Sunday, July 4, 2021

The Ebook Turns 50

On July 4, 1971, Michael Hart made the text of the Declaration of Independence available on arpanet (which is now the Internet), using the gopher protocol (look it up). Although books in digital form certainly existed before that, many of us regard the beginning of Project Gutenberg as the birth of the ebook. There were computer-readable books on magnetic disks, punch cards and the like, but the revolutionary element of Project Gutenberg was the distribution method. Printed books, after all, are a digital media, it just that the bits are embodied by the presence or absence of ink rather than electrons on a transistor gate. Sending the bits over a wire or a fiber is what puts the 'e' in ebook.

The birth of the ebook was a political event as much as a technical achievement. The choice of the "Declaration of Independence of the United States" as etext #1 couldn't have been solely an expression of patriotic fervor. Rather, I think it was a manifestation of the radical belief that everyone should have access to the printed word, without having to pay for the privilege. (Yes, libraries are radical in this way, too!).

As Thomas Jefferson put it:

... it becomes necessary for one people to dissolve the political bands which have connected them with another, and to assume, among the Powers of the earth, the separate and equal station to which the Laws of Nature and of Nature's God entitle them.

In the context of 1971, the "bands" that needed dissolving were expensive services such as Dialog. The idea that users had to pay Dialog per word to read the Declaration mush have been galling to Hart. (Let's overlook the fact that he and other denizens of the 1971 arpanet got their access for "free" because someone else was paying.) Books are things in their own right; stripping ebooks of their "bands" to a single device or service is what put the "book" into ebook.

Although Project Gutenberg is now delivering about 50 million ebooks a year, about 2% of global ebook unit sales, until at least 2009 it delivered the majority of the world's ebooks. Today, that position has been taken by Amazon's Kindle. Just as the United States can't ignore the ideals that led to its founding, the stakeholders of the ebook ecosystem- authors, publishers, distributors, libraries, and readers, all of us need to remember that the ebook was born out of a desire for freedom.


Note: Though I've been helping Project Gutenberg modernize its technology, I don't speak for them in any way, though I am certainly in awe of what they've achieved! If you'd like to support my work advancing freedom for ebooks, consider a donation to the Free Ebook Foundation.

Monday, February 22, 2021

Open Access for Backlist Books, Part II: The All-Stars

Libraries know that a big fraction of their book collections never circulate, even once. The flip side of this fact is that a small fraction of a library's collection accounts for most of the circulation. This is often referred to as Zipf's law; as a physicist I prefer to think of it as another manifestation of log-normal statistics resulting a preferential attachment mechanism for reading. (English translation: "word-of-mouth".)

In my post about the value of Open Access for books, I suggested that usage statistics (circulation, downloads, etc.) are a useful proxy for the value that books generate for their readers. The logical conclusion is that the largest amount of value that can be generated from opening of the backlist comes from the books that are most used, the "all-stars" of the library, not the discount rack or the discards. If libraries are to provide funding for Open Access backlist books, shouldn't they focus their resources on the books that create the most value?

The question of course, is how the library community would ever convince publishers, who have monopolies on these books as a consequence of international copyright laws, to convert these books to Open Access. Although some sort of statutory licensing or fair-use carve-outs could eventually do the trick, I believe that Open Access for a significant number of "backlist All-Stars" can be achieved today by pushing ALL the buttons available to supporters of Open Access. Here's where the Open Access can learn from the game (and business) of baseball.

"Baseball", Henry Sandham, L. Prang & Co. (1861).
  from Digital Commonwealth


Baseball's best player, Mike Trout, should earn $33.25 million this year, a bit over $205,000 per regular season game. If he's chosen for the All-Star game, he won't get even a penny extra to play unless he's named MVP, in which case he earns a $50,000 bonus. So why would he bother to play for free? It turns out there are lots of reasons. The most important have everything to with the recognition and honor of being named as an All-Star, and with having respect for his fans. But being an All-Star is not without financial benefits considering endorsement contracts and earning potential outside of baseball. Playing in the All-Star game is an all-around no-brainer for Mike Trout.

Open Access should be an All-Star game for backlist books. We need to create community-based award programs that recognize and reward backlist conversions to OA. If the world's libraries want to spend $50,000 on backlist physics books, for example, isn't it better to spend it on the the Mike Trout of physics books than on a team full of discount-rack replacement-level players?

Competent publishers would line up in droves for major-league all-star backlist OA programs. They know that publicity will drive demand for their print versions (especially if NC licenses are used.) They know that awards will boost their prestige, and if they're trying to build Open Access publication programs, prestige and quality are a publisher's most important selling points.

The Newbury Medal

Over a hundred backlist books have been converted to open access already this year. Can you name one of them? Probably not, because the publicity value of existing OA conversion programs is negligible. To relicense an All-Star book, you need an all-star publicity program. You've heard of the Newbury Medal, right? You've seen the Newbury medal sticker on children's books, maybe even special sections for them in bookstores. That prize, award by the American Library Association every year to honor the most distinguished contributions to American literature for children, is a powerful driver of sales. The winners get feted in a gala banquet and party (at least they did in the before-times). That's the sort of publicity we need to create for open access books.

If you doubt that "All-Star Open Access" could work, don't discount the fact that it's also the right thing to do. Authors of All-Star backlist books want their books to be used, cherished and remembered. Libraries want books that measurably benefit the communities they serve. Foundations and governmental agencies want to make a difference. Even publishers who look only at their bottom lines can structure a rights conversion as a charitable donation to reduce their tax bills.

And did I mention that there could be Gala Award Celebrations? We need more celebrations, don't you think?

If your community is interest in creating an Open-Access program for backlist books, don't hesitate to contact me at the Free Ebook Foundation!

Notes

I've written about the statistics of book usage here, here and here.

This is the third in a series of posts about creating value of Open Access books. The first two are:

Tuesday, February 16, 2021

Open Access for Backlist Books, Part I: The Slush Pile

"Kale emerging from a slush pile"
(CC BY, Eric Hellman)
Book publishers hate their "slush pile": books submitted for publication unsolicited, rarely with literary merit and unlikely to make money for the publisher if accepted. In contrast, book publishers love their backlist; a strong backlist is what allows a book publisher to remain consistently profitable even when most of their newly published books fail to turn a profit. A publisher's backlist typically consists of a large number of "slushy" books that generate negligible income and a few steady "evergreen" earners. Publishers don't talk much about the backlist slush pile, maybe because it reminds them of their inability to predict a book's commercial success.

With the advent of digital books has come new possibilities for generating value from the backlist slush pile. Digital books can be kept "in print" at essentially no cost (printed books need warehouse space) which has allowed publishers to avoid rights reversion in many cases. Some types of books can be bundled in ebook aggregations that can be offered on a subscription basis. This is reminiscent of the way investment bankers created valuable securities by packaging junk bonds with opaque derivatives.

Open access is a more broadly beneficial way to generate value from the backlist slush pile. There is a reason that libraries keep large numbers of books on their shelves even when they don't circulate for years. The myriad ways that books can create value doesn't have to be tied to book sales, as I wrote in my previous post.

Those of us who want to promote Open Access for backlist ebooks have a number of strategies at our disposal. The most basic strategy is to promote the visibility of these books. Libraries can add listings for these ebooks in their catalogs. Aggregators can make these books easier to find.

Switching backlist books to Open Access licenses can be expensive and difficult. While the cost of digitization has dropped dramatically over the past decade, quality control is still a significant conversion expense. Licensing-related expenses are sometimes large. Unlike journals and journal articles, academic books are typically covered by publishing agreements that give authors royalties on sales and licensing, and give authors control over derivative works such as translations. No publisher would consent to OA relicensing without the consent and support of the author. For older books, a publisher may not even have electronic rights (in the US, the Tasini decision established that electronic rights are separate from print rights), or may need to have a lawyer interpret the language of the original publishing contract. 

While most scholarly publishers obtain worldwide rights to the books they publish, rights for trade books are very often divided among markets. Open-access licenses such as the Creative Commons licenses are not limited to markets, so a license conversion would require the participation of every rights holder worldwide. 

The CC BY license can be problematic for books containing illustrations or figures used by permission from third party rights holders. "All Rights Reserved" illustrations are often included in Open Access Books, but they are carved out of the license by separate rights statements, and to be safe, publishers use the CC BY-ND or CC BY-ND-NC license for the complete book, as the permissions do not cover derivative works. Since the CC BY license allows derivative works, it cannot be used in cases where translation rights have been sold (without also buying out the translation rights). A publisher cannot use a CC BY license for a translated work without also having rights to the original work.

The bottom line is that converting a backlist book to OA often requires economic motivations quite apart from any lost sales. Luckily, there's evidence that opening access can lead to increased sales. Nagaraj and Reimers found that digitization and exposure through Google Books increased sales of print editions by 35% for books in the Public Domain.  In addition, a publisher's commercial position and prestige can be enhanced by the attribution requirement in Creative Commons licenses.

Additional motivation for OA conversion of the backlist slush pile has been supplied by programs such as used by Knowledge Unlatched, where libraries contribute to to a fund used for "unlatching" backlist books. (Knowledge Unlatched has programs for front list books as well.) While such programs can in principle be applied for the "evergreen" backlist, the incentives currently in place result in the unlatching of books in the "slush pile" backlist. While value for society is being gained this way, the willingness of publishers to "unlatch" hundreds of these books poses the question of how much library funding for Open Access should be allocated to the discount bin, as opposed to the backlist books most used in libraries. That's the topic of my next post! 

Notes

This is the second in a series of posts about creating value of Open Access books. The others are:

Friday, February 12, 2021

Creating Value with Open Access Books

Can a book be more valuable if it's free? How valuable? To whom? How do we unlock this value?

a lock with ebooks
I've been wrestling with these questions for over ten years now.  And for each of these questions, the answer is... it depends. A truism of the bookselling business is that "Every book is different" and the same is true of the book freeing "business".

Recently there's been increased interest in academic communities around Open Access book publishing and in academic book relicensing (adding an Open Access License to an already published book). Both endeavors have been struggling with the central question of how to value an open access book. The uncertainty in OA book valuation has led to many rookie mistakes among OA stakeholders. For example, when we first started Unglue.it, we assumed that reader interest would accelerate the relicensing process for older books whose sales had declined. But the opposite turned out to be true. Evidence of reader interest let rights holders know that these backlist titles were much more valuable than sales would indicate, thus precluding any notion of making them Open Access. Pro tip: if you want to pay a publisher to make a books free, don't publish your list of incredibly valuable books!

Instead of a strictly transactional approach, it's more useful to consider the myriad ways that academic books create value. Each of these value mechanisms offer buttons that we can push to promote open access, and point to new structures for markets where participants join together to create mutual value.

First, consider the book's reader. The value created is the reader's increased knowledge, understanding and sometimes, sheer enjoyment. The fact of open access does not itself create the value, but removes some of the barriers which might suppress this value. It's almost impossible to quantify the understanding and enjoyment from books; but "hours spent reading" might be a useful proxy for it.

Next consider a book's creator. While a small number of creators derive an income stream from their books, most academic authors benefit primarily from the development and dissemination of their ideas. In many fields of inquiry, publishing a book is the academic's path to tenure. Educators (and their students!) similarly benefit. In principle, you might assess a textbook's value by measuring student performance.

The value of a book to a publisher can be more than just direct sales revenue. A widely distributed book can be a marketing tool for a publisher's entire business. In the world of Open Access, we can see new revenue models emerging - publication charges, events, sponsorships, even grants and memberships. 

The value of a book to society as a whole can be enormous. In areas of research, a book might lead to technological advances, healthier living, or a more equitable society. Or a book might create outrage, civil strife, and misinformation. That's another issue entirely!

Books can be valuable to secondary distributors as well. Both used book resellers and libraries add value to physical books by increasing their usage. This is much harder to accomplish for paywalled ebooks! Since academic libraries are often considered as potential funding sources for Open Access publishing it's worth noting that the value of an open access ebook to a library is entirely indirect. When a library acts as an Open Access funding source, it's acting as a proxy for the community it serves.

This brings us to communities. The vast majority of books create value for specific communities, not societies as a whole. I believe that community-based funding is the most sustainable path for support of Open Access Books. Community supported OA article publishing has already had plenty of support. Communities organized by discipline have been particularly successful: consider the success that ArXiv has had in promoting Open Access in physics, both at the preprint level and for journals in high-energy physics. A similar story can be told for biomedicine, Pubmed and Pubmed Central. A different sort of community success story has been SciELO, which has used Open Access to address challenges faced by scholars in Latin America.

So far, however, sustainable Open Access has proven to be challenging for scholarly ebooks. My next few posts will discuss the challenges and ways forward for support of ebook relicensing and for OA ebook creation: