Google Book Search settlement and access to out of print books

Google Book Search settlement and access to out of print books
Tuesday, June 2, 2009 at 8:45 AM
Posted by Derek Slater, Policy Analyst

A few weeks ago we blogged about how the Google Book Search settlement agreement will expand access to books for readers in the United States. We recently took another important step forward, when the University of Michigan announced an expanded agreement with Google that will take advantage of the settlement to expand public access to millions of works from the University's collection.

Today we want to explain one area of the settlement in more detail: how the agreement increases access to out-of-print books, including books that some refer to as "orphans."

Out of Print Books

The settlement covers books that Google scans from libraries’ collections, the majority of which may be in copyright but are out of print. These books would ordinarily be hard to access, and one of the principal benefits of the settlement agreement is that it allows people to search, preview, and purchase access to them.

The settlement will also make it far easier for anyone -- including Google's competitors -- to license the use of most out-of-print books. As authors and rightsholders claim their books under the settlement, information about what books have been claimed and who claimed them will be made publicly available, allowing others to take advantage of this information. What's more, the settlement creates an independent, not-for-profit Book Rights Registry run by authors and publishers that will be able to license other services on behalf of rightsholders who want it to do so.

Today, it may be costly for someone to track down the rightsholders of many older works, and there's not much reason for those rightsholders to make themselves easy to find, because they can't earn any money from their work by selling it in stores.

The settlement agreement addresses this conundrum in concrete ways. Because out-of-print books will get a renewed commercial life through Book Search and other services licensed by the Registry, rightsholders are more likely to claim their books. In this way, the settlement creates real financial incentives for owners of out-of-print works to come forward.

In addition, one of the Registry's core missions is to locate the owners of unclaimed books in order to help authors and publishers claim their works and the revenues due them under the settlement. There is an extremely broad notification program already underway.

Orphan Works

While the majority of all book titles are out-of-print, only a minority of them are what some people call “orphans.” This term isn’t defined in U.S. law and people disagree on the definition, but it typically refers to in-copyright works whose owners cannot be identified or found.

As “parent” rightsholders claim their books through the Book Rights Registry, we think it will become clear that most out-of-print books are not actually “orphans.” Books that were once difficult for anyone to license will become books that are very easy for everyone to license, either through the Book Rights Registry or directly from their owners. Furthermore, many books that some think are in-copyright orphans (including a large percentage from 1963 or before) are actually out-of-copyright, and Google is working to make more information available that can clarify their copyright status.

Of course, some rightsholders may still be too difficult to find. Under the settlement Google will be able to open up access to truly orphaned books, but we still think more needs to be done to allow anyone and everyone to use these works. Any company or organization that wants to open up access to this untapped resource should be able to do so. The settlement is not a panacea, since it only covers a subset of orphaned works, provides only certain uses, and is not able to extend these uses to other providers. The need for comprehensive orphan works legislation is not diminished.

That's why Google has been working for years to pursue legislation to provide meaningful avenues for any entity to use orphaned works. We first explained our views to the Copyright Office on this subject over four years ago, and it will remain one of Google's priorities to work to pass effective orphan works legislation.

Fortunately, there isn't an either-or choice between legislation and the settlement. While we work with others towards a comprehensive orphan works solution, the settlement agreement takes one important step towards opening up access to orphaned books in the meantime. If the agreement is approved, anyone across the nation -- from a school child in rural America to a blind PhD candidate -- will have an easy way to go online and read books that would otherwise be hard to access. We are excited about making that possible.

Publishers Lunch Report on BISG Sales Information

The Problem with Statistics: BISG's New Method Still Inflates the Unknown

It's that time of year again, when we all try to reckon with a wave of imprecise measures of our increasingly complex and nuanced business. The need for true data seems to increase in direct proportion to the supply of inadequate evaluations.

In recent weeks, we've looked a little bit at the confusingly broad distinctions in the Bowker counts of new titles published last year and tried to reckon with Amazon's unsubstantiated glimpse into rising Kindle sales. Discerning readers already know that on a standing basis, we very consciously do not report periodic numbers from the AAP, Census Bureau, and IDPF since they are so incomplete (and sometimes inconsistent) as to be more confusing than illuminating. Nielsen Bookscan is great for what it covers, but is also incomplete (and doesn't capture certain things, like ebook sales, at all), and Bowker's PubTrack is interesting at reflecting very specific buying patterns and demographics but is no substitute for actual market data.

Today come the preliminary 2008 numbers from the Book Industry Study Group's annual BOOK INDUSTRY TRENDS (the printed study will not be available until July), which once upon a time served the role of blending a variety of data streams together. In recent years, the BISG developed and pushed their interest in trying to calculate the size of the so-called under-represented/"under-the-radar" publishers, blending surveys and multiplication with actual data.

For 2009 the group set out to revamp the entire TRENDS methodology. The goals were noble, and included trying to more accurately track new and growing segments such as electronic books and digital audio, get a better understanding of both actual book exports as well as rights sales, and new divisions that aim to better parse sales by channel (internet; wholesale; bookstore; specialty; public libraries; educational/academic; organizations/direct). The ever-changing university press category gave way to a broader look at "scholarly publishing," the dwindling book clubs have been merged into the channel data, and mass-market is now included in trade (which breaks out sales by format). Other interesting changes include data on backlist vs. frontlist sales for the first time.

As a vocal critic of certain aspects of the previous process, I was kindly invited to join an advisory group and did so happily, understanding that the BISG would still make their own choices about how to proceed.

While I applaud many of the new goals, I was--and remain ever more--a skeptic on multiple points:

- I still do not believe the aggressive estimates of sales from the thousands of publishers with sales of under $50 million annually (e.g. almost all) that somehow do not appear in any other known sources of sales. This year they proclaim such companies comprise 32 percent of total industry revenue, or almost $13 billion in all.

But that is based on less hard data than ever. In fact, the number of companies responding to this year's survey was much lower than in previous years (when only the smaller publishers were surveyed). Only about 500 companies responded, of 70,000 queried by e-mail. Another 50,000 so-called publishers registered with Bowker don't even have e-mail addresses listed. But through the magic of multiplication (and fancy statistical methods to align and "triangulate" the new information versus the previous, also unproven, numbers) a fraction of real information turns into billions of dollars of revenue.

- I was skeptical about the ability of e-mail surveys to produce the kind of data and response that was sought, and remain so. If the industry wants better data, it may take considerably more resources to pull it off.

- I agree that it's interesting to further study this large segment of elusive companies; I continue to believe that it pollutes the known data to blend these two categories together, however. Making things worse, there is no longer any "traditional" data line in the new TRENDS report. Your only option is the magically multiplied numbers.

- I've seen first-hand the earnest intentions of the BISG to create a better map of the landscape. But just as you can't be a little bit pregnant, data like this can't be a little bit flawed--it's right, or it isn't.

The Numbers
Per above, with stated reluctance, here are the numbers released today by the new Book Industry TRENDS study. By their count, the publishing industry as a whole reached sales of $40.32 billion in 2008 (including about $13 billion of imagined revenue for small companies that didn't actually report their numbers).

Take away that $13 billion and you get $27 billion--close to the same ballpark as the AAP's estimate of $24.255 billion for the year. (Last year's BISG "traditional summary" total was about $30 billion; trade and mass market combined comprised about $9.5 billion in that report.) The new BISG breakdown is:

Trade 14,734 (down 2.1%)
Adult 11,125 (down 2.3%)
Juvenile 3,609 (down 1.3%)

Religious 2,318 (down 10.0%)

Professional 8,696 (up 4%)

Scholarly 1,428 (up 3.9%)

Elhi 7,373 (up 4.5%)

College 5,772 (uo 4.5%)

TOTAL $40,321 billion

Compare that to the AAP:

Trade 10,049
Adult 4,800 (down 5.5%)
Juvenile 3,278 (Down 4.8%)
Book clubs 600
Mass Market 1,086
Audio 172
eBooks 113

Religious 783

Professional 3,457

El-hi 6,076

Higher Ed 3,777

Other 168

TOTAL $24,255 billion

In unit data, the BISG says that adult trade comprised 1.348 billion units and juvenile trade comprised 889 million units, both up less than one percent from the previous year.

It's hard construct a completely comparable set of unit data from BISG versus Bookscan, but to come relatively close, we looked at BISG's bookstores, internet, and wholesale numbers combined, which assert trade unit sales of 1.244 billion units in 2008. Bookscan's own number for the outlets they cover was 700 million units--meaning Bookscan would have registered only 56% of sales in overlapping outlets. Take that as you will. But the BISG also asserts that total trade units sold were 2.134 billion--meaning Bookscan would only be capturing less than a third of sales.

From within some of their new features, TRENDS says that adult trade backlist revenues declined by 4.7 percent, while frontlist held nearly flat. By their count, adult frontlist comprises 53 percent of sales. They say juvenile frontlist rose 37 percent while backlist fell 11 percent, in this case with frontlist comprising under 29 percent of the market.

Some of the new line items would seem to underestimate rather than over-report, however; they say that adult trade electronic sales of $113 million grew only 6.8 percent over the prior, and they put all trade audio at just $50 million (with almost no children's audio revenue recorded at all). That's overtaken in their count by $88 million of religious audio revenue.