity of books and knowledge, directly contrary to the purpose of libraries.

The licensing strategies vary, but reliably set prices so exorbitantly high libraries can only afford to license a very limited selection. To use the example from American Library Association’s congressional brief (and note that this is in the context of commercial ebooks as lent by Overdrive):

Competition in Digital Markets Abusive pricing for libraries also is typical from the Big 5 publishers. For example, The Codebreakers by David Kahn and published by Simon & Schuster was quoted for $59.99 as an eBook for a consumer purchase — which means lifetime access. By contrast, the price to libraries for the very same eBook is $239.99 — and this is for one copy (i.e., it can be loaned out to one person at a time, simulating the print loan model) and lasts for only two years. If a library wanted access for four years, it would pay $479.98. If the library wanted access for 20 years, it would pay a staggering $2,399.90—for one copy, lending that eBook to one person at a time.

As another example, All the Light We Cannot See: A Novel by Anthony Doerr, is priced as an eBook for $12.99 to consumers. The library price is $51.99—for two years or $519.90 for 20 years—for one copy.

Similar evidence can be found in Readers First’s Position Paper (although they’re being far more charitable to publishers than I am):

eLending Position Paper — Readers First To understand exactly how the various models developed, I pulled a set of purchase order detail reports spanning the last ten years from OverDrive Insights. I included at least one purchase order from each month in the time period and the final set totaled 11,357 purchases of titles.

In brief, the average price per copy has tripled in nine years at the same time that license models have become much more restrictive. In 2011, the average price per copy from the major publishers was $13.37, offered with a perpetual license from most publishers. In 2020, no major publishers offer a perpetual license, and the average cost per copy is $39.96.

Roughly the timeline is as follows:

  • 2011: Most major publishers were offering eBooks to libraries for sale at approximately retail prices on a perpetual license (although some did not make their full catalog available at that time). …
  • 2013: Prices rose to a multiple of the retail price, which aligns with the proposed model that a perpetual use license should cost approximately 4 times retail. …
  • 2018: Publishers switched license models from perpetual use to metered access but did not lower their prices. A combination of metered access at prices at 2–3 times the price of retail is unfair to libraries.

It doesn’t come as surprise to anyone familiar with monopolized markets. Although the issue of price hikes has already hit us, people have been warning against this danger for decades.

Quoting from The First Sale Doctrine in the Era of Digital Networks, published over two full decades ago in 2003:

Reese, R. A. (2003). The First Sale Doctrine in the Era of Digital Networks …for consumers who want to acquire their own copies of a copyrighted work, a shift to digital transmission might result in generally higher prices than are charged today in markets in which works are distributed in copies and the first sale doctrine allows those copies to circulate freely. Consumers may have good reasons for wanting to own a copy rather than to acquire only limited access to copyrighted works. If a consumer is certain to access a work repeatedly, it may be cheaper to pay once for a copy of the work and obtain the right and ability to use the work as much as desired, rather than to pay a per-use charge or an ongoing monthly subscription fee. Buying a copy also offers certainty as to price and availability. A consumer who buys a copy knows up front the price she must pay for unlimited access. A consumer who pays for each use of the same work, or who pays on a monthly basis, must take the risk that the copyright owner will raise the price of access, or reduce the availability of the work.

A library might, for example, contract with the provider of an online encyclopedia or dictionary, such that anyone with borrowing privileges from that library would be able, from any computer, to access the work. It is unclear as a general matter whether the cost to a library of providing such online access would be higher or lower than the cost of purchasing copies to lend to its patrons. The cost per patron use of online access might easily exceed the cost of a copy of the work, spread across the number of patron uses of that copy. On the other hand, the cost per use might be no more expensive than the cost of buying a copy, or might actually be cheaper. Much would depend on the prices charged, how those prices compare with the price of copies, whether the price is charged as a periodic subscription or on a per-use basis, what restrictions are placed on the use of the work by the library and its patrons, and how heavily a work is used. In addition, libraries that provide access in this manner, rather than buying copies to lend, could remain vulnerable to a copyright owner’s pricing decisions, particularly if libraries contract with copyright owners for relatively short terms. Whenever such a contract is up for renewal, a copyright owner might raise the price to the library to a level that the library cannot afford. If the library therefore chooses not to renew, access to the works could become entirely unavailable to the library’s patrons. Had the library bought copies of the works instead of paying a recurring fee for mere access, it would, of course, be able to circulate those copies regardless of price hikes by the owner for more copies or for new works.

The publishers figured this out too. But they ran with it, to the point of refusing to sell books outright at all. They consciously rejected the traditional model of libraries purchasing and lending media in favor of a new rentier system that empowered them to siphon money from libraries indefinitely, just for access to their “intellectual property”. That’s why they intentionally denied libraries the ability to purchase copies of works, to cut off the one relief. The only option left is leasing, the option by which publishers can extort the most economic rent.

With Overdrive, libraries are forced to become renters, mere tenants of publishers instead of individual actors in their own right. Often ebooks are priced yearly as above, but many publishers double-dip, like HarperCollins’s “you lose your ebooks after one year or 26 checkouts, whichever comes first” model. “Perpetual licenses” that allow a library to pay once for a book often aren’t even offered at all, and when they are the prices are prohibitively expensive. In order to continue effectively performing their mission at all, they’re forced to sacrifice the fundamental concept of owning a collection of books, and instead subject themselves to the moment-to-moment whims of publishers eager to maximize revenue generation.

Ebook Defense Bills🔗

The state of ebook pricing is so bad that state governments are already trying to pass defensive laws to protect libraries from further exploitation. A highlight of these is Maryland’s H.B.518, which codifies library exploitation under the definition of “UNFAIR, ABUSIVE, OR DECEPTIVE TRADE PRACTICES”:

H.B.518 requiring a publisher who offers to license an electronic literary product to the public to also offer to license the electronic literary product to public libraries in the State on reasonable terms that would enable public libraries to provide library users with access to the electronic literary product

An overwhelmingly popular bill, HB518 passed unanimously in both the house (130 to 0) and the senate (47-0).

You’ll never guess the one faction who isn’t happy about it. Publishers sued the state of Maryland over this bill, taking offense at the requirement to be, quoting the bill, “reasonable.” So, if you’re ever tempted to think publishers don’t view libraries as their enemy, that’s the kicker.

(Specifically, they claimed that regulating trade practices infringed on their copyright, which is ludicrous, but also an example of the problems with the intellectual property mentality. These laws are about commerce, not copyright.)

Note that this bill doesn’t represent the state trying to renegotiate any of its existing contracts, but is rather a defensive measure against projected abuse.

Matt Enis, AAP Sues Maryland Over Law Requiring Publishers to License Ebooks to Libraries Under “Reasonable Terms” [Michael Blackwell, director of the St. Mary’s County Public Library, MD, and an organizer of the ReadersFirst coalition] said that, in his opinion, publishers that currently license ebooks to libraries are already compliant with the “reasonable terms” standard, even though many librarians may be unhappy with pricing, circulation caps, or other aspects of ebook licensing.

“This doesn’t really go as far as a lot of people in the library community would like,” he said. “I think it’s fairly modest. The state has a legitimate interest in guaranteeing that its citizens can read ebooks that are otherwise [for sale] to the public.”

Blackwell added: “We in libraries look forward to working with any publisher on mutually beneficial agreements for the good of all readers and learners. Our bill is not anti-publisher…. It is reasonable. Existing licenses under which we have been working will not need revisions.”

Maryland isn’t alone in trying to defend its libraries against abusive publishers though. I recommend reading the excellent summary at EveryLibrary Institute for more on this.

Why Does Overdrive Exist at All?🔗

If publishers think that libraries are diametrically opposed to their revenue-generation goals, why offer them ebooks at all? Why pour so much time and energy into an ecosystem while they simultaneously insist that their “current analysis on eLending indicates it is having a direct and adverse impact on retail eBook sales”? I think the answer is fairly obvious: it’s an opportunity to lock libraries into a new, extractive system.

The ways in which publishing companies have refused to cooperate with lending libraries can be understood fully as part of the much larger political war on individual ownership of property. We’re currently in an age where companies are leveraging new technology in order to recreate a feudalist system where rent is extracted perpetually from users who own nothing themselves, while corporations retain the real control. Corporations want everyone to be forced to license their property from corporate owners, who can revoke those licenses or set conditions (locked to a specific brand of devices, etc.), and even demand recurring fees on products that don’t need them.

To publishers, the great value of normalizing the Overdrive model is not just in sapping revenue from libraries, it’s in eroding the concept of consumer ownership, which in the era of digital technology is a barrier to corporate profit. They’re already mad they can’t put a chokepoint on physical media, and they see an opportunity to chokepointify books. We know they’re against lending because the ebooks they sell through Amazon and similar already have technical controls preventing it. Tradable property has only ever been something they’ve tolerated, and now that they have the tools they’re finally working on eliminating it.

So with publishers already publicly professing they’re unwilling to commit to “reasonable” licensing terms, why would we trust them with monopoly power to define licensing terms for library books? They’re already actively trying to punish libraries, which they resent as competitors.

Monopolists always insist they can manage the entire market single-handedly, like tzars of a command economy. They insist they’re managing in a way that benefits everyone involved, not just themselves. It’s never true, but they always say it. But for books we already know we’re not getting that behavior. The harms aren’t projected; they’re already inflicted, and the attacks are only ramping up in scale.

Overdrive’s book licensing monopoly is yet another case of a business model being elevated to the position of a law. Corporations use their copyright-granted production monopoly right to force libraries into a constructed, extralegal exploitation system where rental is the only option available. But they don’t have a right to that system, and if libraries can find a legal way out, publishers should have to go pound sand.

This is why controlled digital lending is such a desperately needed relief for libraries. CDL threatens to give libraries a defense against this exploitation by creating a new option for libraries, but instead of calling feudalism a loss and trying to find a way to create value, publishers leapt at their throats like wild dogs.

CDL as Alternative🔗

But what is the nature of the “relief” controlled digital lending provides libraries. Moreover, what does Overdrive have to do with CDL? I’ve kept saying that CDL is concerned with print books, not first-class ebooks, and that’s still true. But is CDL an escape valve for the predatory practices seen in Overdrive? Are CDL copies an “alternative to” ebooks? The answer is yes and no.

CDL loans are an “alternative” to ebooks in some cases because they fill some of the same needs, not because they have all the same features. It’s like calling water an “alternative” to soda: they both hydrate you, but the only way water offers something soda doesn’t is if there’s a terrible problem with your supply chain and your basic needs aren’t being met. CDL books “scratch the same itch” as ebooks but only because there was a giant gaping hole where the remedy for “no, I don’t need the luxury product, I just need the information for research” was supposed to be.

Of course, lots of things are “alternatives” to licensing an ebook from a publisher, depending on your needs. You could borrow a physical copy from a local library or purchase a physical copy outright. Or you might just make up the information and go ignorant. Yes, ultimately using a book you already purchased and own is an “alternative” to buying a new ebook copy. But then, that’s an argument that recurses forever, isn’t it? Never use, only pay.

But a general defensiveness against alternatives is unwarranted. While a sufficiently self-agitated corporation might muster up the energy to take offense at alternatives to their product existing, they do not have a general right to “not be competed with”, even if they hold a copyright. There are only specific rights over specific works, all of which are being honored. Publishers take offense at the threat to their money and backfill that emotion with post-hoc justification.

So, yes, CDL threatens to disrupt this system by providing book access to people who just need a temporary copy of a book they can read digitally without the library having to negotiate with Overdrive.

Normally the library doesn’t have to “negotiate” with anyone: a status quo Overdrive disrupted but CDL restores. The Overdrive system is this outlier that’s held up by restrictions that shouldn’t exist in the first place: of course libraries’ response to a burdensome, exploitative trap is to work around it, as it should be! Abusive monopolies are made to be disrupted; it takes violence to prevent that.

But CDL isn’t just a cludgy workaround to avoid paying for a service, it’s a genuinely valuable solution, leveraging new technologies and existing rights to create a new, useful option.

Anti-CDL Propaganda🔗

So why am I having to be so defensive about the need for CDL and its legitimacy as a practice? Because publishing corporations hate lending with a venom, and need you to too. Not just controlled digital lending specifically, but lending and individual ownership of property as a general practice.

As is very often the case with these issues, there’s a small population of people who are against CDL for their own personal gain and another much larger population that’s been hornswoggled by the first.

In order to understand how publishing companies go about orchestrating an outrage campaign against libraries, we have to start with The Authors Guild7.

“The Authors Guild”, is a political advocacy organization for published authors and journalists focused on strengthening the IP rights of authors through lobbying and legal action. It has been a power player behind a number of copyright lawsuits, including major cases against Google Books and the Internet Archive.

The theory is that a “strong” copyright regime increases the value of the work of authors. In practice though, all the copyrights created by authors get vacuumed up by the publishing companies8, namely the Association of American Publishers (AAP). So in its fights to strengthen copyright protections, TAG ultimately works for them first, even when that means putting authors second.

And so, because CDL cuts off a source of enrichment for publishing companies, of course The Authors Guild is blue in the face with rage over it. But they’re wrong to be, and so their complaints turn out wrong too.

In describing its complaint with the Internet Archive, The Authors Guild starts things out by calling Open Library larger than “the most flagrantly illegal pirate websites.” This is an obvious lie: not only is Open Library’s 1.4  m book catalog smaller than Libgen’s 6 m + 84 m, or Z-library’s 22 m, but The Authors Guild knows they’re lying here, because they make the Z-library comparison themselves later in the same brief. This outburst is not only objectively false, it’s a schoolchild-tantrum-level “you’re worse than Hitler!” outburst that, right out of the gate, utterly delegitimizes TAG as a serious participant in the conversation. But so does everything else they do, so let’s keep going.

The Authors Guild Newspost “Internet Archive’s wholesale scanning and posting of copyrighted books without the consent of authors, and without paying a dime, is piracy hidden behind a sanctimonious veil of progressivism,” said Douglas Preston, author and President of the Authors Guild. “The Internet Archive hopes to fool the public by calling its piracy website a ‘library’; but there’s a more accurate term for taking what you don’t own: ‘stealing.’”

Preston continued, “What Internet Archive is doing is no different than heaving a brick through a grocery store window and handing out the food—and then congratulating itself for providing a public service. It’s not a public service to violate the rights of thousands of hard-working authors, most of whom desperately need the income. We authors want everyone to have access to books, but there are already thousands of excellent public libraries in the United States devoted to providing free access to e-books. Legitimate libraries pay for those e-books, and a portion of that flows back to authors as royalties, helping ensure they can continue to write.”

First off, the only source they have to quote to support their idea is themselves, which is very funny. And they’re not framing it as repeating themselves, they’re framing it as an appeal to authority, except the authority is Douglas Preston, both the president of the organization and also responsible for this post, citing himself.

Second, everything else is wrong. IA is a library by every definition, including its function but also its legal status: IA is fully accredited as a library with the state and federal government (which, you might remember, love it.) Douglas Preston saying IA isn’t a library here is just lying. And it’s the same with every other word that comes out of his mouth: IA isn’t stealing books, it bought them with money it paid to publishers. Authors’ rights aren’t being violated, their books are being purchased. Existing public library access to ebooks isn’t sufficient, it’s in terrible peril.

Possibly the most backwards thing here though is the “heaving a brick through a grocery store window and handing out the food—and then congratulating itself for providing a public service” metaphor because it doesn’t describe the internet archive (or even what they’re accusing them of, it’s a very ill-fitting metaphor), it describes what publishers are doing.

Dougie is accusing the Internet Archive of committing the broken window fallacy: claiming that causing damage is a net benefit to society because of some virtues found in recovering from it. This is not what IA is doing, but it is a reasonable metaphor for what publishers are doing: refusing to sell libraries books outright, thus causing a shortage, and then calling your willingness to rent them ebook access economically valuable. I’d have gone with developing a cure for a disease but refusing to sell it so you can continue to sell pain relief, but the broken window metaphor fits nicely too. It doesn’t fit IA, of course, just the publishers. But, as metaphors go, the broken window fallacy is actually a much better fit for… well, we’ll get to that.


So why is the TAG/AAP coalition trying to manufacture this us-vs-them tribal conflict of authors versus libraries, anyway? It doesn’t reflect reality: authors want their books in libraries. So why are publishers so eager to have “libraries” around as a subject to blame? Well, according to TAG’s own research, authors’ livelihoods are9 indeed under dire threat:

The Authors Guild Survey Shows Drastic 42 Percent Decline in Authors Earnings in Last Decade The Authors Guild’s 2018 Author Income Survey, the largest survey of writing-related earnings by American authors ever conducted finds incomes falling to historic lows to a median of $6,080 in 2017, down 42 percent from 2009.

The Authors Guild calls this a “crisis of epic proportions for American authors” that threatens their ability to make even a living wage. And, frankly, yes: a 42% cut in wages is a catastrophe. By my estimate, though, The Authors Guild has actually severely under-counted the damage suffered by authors by not properly adjusting for inflation. Adjusted for inflation10, that survey means authors are making just under 51% of what they did in 2009. For the authors affected, this is nightmarish!

The Authors Guild’s explanation for this can be found under “The Causes”. It’s shockingly bad, but before I get into what it says, it’s probably more important to look at the one factor this “analysis” conspicuously glosses over entirely: publishing companies. Yes, really: this “study” refuses to even consider the possibility that the reason authors aren’t paid enough anymore is that the publishers aren’t paying them enough anymore.

There are several lines blaming Amazon, saying it’s able to “lock publishers into a vise, relentlessly demanding increasing discounts and narrowing margins” which forces “publishers to pass on [their] losses to authors”. But other than this implication that publishing companies are the ones being bullied by big tech, there’s really no mention of the overall health of publishing companies and their role in aggregate price-setting here. And book sales aren’t declining, as The Authors Guild points out itself — in fact, they’re far exceeding expectations11. With these conclusions, you’d have to assume the publishing companies are barely solvent, or else they wouldn’t have stopped paying their authors. So what’s behind that curtain?

Estimated net revenue of the book publishing industry in the United States from 2008 to 2022
(in billion U.S. dollars)

Estimated net revenue of the book publishing industry in the United States from 2008 to 2022 Statista


In stark contrast to the hardships they’re supposedly “passing on” to authors, there is no 42%12 chunk missing from publishing company revenues. With regard to costs, the popularity of ebooks means less material costs, so except for a brief trough during pandemic shortages, costs have dipped down, not spiked. So, with the exception of a slight dip 2016-2020, publishers’ revenue has held steady, and 2021-2022 is back up to record profits of 28-29 billion dollars: higher than anything previous in all recorded data. More on this later. Publishing companies haven’t had their revenue slashed in half, they’ve just slashed the pay of their authors and pocketed it.

And it’s too late to argue that it’s okay for the publishing industry to be strong while authors suffer, because the whole survey argues the opposite! Publishing companies could make all sorts of arguments that this is fine — there are whole categories of rhetoric available to this effect — except it’s too late for them to do that because they already published a report about what a disaster it is.

So if authors are going broke because the publishers are paying them a pittance, and the publishers aren’t losing any money at all (in fact, they’re making record profits), where’d the missing money go? …Well, according to TAG, greedy book-owners and libraries!

Yes, in one of the most shameless acts in this whole saga, on the end of this very study, right after they slap “Amazon sells used books” under Causes, The Authors Guild recommends

  • Royalties should be paid by resellers to authors for resellers’ sales of new books.
  • U.S. should establish a federally funded equivalent of a public lending right13 to provide authors a benefit from the public use of books; and libraries should be better funded.
  • Publishers should pay higher royalties on ebooks and deeply discounted books; and they should destroy all bookstore returns to prevent them from getting into the secondary market.

Remember the broken window fallacy?

Just a few paragraphs ago, Douglas “my source is Douglas Preston” Preston accused the Internet Archive of “heaving a brick through a grocery store window and handing out the food.” Meanwhile, their actual, take-us-seriously-please policy proposal is to burn all the food so they can sell more.

Their honest-to-god policy proposal, that they wrote up and published as if they were respectable people, is to require used books to be destroyed, brazen broken window fallacy and all. They really won’t pay authors a living wage, and their answer to that problem is “we should do a Fahrenheit 451 on it”, expecting you to believe that their accumulating more money will make them change their mind and want to pay authors again!

This is the instinct the attack on controlled digital lending comes from, and it is perverse. Attacks on CDL are born from this very same outrageous eagerness to destroy private property in order to create demand for more sales. As shameless, as outrageous as it is, publishers have embraced the ghoulish notion that destroying private property people purchased from them in order to destroy value and create demand is an effective business strategy. Attacking CDL diminishes the value of private property by preventing forms of lending: you bought it, but now it’s not usable for as much by simple fiat.

When publishers demand libraries destroy their CDL collections, it’s for the same reason they want bookstores to destroy their stock. It’s window-breaking. Publishers aren’t simply content to leverage their gatekeeper position to extract ever-increasing profits from customers without providing additional value: they’re so dedicated to that model that they have to actively seek out value that exists — even in the form of books they’ve already sold people — and destroy it! The penchant for wanton destruction is not a drive to protect the rights of authors, but a willingness to cause any amount of destruction, even to their own industry, if it seems like there’s money in it for them. It’s cancerous.

In writing this essay, I’m trying to avoid painting a cartoonish caricature of a conflict between noble, archivist librarians and book-burning robber baron publishers. But, as the facts stand, one of the sides is that absurd extreme, and so I’m forced to call it honestly, even if it means superlative language. If you just sit down and look at what the anti-library faction is actually saying, they’re climbing over each other like pigs in mud to make arguments about why private book collections should be illegal and how we need to be destroying whatever books aren’t generating them new profits! Every argument is just “this might make me some money”!

Discord: The Authors Guild Against Authors🔗

Unsurprisingly, The Authors Guild finds that their primary opponent in pushing their anti-book, anti-bookowner, anti-library policies are the very authors they’re supposed to represent. I talked earlier about how The Authors Guild aligning themselves with publishers means they’re no longer representing their own members, and here’s the balance due.

The best summary of the vast support authors themselves have shown for CDL is likely Fight For The Future’s “Authors For Libraries” open letter, signed by more than 1000 authors. I’ll include the full body because it’s excellent:

We urge all who are engaged in the work of getting books into the hands of readers to act in the interests of all authors, including the long-marginalized, midlist, and emerging authors whom librarians have championed for decades. We write to ask that all publishers, distributors, and trade associations:

  1. Enshrine the right of libraries to permanently own and preserve books, and to purchase these permanent copies on reasonable terms, regardless of format. Many libraries would prefer to own and preserve digital editions, as they have always done with print books, but these days publishers rarely offer them the option. Instead, when libraries have access to ebooks at all, the prices libraries pay to rent ebooks are often likened to extortion.

    Digital editions are more affordable to produce and often more accessible, but libraries are already relying on emergency funds and may only be able to license a small selection of mainstream works in the future. In turn, readers will have fewer opportunities to discover the more diverse potential bestsellers of tomorrow.

    It is past time to determine a path forward that is fair to both libraries and authors—including a perpetual model for digital ownership based on the cost to maintain a print edition.

  2. End lawsuits aimed at intimidating libraries and diminishing their role in society. The interests of libraries are the interests of the public, and of any author concerned with equity and longevity for themselves and their fellow writers. We are all on the same side. Yet a unanimously passed Maryland state law ensuring libraries pay “reasonable fees” for digital editions died after the AAP sued. And after a previous suit failed, several publishers are currently suing the Internet Archive Library in an attempt to prohibit all libraries from lending out scanned copies of books they own. While undermining libraries may financially benefit the wealthiest and most privileged authors and corporations in the short term, this behavior is utterly opposed to the interests of authors as a whole.

  3. End smear campaigns against librarians. Recent comments likening library advocates to “mouthpieces” for Big Tech are as tasteless as they are inaccurate. Also concerning are the awards recently given to legislators who have advocated in favor of the dangerous surveillance of library patrons, and of laws that criminalize librarians. As a last bastion of truth, privacy, and access to diverse voices, libraries’ digital operations grow ever more essential to our society—and their work should be celebrated, not censured.

We fear a future where libraries are reduced to a sort of Netflix or Spotify for books, from which publishers demand exorbitant licensing fees in perpetuity while unaccountable vendors force the spread of disinformation and hate for profit. Publishers must balance profits for the most prominent authors and shareholders with the right of the public to free, unsurveilled access to knowledge and information—as well as the right of emerging authors to be collected, preserved, and discovered.

“Anyone who tells you libraries and authors are on the opposite side of any issue has grossly misunderstood the nature of libraries, or authors or both,” Cory Doctorow, co-author of Chokepoint Capitalism, which explores the harms of big content creators, and signatory of the Fight for the Future letter, wrote in a statement accompanying the letter.

Traditional librarians love CDL too. Quoting from The Internet Archive Is a Library by Dave Hansen, Deborah Jakubs, Chris Bourg, Thomas Leonard, Jeff MacKie-Mason, Joseph A. Salem Jr., MacKenzie Smith and Winston Tabb (can you tell from that citation we’ve entered academia yet):

The Internet Archive, a nonprofit library in San Francisco, has grown into one of the most important cultural institutions of the modern age. …

Right now, we are at a pivotal stage in a copyright infringement lawsuit against the Internet Archive, still pending, brought by four of the biggest for-profit publishers in the world, who have been trying to shut down core programs of the archive since the start of the pandemic. For the sake of libraries and library users everywhere, let’s hope they don’t succeed.

The Internet Archive … is a one-of-a-kind independent research library, with its holdings fully available in digital form. Its substantial physical and digital collections are unique. It employs librarians and other information professionals. It is open to all interested readers. It cooperates with peer libraries in support of archiving the information and contemporary discourse as manifested in the World Wide Web. It has an active community of researchers who depend on its collections. And it is an engaged, responsive, resource-sharing partner to hundreds of peer libraries. It is also now an integral part of the interlibrary loan system, sharing its holdings with other libraries worldwide. It shares the keystone values of all libraries: preservation, access, privacy, intellectual freedom, diversity, lifelong learning and the public good. And it does all this without commercial motive as a mission-driven not-for-profit organization.

Those of us who have worked with the Internet Archive or drawn on its many offerings have long seen the organization as a peer. The Internet Archive fulfills the mission of a library in ways we could only dream of a few decades ago.

We cannot defend against the publishers’ lawsuit. We can, however, stand with Internet Archive as it fights for the right to buy, preserve and lend books, which is what libraries do.

So, that’s what authors think, but it’s clear that publishers don’t care. The publishers are actively working against authors’ interests, and TAG is blatantly disregarding its own mandate to help them. So we can definitely discount publishers’ claim to be generally acting on behalf of writers’ interests; we know the opposite is true.

The Authors Guild being so entwined with the Association of American Publishers is a serious red flag of a dysfunctional system. In the real world, authors’ and publishers’ goals aren’t aligned; they’re actively antagonistic. There is tension between authors (who want the money from book sales) and publishers (who want the money from book sales) that’s resolved with negotiation.

But the TAG/AAP partnership does not reflect this dynamic at all. For instance, in their literature, TAG complains about the rise of self-publishing with Amazon: something good for publishing authors, but bad for the entrenched publishing monopoly, who The Authors Guild sides with over, y’know, authors. It’s evidence of institutional corruption, where the guild is working for the enrichment of the guild itself rather than its members.

Discord: Authors Against CDL?🔗

But, admittedly, authors don’t uniformly show this overwhelming support for the Internet Archive. So what about the authors who do support TAG/AAP? When we look at the actual writers TAG/AAP drags out to support their positions, their stances are, in a word, telling. Categorically, their authors don’t have a grasp on any of these concepts, and are instead just whipped up into an emotional haze:

Sandra Cisneros When I went on the Internet Archive’s website and saw that scans of my books were being distributed to anybody who wanted them for free — without my permission or any payment — I was appalled. I found the experience so viscerally upsetting that I could not stay on the website for long. It was like I had gone to a pawn shop and seen my stolen possessions on sale.

To this day, I am angry that Internet Archive tells the world that it is a library and that, by bootlegging my books, it is simply doing what libraries have always done. Real libraries do not do what Internet Archive does. The libraries that raised me paid for their books, they never stole them. The libraries that raised me paid for their books, they never stole them. Any libraries that want to provide eBook versions of my books to the public for free can do so because I have authorized Penguin Random House to license my work to any library that is willing to pay for the authorized digital formats. I consider Internet Archive’s distribution of my books to be a terrible violation of the control I have worked so hard to establish over my work.

There is also no doubt in my mind that I lost money because Internet Archive has posted my work online. It is common sense that if someone is willing to download one of my books for free from the internet, they are not going to pay for it later. … I have lost money because Internet Archive has not paid me any of the fees that actual libraries pay to license my work and lend it to the public.

For the most part, this is just using emotional language to lampshade Sandra saying she’d like more stuff but doesn’t really care who it comes from. Sandra is wrong on the facts in ways I’ve already explained: the books weren’t stolen, she was paid for them, Internet Archive is a real library, the books can’t be downloaded, traditional libraries don’t pay licensing fees to lend books, etc. She’s even fundamentally wrong about libraries being able to loan ebooks, as I discussed already when describing Overdrive. It’s all wrong.

That’s why, in order to use this wrong-but-emotionally-compelling story in court, someone had to side-channel it through a testimonial declaration by a non-expert rather than pose it as a direct legal argument that would be subject to factual scrutiny. That’s how you get away with having someone just guess that they think something probably cost them money and submit that to the court as if someone being ignorant is in-and-of-itself evidence that they’re right.

But I don’t just find it interesting that she’s wrong, I find it interesting the emotional way she’s wrong, as shown by the language she uses here. This is language of trauma. Assuming these aren’t crocodile tears and Sandra isn’t putting on an affect in her language, Sandra has been lead to believe that the IA scans are a personal violation. Of course, IA didn’t steal the books; they bought them just like the other libraries did. But that doesn’t change that fact that she feels like she’s been violated. So why is that?

TAG/AAP is actively cultivating this trauma by flooding authors with misinformation and reinforcing error when it serves their financial interests. And now that the damage is done, TAG/AAP’s campaign against CDL is actively preventing any healing and is instead stoking that trauma like a flame so they can use it for profit. Sandra doesn’t understand what’s going on because she’s not here to understand what’s going on; she’s here to look hurt because publishers want it to look like CDL hurt somebody.

And of course this is spewed all across social media.

And that’s just a tiny sample; I think most of those conversations took place within the same week. There’s just been so much of it.

Reinforced Error in Public Discourse🔗

This is another side of the conflict that has a nuance that interests me: what we’re seeing isn’t quite astroturfing. For now I’m calling it reinforced error. It’s a technique where the propagandist can exploit an existing fallacy instead of having to manufacture consent from whole cloth.

It starts with a reasonable first-glance understanding. (In this case, it looks like “a site is distributing a commercial book for free, which is bad”). That conclusion is wrong, but it’s wrong for complex, nuanced reasons, and it’s certainly not an unreasonable idea. It just isn’t correct. But because the baseline conclusion can arise naturally, the misinformation campaign doesn’t require propagandists to manufacturer an entirely artificial narrative, it just requires them to stimulate a natural fallacy and discourage people who have reached the naïve conclusion from thinking through the problem further.

It’s the geocentrist fallacy. “The sun goes ‘round the earth.” How hard is it to convince a population that their first inclination was right when you know it wasn’t?

Naïve writers are being worked up into a frenzy against a manufactured enemy that intuitively seems to be harmful but isn’t, while the publishers who fan the flames of outrage against an innocent opponent scoop up all the money. It’s not a new strategy. It’s astroturfing, except the soil is already fertile and primed by the fact that, if you squint hard enough, libraries seem to be an opponent, even without the propaganda campaign.

The insidious thing is how close it gets. If you made all the assumptions publishers want you to, and didn’t know all the books had been purchased and lending was controlled, outrage would be justified. It’s just that it isn’t.

Library Permanence and Archival🔗

Libraries are counterexamples to the neofeudalist no-one-needs-to-own-anything narrative. While I would argue that it’s important for individuals to own books too, libraries are institutions that must own their materials outright. So as publishers try to force such a model on the book market, it makes complete sense that libraries would be one of the early cases that jams the gears and demonstrates that private ownership — not corporate control — is not just preferable but necessary.

Preservation and Archival is a Core Function of Libraries🔗

As part of their mission to provide educational resources to the public, an important part of libraries’ work is preservation and archival. The Internet Archive is a specialized research library that focuses on archival and preservation in particular, but ensuring materials remain accessible to the public is a main function of every library. This comes in many forms: libraries make it possible to access more information than you could purchase outright, but also ensures accessibility in other key ways, like distributing out-of-print books.

It seems trite to point out “accessibility” as a good to maximize with public policy, as if the only priority should be public distribution of work at the expense of the people who make it. But we can’t think about the question of copyright without considering value judgements because the value judgement to encourage the arts and sciences is baked into the foundations of copyright already! Copyright, as a doctrine, is exclusively designed to promote science and the creation and ultimate accessibility of information.

The Surprisingly Big Business of Library E-books | The New Yorker Books, like music and movies and TV shows, are increasingly something that libraries and readers do not own but, rather, access temporarily, from corporations that do.

The high prices of e-book rights could become untenable for libraries in the long run, according to several librarians and advocates I spoke to—libraries, venders, and publishers will probably need to negotiate a new way forward. “It’s not a good system,” Inouye said. “There needs to be some kind of change in the law, to reinstate public rights that we have for analog materials.” Maria Bustillos, a founding editor of the publishing coöperative Brick House, argued recently in The Nation that libraries should pay just once for each copy of an e-book. “The point of a library is to preserve, and in order to preserve, a library must own,” Bustillos wrote.

Again, this isn’t unique to libraries, but libraries are a prominent example of it. It is ownership of books — specifically, the kind of ownership of copyrighted media allowed by the exhaustion of rights — that generally contributes to preservation of knowledge by removing the publisher as a single-point-of-failure for literary culture at large.


Corporate ebook markets are insidiously precarious, designed so that publishers — who sometimes want what consumers consider disastrous failure — have access to kill switches.

True sales of media, like with physical books, create a naturally distributed system. Since every copy of the work is independently available, the accessibility of a work scales with how many copies exist in usable condition. And if any one copy is damaged, another can replace it. Simple enough.

Each copy of a book is also self-sufficient and doesn’t have any special dependencies it requires to work. If a book is out of print — because it isn’t as profitable to sell as other options available to the publishing company at the time, or because the rightsholder is no longer solvent at all — books in a library’s collection remain available as a research resource or simply for general consumption. These are all qualities that give ownership of a book genuine utility in its own right.

Streaming and DRM-based media ecosystems are the opposite: the systems are intentionally architected to maximize profit, which means they prioritize preventing piracy and unauthorized access at the cost of making the system intentionally fragile. For example, streaming-only books (like many textbooks) require an active connection to a web server that actively verifies the reader’s entitlements before letting them read their books.

This creates a series of tightly-chained architectural components that depend on each other: if at any point there’s a failure in the payment processor, the entitlement database, the user’s network connection, the app store, or the reader application, the book will be unavailable.

This is called a fail-closed system: if anything in the verification process fails, access is prohibited by default. The purpose of DRM is to limit the use of digital works, so any competent DRM must fail closed, or else it would be impotent. Since companies want to “verify ownership” with digital content above and beyond possession, anything that prevents that first-party verification intentionally locks the content. (More on this dynamic in a future essay.)

This kind of fail-closed system is good for strong security, but it’s completely antithetical to preserving availability and access. A true owner of an item — for instance, a library that owns its collection of books — cannot have their access to that item dependent on a third party. A library’s collection must not depend on the active consent of outside publishing companies, because, as an obvious example, it must be able to outlast the existence of that company. A book that cannot function without getting authorization from the server of a company that went defunct years ago is utterly worthless in an archive, except as a monument to the failures of capitalism.

This creates a perverse system where the economic incentive for publishers is to create a significantly worse product. Publishers believe that this DRM and artificial fragility is necessary in order to maximize profit, and so ebooks are designed to be intentionally fragile, even though that costs more to maintain than a simple, portable piece of media.

Even though mainstream ebooks are transmitted to the public and read by millions of consumers, since the law doesn’t acknowledge that those consumers were ever in possession of a copy of the book, they have no right to keep and read their own books, no matter how much they paid for it. Only the right of first sale does that, and with online-only or DRM-locked ebooks on the Kindle storefront or the like, licensing conditions explicitly deny customers that crucial right.

As far as preservation is concerned, this means all the power — and responsibility — is completely concentrated in a few entities, each of which represents a single point of failure. With DRM’d ebooks, “out-of-print” is just a switch in a database that removes books from the storefront but also removes your “license” to read them on your own device. Publishing companies, you’ll remember, don’t want out-of-print books to be available at all: they’re against libraries providing this function, but they’re also waging legal campaigns against used book stores. In their minds, if they’re not selling a book directly, it’s stealing their customers.

Publishers Cannot be Trusted🔗

One of the contingencies decentralization and the first sale doctrine addresses is the case where the copyright holder — for whatever reason — ceases to make work available themselves. It’s good to distribute the ability to preserve media among all the individual media owners, because consolidating it in one party — like the copyright holder — is a critical mistake.

The full duty and responsibility for preservation must not be held by a few corporations alone, because it’s a duty they’ve abdicated and a responsibility they can’t hold. I don’t mean this as an insult, I mean that they’ve very clearly chosen not to take on the responsibility, and they are, definitionally, unable to be responsible in the event of their own failure. A publishing company’s first responsibility is to its own profitability: this is the nature of the beast.

Publishers let works go out of print regularly. No news there. This is generally an “economically rational” decision: the publisher believes spending those same resources elsewhere, like new books, will make a greater profit. Importantly, the water level for “not economically rational to print” is not zero. There can still be significant demand for the work, but external factors alone can make publishers choose to discontinue it.

Likewise, there is no incentive for a corporation to take precautions for the event of its own failure. Not only is there no contingency plan for preserving works after a firm folds, a corporation actively loses value by having such a plan. Not only would executing such a plan not deliver any profit, tying your existence to the usability of your products actually incentivizes consumers and the government to keep you afloat, for the public good.

Like most IP rights, the enormous responsibility of ownership over single points of failure in the ebook ecosystem comes with no obligation of care. In the eyes of the law, you have the right to neglect or destroy your own property, even if other people want or need it. As copyright is erroneously treated like property, this mentality begins to apply to things like books. Corporations argue that they have every right to destroy media if they judge it profitable to do so, even though the purpose of copyright is to encourage publication.

We can’t expect publishers to preserve their own work, even if that work could actively be making them money, and we have remarkably good data demonstrating this. Works copyrighted before 1978 had their period of copyright protection divided into two terms, the second of which required an active renewal filed with the copyright office. Even though the renewal fee was nominal, only a small proportion of copyrighted works were renewed such that the protection period lasted the full two terms. Empirical analysis from Landes, William M. and Posner, Richard A., Indefinitely Renewable Copyright (August 1, 2002) shows the proportion of eligible works that were renewed was as low as 11% and only reaching as high as 22%.

Rightsholders were so uninterested in the sustainability of these works after the peak of their commercial value that they didn’t bother retaining the rights at all. What we see here is a rejection of the “long tail” model: publishers decided it wasn’t worth preserving these works, even in a form that could actively make them a profit, when they could instead invest time and energy into new hits.

Here’s what really kills me, though. Read these excerpts from Reese, R. A. (2003). The First Sale Doctrine in the Era of Digital Networks:

…the potential impact of digital distribution on the preservation effects of the first sale doctrine is very significant. In the world of copy distribution, a copyright owner’s economic decision not to expend resources to preserve a work might have had a relatively limited effect on the work’s survival, because many other people owned copies that they might choose to preserve (particularly if those other owners were libraries or archives). But if the digital-transmission copyright owner forgoes expenditures that help preserve the work, it is not clear that anyone else will be in a position to engage in such preservation.

A shift to digital dissemination also seems likely to threaten the first sale doctrine’s contributions to preservation. If a work is distributed only by transmission, then the only copies of the work will generally be copies stored on the copyright owner’s computer server, which transmits the work to the public over the network (as well as any other copies, such as printouts, kept by the copyright owner). Even if the work is seen or heard by millions of consumers, those consumers generally will not have a copy, just as radio listeners and TV viewers today do not generally have copies of the works they hear and see. This lack of widely distributed copies means that the copies maintained by the copyright owner must bear all of the risk of damage, loss, disappearance, or destruction.

In the past, preserving copyrighted material has generally involved mostly properly storing and conserving copies owned by the preserver—acts that, thanks to the first sale doctrine, did not generally run afoul of copyright owners’ rights. Preserving digital works will often require migrating those works to a more contemporary format, which will involve acts of reproduction (and perhaps adaptation) generally reserved to copyright owners.

Even libraries will not necessarily be able to exercise their reproduction privilege for preservation purposes effectively. Technological protection measures used by copyright owners will often prevent effective library copying for preservation purposes.

Courts might be especially cautious in characterizing a copyright owner’s transaction with a consumer as a license rather than a sale, giving careful scrutiny to the actual reality of the transaction rather then[sic] to any labels used by the copyright owner. If the consumer essentially obtains permanent dominion over the physical object that is the copy, the transaction should probably, absent compelling reasons to the contrary, be characterized as a sale, thus conferring on the consumer the rights of a copy “owner” under section 109(a)—the rights to lend, resell, or rent her copy.

This is from Reese: a study published by U of Texas Law in 2003. Credit where credit is due, it correctly identified the problem, but we’ve predicted this problem for two decades.

Publishers cannot be trusted to preserve their works. I don’t mean this as a petty insult: factually, they have made no commitment and have no responsibility to preserve their works, so there’s nothing to trust. Trusting someone to do something they don’t do now and haven’t committed to do in the future isn’t trust, it’s negligence.

But publishers’ entire argument for giving them this incredible, unprecedented power over virtually all of civilization’s knowledge is just “dude, trust me.” They don’t make any commitments. They don’t promise to do anything specific. They demand you trust their judgement — which we already know is very bad — with some of the most important power in society.

If you follow up and ask “trust you to preserve works, you mean?” the answer is “well, no, sometimes we decide it’s better not to do that.” If you’re trusting publishers, you need to be ready to respond to them saying “no, it’s fine we did that” after every disastrous mistake. What are you going to do? Society is just hosed.

And so, instead of companies that sell people books they then own, we’ve constructed this ridiculous mousetrap. In order to read books at all, people have to accept the sword of Damocles: your access to what should be your own property is contingent on dozens of new factors that should be entirely irrelevant to the process, like the solvency of the company you bought it from, your account’s online moderation status, or even censorship prompted by reactionary cultural movements. Meanwhile, publishers have their fingers hovering over the killswitch, just waiting for the moment when they project they can make more money by revoking licenses from people who purchased them.

National Emergency Library🔗

Unfortunately, I need to talk about the Internet Archive’s one misplay: the National Emergency Library.

I resent having to talk about the NEL at all, not only because it was a flub made by the heroes of this story, but because it is, for all intents and purposes, irrelevant: no policy issue hinges on the event itself. Too many people make the mistake of granting that the question of CDL or the associated conflict hinges on the NEL: they don’t, and any implication that they do is either a blundering concession or an intentional misdirect. When it counts, even the attacking publishers claim the NEL is irrelevant, so on paper everyone is technically in agreement. But the optics effectively feed back into the reinforced error propaganda engine, and it’s been used — quite effectively — to unjustly cast blame on the blameless aspects of CDL.

The “National Emergency Library” was a temporary program launched March 24, 2020, during the height of COVID, when access to physical libraries was indefinitely shut down by isolation orders. The program ran for only twelve of the scheduled fourteen weeks, due to other book options emerging quickly.

CDL itself, as I’ve already mentioned, was developed well before the pandemic. The whitepaper was published in 2018, although the idea had been in construction for longer.

So, what did NEL actually do? It was a policy change. During the NEL period, the Internet Archive suspended the strict owned-to-loaned requirement, and loaned out DRM-protected CDL books, even if all the copies it owned were already virtually checked-out.

The Case of the Internet Archive vs. Book Publishers The archive had been lending book scans for years. Publishers did not like it but did not sue. What made the pandemic emergency library different was that the brakes were removed. If 10 people, or 100 people, wanted to read a particular book, they could all do so at once.

The emergency library “was as limited as a small city library’s circulation level,” Mr. Kahle insisted. “This was always under control.”

Here’s how the EFF describes the events:

In response to urgent pleas from teachers and librarians, and in consultation with the library community, the Internet Archive temporarily suspended technical enforcement of the one-to-one owned-to-loaned policy. The Internet Archive called this temporary program the “National Emergency Library.” Twelve weeks later, other options had emerged to fill the gap, and the Internet Archive was able to return to the traditional CDL approach.

School districts and university libraries reached out to the Internet Archive concerned that students and teachers could no longer physically access books — including assigned books that schools had already purchased. … the closure of libraries took 650 million books out of circulation.

Since all of the libraries were closed, the Internet Archive reasoned, there were surely more non-circulating copies locked up in shuttered libraries than would be borrowed via the Internet Archive even without those technical controls in place. The Internet Archive was uniquely positioned to be able to address this problem quickly and efficiently. (Indeed, Plaintiffs testified that negotiating workarounds for access issues caused by the pandemic took weeks and sometimes months.)

In my view, the National Emergency Library, even by controlled digital lending’s standards, constituted copyright infringement. And, because the CDL prevents such infringement, this also means it violated the strict requirements of CDL. It seems to me that each copy loaned in excess of the Internet Archive’s stock of physical books backing the digital copies constituted an infringement of copyright. Securing rights to the “non-circulating copies locked up in shuttered libraries” could have mitigated this problem, but in its haste to open and provide emergency access as soon as possible, the Internet Archive did not explicitly secure the necessary rights to those copies.

Even though this infringement did not constitute any meaningful market harm (see Case 1:20-cv-04160-JGK-OTW Document 106), in my opinion, the Internet Archive does owe publishers damages on those excess loans, few though they may be.

But, as I described earlier, instead of addressing the specific violation, publishers exploited this incident to defame the whole CDL concept. But, again, this was mostly done by proxy using reinforced error. In cases like this, where instead of having to tell lies yourselves people will be wrong for you for free, social media is an unimaginable boon.


On June 1, 2020, all four major publishing companies (Hachette, HarperCollins, Wiley, and Penguin Random House) sued the Internet Archive on the basis of copyright infringement. The publishing cartel (the trust which makes up the Association of American Publishers) accuses Internet Archive’s Open Library and the entire concept of controlled digital lending of being in violation of publishers’ copyright.

The IA case represents an inflection point in the attack on property rights, shifting from passively eroding rights opportunistically through control over how technology is deployed to demanding that they’re actively entitled to the ground they’ve encroached on.

At the time of writing, there is a judgement in favor of the plaintiffs (the publishers), although this is being challenged.

Target is Lending, not Digital Formats🔗

The lawsuit’s existence implies a number of falsehoods, like the background idea that the appetites of publishers for profit would be sated if not for this.

One vital thing to note is that, though the complaint seems to be about the Digital in controlled digital lending, meaning the digital formats used, it’s really more concerned with the Lending aspect. Book lending is something people do without paying publishers, so they’re against it on that principle.

EFF Press Release, “Internet Archive Seeks Summary Judgment in Federal Lawsuit Filed By Publishing Companies” “The publishers are not seeking protection from harm to their existing rights. They are seeking a new right foreign to American copyright law: the right to control how libraries may lend the books they own,” said EFF Legal Director Corynne McSherry. “They should not succeed. The Internet Archive and the hundreds of libraries and archives that support it are not pirates or thieves. They are librarians, striving to serve their patrons online just as they have done for centuries in the brick-and-mortar world. Copyright law does not stand in the way of a library’s right to lend its books to its patrons, one at a time.”

With measures like DRM-controlled copies and IA’s five-year embargo, controlled digital lending is actually far less objectionable to traditional book lending. The lawsuit is not driven by an ideological objection to digital lending, it’s an opportunistic attack. Their argument has to be that it’s not analogous to lending physical media, but they don’t really think that. The ways in which it’s analogous to lending one’s own property are the aspects they take offense at. What they are attacking with this is the notion of books-as-property itself.

Eevee, “Copyright is broken” You could fabricate a philosophical reason for this, but here’s my much simpler one: they make more money this way, and they can get away with it. Why sell only one of a thing, when you can sell several of it, possibly even to the same customer?

The technical question of whether an infringing reproduction is involved is a technicality they are not actually interested in or care about. They know perfectly well it isn’t a real reproduction. That’s just a stepping stone to them.

Not NEL🔗

Another implication is that this is somehow a reaction to the National Emergency Library or is only objecting to parts of CDL as a practice. Not so!

Despite publishers’ Olympian gymnastics to cast the National Emergency Library as the inciting incident in the public eye, this was not the case. That was always the implication trickled down to the public discourse, but it was never what they actually argued. As far as the actual policy and lawsuit were concerned, the NEL was a complete red herring, and they knew it.

What’s more, we know they knew it. This lawsuit was not provoked by the emergency library, so much so that the publishers go out of their way to make that point themselves:

Maria A. Pallante, AAP CEO As a point of clarity, we sued Internet Archive on June 1, 2020, for its entire practice of “controlled digital lending,” not only the extra-extreme version that it rolled out in March 2020 with its hyperbolic “National Emergency Library” (NEL) and shut down on June 16, 2020, shortly after the U.S. Copyright Office suggested it was likely outside the bounds of fair use. We previewed a suit in February 2019 with this public statement, which regrettably was ignored. When the pandemic hit, the underlying suit was already being prepared.

No, this suit attacked the entire concept of controlled digital lending. The national emergency library is a very effective red herring, but it’s a red herring nonetheless, and having thoroughly proved that, I’m not falling for that trap. Let’s look at what’s actually at stake, not just the PR.

The Complaint🔗

But enough about what the complaint isn’t, what is the actual complaint? What’s in Hachette Book Group v. Internet Archive? Well, some really incredible stuff. This whole document is a vile, hateful screed, which further takes that hate as a point of pride rather than the grave embarrassment it is.

Originally I went over the entire complaint and every argument it made, systematically debunking all of them. And I had all of that inline here. But, in a futile attempt to keep this essay somewhat constrained and readable, I’ve moved that deep-dive to a separate article.

If you want to see all the points the publishers made, and why each of them are wrong, flip over to

CDL: The AAP is Wrong About Everything / GioCities

If I haven’t exhausted your interest already, do read it. There’s a lot wrong here.


The initial Internet Archive Opinion & Order ruling in favor of the publishers against the Internet Archive came just four days after the hearing.

The publishing cabal found a sympathetic audience in the 77-year-old John G. Koeltl, a man who got his law degree before the conception of ARPANET and took office before the release of Netscape Navigator. This is exactly who publishers needed: someone who cared more about the stability of industry than unimportant things like what must seem to him like an unimportant edge case compared to “real” books.

The order itself mostly consists of uncritically parroting the same arguments from the publishers which I’ve already debunked. So much so, in fact, that I’ve just been including snippets from the ruling as examples of wrong arguments in favor of the publishers, so there’s really nothing new to go over. It’s just wrong for all the reasons we know about already.

The Specter of ReDigi🔗

One of the great problems of the common law system is that all previous decisions are expected to be correct, and extreme weight is put on upholding precedent. Past precedent is assumed to be correct and is therefore, in effect, elevated to a sacrosanct position; not only can it not be effectively overturned, it has to be taken as axiomatically true by subsequent cases. Every overturning of past precedent comes at the expense of the legitimacy of the entire judicial system, and so such overturnings are dramatic and rare. But precedent is also treated as universally applicable, even though the ruling is based on specific details pertinent only to the situation in question.

So, if a legal topic’s history includes a really bad ruling, that cripples you going forward. As for digital first sale, this is precisely what Capitol Records, LLC v. ReDigi Inc. is: poison in the water.

I talked about the enumeration of rights back under New uses invalidate standing pricing assumptions, and also back in You Haven’t Seen Copyright

This monopolist argument is precisely the error ReDigi makes. What makes the ReDigi decision a disastrously bad interpretation is that it inverts the enumerated rights of copyright, saying that copyright is the rule, not the exception, and any rights owners’ might have over their own property must be enumerated in law to exist. This explicitly includes rights that did not exist when the law was written, meaning the exclusive right for new use created for a piece of media belongs, automatically, to the copyright holder, not the person. It locks first sale rights to the mediums that existed when the law was written, and demands Congress amend copyright law for every new technology, or have the power fall to corporations by default.

Ultimately it fundamentally misunderstands copyright as a system of enumerated rights reserved to creators, and instead sees an individual’s right to property as a bizarre outlier to be kept rare and controlled for the sake of continued corporate control.

This is a monstrously bad legal doctrine. It’s totally wrong, including procedurally: the case ignored both prior decisions (hathitrust) and an amicus brief from the American Library Association that predicted the problems the ruling caused. IA tried to avoid ReDigi by avoiding the format shifting argument, because they realized they were hamstrung, but I’m not going to pretend that there’s any virtue in trying to work around it when the entire conceit desperately needs to be overturned.

Having this poison on the books cripples the Internet Archive, as Judge Koeltl continually discards the law itself in favor of ReDigi’s incorrect interpretation of it.

Statement on Flawed Theory of “Controlled Digital Lending” The White Paper further asserts that CDL should be favored on public benefit grounds because it is “aligned with the principles” of first sale limitations under §109(a), even if it does not actually qualify for application of the first sale defense by statutory definition. In December 2018, the Second Circuit in ReDigi expressly rejected the notion that §109(a) is a statement of broad “principles,” as the White Paper asserts, instead holding that it is a provision “for which Congress has taken control, dictating both policy and the means of its execution.” … In short, ReDigi makes clear that there is no broad “principle” underlying §109(a) as enacted, and therefore fair use cannot be used as a way of implementing such a “principle” without doing violence to the Copyright Act as Congress has written it.

Internet Archive Opinion & Order Section 109(a) does not excuse IA’s unauthorized reproduction of the Works in Suit. The first sale doctrine limits a copyright owner’s distribution right under § 106(3), but Section 109(a) “says nothing about the rights holder’s control under § 106(1) over reproduction of a copy or phonorecord.” ReDigi, 910 F.3d at 656.

In ReDigi, the Court of Appeals plainly held that the first sale doctrine has now been codified in Section 109(a), that it does not include a right of reproduction, and that any broader scope of the first sale doctrine should be sought from Congress, not the courts.

This “Congress, not the Courts” line is especially shameless, because the rights are already correctly written in the law. It’s the courts — ReDigi — who got it wrong, and all you need to do is read the law congress wrote to fix that. Taking this incorrect interpretation and applying it to CDL, even when the law says otherwise, is not some neutral decision, it’s a deeply political decision.


The Internet Archive has appealed the ruling, but at time of writing it’s still haunting us.

Currently, all books produced by any of these major publishers and distributed through standard channels, including ebook forms, cannot be owned in digital form at all. The IA ruling re-enforces this, saying that it is a violation of federal law to consider any digital form of a book — including one you create yourself using property you purchased and rights you have — to be property you own.

David Streitfeld, “The Dream Was Universal Access to Knowledge. The Result Was a Fiasco.” …the Archive announced that as a result of the ruling, “the Publisher Plaintiffs will notify us of their commercially available books, and the Internet Archive will expeditiously remove them from lending,” if those publishers have an ebook version available. And as part of a separate agreement with the Association of American Publishers, the trade group which helped to bring the lawsuit, they will also have to honor takedown requests from any of their member publishing houses—which includes most major publishing houses in the U.S., and potentially puts millions of books in jeopardy.

Our Fight is Far From Over | Internet Archive Blogs We remain steadfast in our belief that libraries should be able to own, preserve, and lend digital books outside of the confines of temporary licensed access. We believe that the judge made errors of law and fact in the decision, and we will appeal.

Statement from Internet Archive founder, Brewster Kahle: “Libraries are under attack at unprecedented scale today, from book bans to defunding to overzealous lawsuits like the one brought against our library. These efforts are cutting off the public’s access to truth at a key time in our democracy. We must have strong libraries, which is why we are appealing this decision.”

What the Hachette v. Internet Archive Decision Means for Our Library | Internet Archive Blogs Separately, we have come to agreement with the Association of American Publishers (AAP), the trade organization that coordinated the original lawsuit with the four publishers, that the AAP will not support further legal action against the Internet Archive for controlled digital lending if we follow the same takedown procedures for any AAP-member publisher.

So what is the impact of these final orders on our library? Broadly, this injunction will result in a significant loss of access to valuable knowledge for the public. It means that people who are not part of an elite institution or who do not live near a well-funded public library will lose access to books they cannot read otherwise. It is a sad day for the Internet Archive, our patrons, and for all libraries.


Ultimately, publishers are trying to control things that copyright not only gives them no right to touch, but actively demands they leave alone. Their only acceptable response is to keep their grubby mitts off. But instead we have this attempt at an enormous, unprecedented, all-too-predictable power grab.

The proliferation of digitally licensed-only media — as seen in ebooks, but also other forms of media — increases the immediate accessibility and affordability of select works, selected at the discretion of publishers and media companies. But the license-only model and the eradication of the individual property right means this convenience comes at the cost of destroying the very foundations of the system. Accepting media ecosystems — like the ebook market — in which items cannot be owned is to trade an infinite future of culture in exchange for the chance to consume transient experiences more conveniently. And those transient media experiences are becoming lesser and lesser, according to the whims of their publishers — in fact, in some cases the window’s shut entirely, and we’re losing works before they’ve been released at all.

What’s truly immiserating about this is that the supposed trade-off between the permanence of physical copies and the accessibility of digital leasing is entirely manufactured. Digital formats can retain all of the permanence of physical media and more. In fact, a consistent complaint of companies demanding the right to lock media behind DRM is that digital formats are too resilient and preservable and aren’t subject to the same kinds of natural decay physical media is.

Exactly contrary to the permanence/accessibility trade-off we see in the physical ownership/digital leasing dichotomy, digital formats are actually the best option we have for maximizing both of these goods. The only reason ebooks aren’t permanent, aren’t archivable, aren’t as valuable to (would-be) owners is that the handful of companies who control the supply go to extraordinary lengths to take naturally-resilient media and intentionally cripple their usefulness.

Colluded, conspired, or just independently crooked, it makes no difference: a carefully calculated, enormously expensive enterprise ensures the supply of ebooks is tainted at the source. In fact, there is no supply of ebooks because there is no property supplied to use or consume — only licenses. Responsible suppliers can prevent all the problems of fragility and permeability we see in DRM-locked ebooks and ensure accessibility beyond anything we’ve ever seen. Instead, that future is being burned for corporate profit today.

When companies attack people who are going out of their way to do things legally and buy by the rules, they’re not playing by the rules, they’re gunning for a specific outcome and breaking the system to do it. What was offensive to them was not the morality or legality or equity, but that they didn’t like the outcome. But if they’re going to (very publicly) reject legality and morality and equity as standards, we can’t pretend those frameworks are still “in play” when it comes to their work.

This is why being a good faith actor is an important distinction. It matters if the other person is trying to play a fair game or not. That informs everything. And media companies aren’t.

Which way is it for them? It’s not a way at all. In this universe is no standard, no principle, no concrete reality. It’s just whatever benefits the people in power the most at any given second with no regard for something like consistency. Consistency means they would have to treat you as anything other than barriers between them and your money.

  1. In 2020 at time of writing. In 2022, that number had risen to more than 80, and is 96 today. 

  2. In the Internet Archive’s implementation, when patrons download offline copies of DRM-protected loaned books (usually loaned books are read in-browser), those files are PDFs secured with the same Adobe-developed PDF DRM that publishers and other DRM-maximalists use. 

  3. With publisher cooperation, perpetual ebook licenses are easy to issue, and loaned-copy management is a solved problem. 

  4. In economic terms, “perfect price discrimination allows a monopolist to absorb all consumer surplus.” In Gio terms, this is “bad”. 

  5. Or, at least it was when I wrote this line in October 2023.


  6. The Overdrive company may also use the brand name “Libby” depending on the library. 

  7. And yes, it’s The Authors Guild and not The Authors' Guild, even though that throws me every time I type it. But it’s funny: both grammatically and pragmatically, it’s a guild that contains authors, not one authors possess or control. 

  8. See Chokepoint Capitalism by Rebecca Giblin and Cory Doctorow 

  9. Arguably, I’m being overly credulous with some of these examples. But I’m not overly worried about that in this case because if they’re lying that makes me more right, not less. 

  10. $6,080 (2017 US dollars in 2009) = $5321.41 (2009 US dollars)
    5321.41 2009 US dollars ÷ 10,500 2009 US dollars = 0.506
    ≅ 51% 

  11. Despite Pandemic, 2020 U.S. Book Sales on Par with Past Five Years Given the huge impact of the pandemic on the publishing industry, including publication delays; canceled in-person book tours, signings, and readings; and the closure of brick-and-mortar bookstores during the enforced lockdowns in 45 states, many had expected last year’s overall book sales to drop significantly.

    “The Authors Guild is delighted by the news that the book industry continues to thrive and the renewed interest in reading as entertainment,” said Mary Rasenberger, CEO of the Authors Guild. “This bodes well for the future of the writing profession. The Authors Guild is here to ensure that authors, and not just publishers and retail platforms, also benefit from the health of the industry, and to reverse the downward trends of author incomes.”

  12. Note that this graph and the original 42% chunk are both pre-inflation, so they are directly comparable. 

  13. A federal subsidy to publishers whose books are lent in public libraries 

Howdy! If you found my writing worthwhile, the best thing you can do to support me is to share an article you found interesting somewhere you think people will appreciate it. Thanks as always for reading!