orphan books

The post from Thursday about the Google “windfall” provoked a lot of information sources to help me understand the settlement, large parts of which I clearly did not. We’ll go over the answers I got (as I understand them; my understanding seems to be a moving target…) to the questions Michael Cairns and I posed and then I have a few more thoughts about where this leaves us.

1. There are no “rules” about what BRR can keep for its own operations. The notion is that since the Board is composed of representatives of the net recipients, who will benefit from tight cost control, that there is incentive for the Board to manage expenses well. 

2. A share of the money from the Google licensing activity that goes unclaimed because it is attributable to orphan works is first available to pay “inclusion fees” for the opters-in work of $200 a title. (This is not to be confused with the $60 per title scanned before this May 5 which is paid as part of the settlement.) Beyond that, the licensing money is divided among the opters-in on some to-be-determined formula based on usage.

3. The allocation of money to c/r holders is a little bit by volume of material (the up to $200 per book mentioned above) and then thereafter by a to-be-determined measurement of use.

4. The “costs” Google incurs, including sales costs, are all included in the 37% deduction. That 37% is actually arrived at because the split is 70-30 after a 10% expense allocation. There is a provision in the settlement to enable rightsholders to claw back their share of that 10% out of unclaimed revenues. So Google keeps 37% of the total it processes, but declared rightsholders could still get 70% of the revenue attributable to their books with the difference coming out of orphan revenues (before additional payments that could occur out of additional unclaimed revenue.)

The answers to the rest of our questions would be purely speculative. Whether new models are clearly contemplated (like print-on-demand or downloadable ebooks) or not (like licensing orphans for press runs), deals for orphan books can only occur by mutual agreement of Google and the BRR.

And therein lies one big rub. We must assume that each of the three entities with decision-making power: Google, the Authors Guild, and the Association of American Publishers, will act in the best interests of its principal stakeholders. For Google that would be its shareholders; for the others it would mean their author and publisher constituents.

For those (like me) whose primary interest in the settlement is the liberation of all this stranded (orphan) IP, this is discouraging. I don’t believe Google would have reason to object to seeing old books published again, although, for those few that would be, their search “exclusive” could conceivably be compromised. In the overall scheme of things, that would be small beer.

And publishers would also have an interest in allowing those books to be relicensed and published again because, after all, publishers will be the ones relicensing.

Authors, on the other hand, would have no interest in seeing thousands of books come back to active promotional life. If you’re working on a new biography of Franklin Roosevelt, do you really want to see 25 of them published over the last 70 years and long since buried suddenly come back to compete with yours? I see no upside for today’s author in liberating the orphans and I would expect that to be an important consideration for Authors Guild representatives on the BRR board.

What that means is that this settlement does not eliminate the need for legislation to further break up the logjam blocking complete access to the orphans. It makes it important that Google be sincere in its statements that it still supports orphan legislation. My understanding is that it is representatives of non-book IP that have a lot to do with blocking such legislation. Publishers would have reason to favor it. Would authors?

And will approval of this settlement or its rejection make new and constructive orphan legislation more likely? It’s only a guess, and I know more about politics than I do about orphan works legislation, but I’d imagine that the game-changer of this settlement would be a spur to action and rejection of it would leave the matter in the courts.

In the conversations I had yesterday bringing me up to whatever speed I’ve been able to attain, it was pointed out to me that the number of orphans may be large but the usage would be greater of those where copyright is claimed. The books in the database that will get the most use will be those academic publications which don’t even have reversion clauses so the copyright owner is not obscure. Think: you’re doing a book or paper on paranoia and you need to read what people thought in 1937 and 1951 and 1986. It will be situations like that which drive the page views.

Trying to estimate how many titles are involved (the $60 fees will be paid only on those that have been scanned until May 5, though many more will be scanned thereafter) is almost impossible. But even if 25-to-50 percent are claimed, and they amount to a somewhat higher percentage of the usage, the “windfall” is likely to be in the low eight figures annually. A big number.

As to the big question we posed: what happens with that windfall, the answer is that, primarily, it goes to the opters-in and primarily based on usage. How many of them will there be? A million? Several million? Cairns and I have to go back to our economic model in light of what we’ve learned, but we know those books will be sharing windfall revenues of some tens of millions of dollars annually. Assuming some sort of Pareto distribution of the revenues, that could result in some significant found money for select publishers — likely ones that are academic, sci-tech, professional, and have a very lengthy backlist. We’re likely to be talking about a handful of multi-million dollar windfalls.

The brand new position of the BRR will be as a licensor of the opt-in titles for whatever uses it can persuade the copyright owners to allow. BRR will be trying to demonstrate value here, both to copyright owners (so they keep putting new books into the system) and to potential licensors, based on BRR’s position as an aggregator of a massive number of copyrights (imagine if it is a million or more: that’s a lot in one place!) 

From what we see here, the two most important questions (about which reasonable people can certainly disagree, and nobody can really know yet):

Can BRR, given its assets and revenue by fiat and its restrictions by structure, serve a significant function beyond maintaining the database and adjudicating book rights disputes? (Or will it simply serve a 1-time clearance function and then process checks?)

Would acceptance of this settlement be a spur to get more far-reaching stranded IP liberation through legislation? Or would stopping it make it more likely that the politicians would act?

Share this post

Michael Cairns and I have both been frustrated with most of the conversation surrounding the Google Book Search settlement. The principal concerns of most of the participants in the dialogue seem to be: 

1. Has Google unfairly captured a monopoly on some content?

2. Has the “class” of “orphan authors” been dealt with fairly, since they aren’t really “represented” in the negotiations?

3. If this case doesn’t adjudicate questions of “fair use”, does that ipso facto mean that a settlement is a bad idea?

4. Can any settlement of broad public-interest questions about copyright and use be legitimately resolved in any way other than through legislation, since, after all, copyright rules are created through legislation?

We believe it is unfortunate that the attention has been focused there because there are some very real commercial questions that we think need answers to fully appreciate the practical implications of the settlement. We’ve been doing our best to build a model of what revenue will be and where it will go. Trying to do that makes it very clear how much important detail has been omitted from the debate we’ve heard so far (and we’ve both heard a lot of it.) Here’s a starter list of questions that need answers to forecast this business which we hope that people more familiar with the terms of the settlement than we are might be able to answer for us.

By far the most significant questions we have concern how the revenues are divided,  and these are significant questions because the preliminary financial projections we have done indicate that this database of content will produce hundreds of millions of dollars for Google and BRR.

1. We understand that revenue flows from the books in the database to Google and then 63% of that to the BRR.  Are there any rules set yet about what BRR can keep of these revenues for its own operations before it passes on the remainder to rightsholders? We might logically assume that BRR would require a diminishing percentage as revenues rise, but we wonder how those controls will be established.

2. We understand that future orphan claims can be compensated going back five years from the time of the claim. That suggests that the BRR has to hold the orphans’ money in escrow going back five years. The key question we have not heard discussed is: what happens with the money older than five years? We’ll expand on that below.

3. How is the allocation of revenue determined for the copyright owners in the database? Are they paid by the amount of content in the database? Or by the number of pages viewed of their work in the databases licensed? Or on some other basis? Or is that something still to be determined by the BRR?

4. We believe that any sales costs Google incurs, such as hiring another organization to help them sell licenses, would come out of Google’s 37%. Is that correct, or can Google deduct sales costs before dividing the money?

And we have a bunch of questions to which the answer might be, Book Rights Registry (BRR: the entity with a Board of eight — four from the AAP and four from the Writers Guild — that can therefore deadlock) just decides. We want to know if there are any barriers or constraints on any of the following within the terms of the settlement.

5. We know that the database will have greater value and greater use if it is curated and merchandised. Is there a plan for this? Is there even a concept for how a third party could be compensated for doing this curation and merchandising? 

6. We see opportunities for services & solutions providers such as SharedBook and to add value by providing the ability for customization, personalization, and annotation of the IP and then perhaps to have the end product  sold both as a book and as an ebook. Is this a deal that BRR would just be free to make on whatever terms they deemed appropriate?

7. Does BRR get to retain a larger percentage of revenues for ‘home-grown’ product initiatives such as the ones we are describing?  This revenue doesn’t come from Google like the institutional licensing and ebook sales money does, so does Google still get its full 37%?

8. To leverage non-database (non-Google) revenue opportunities we see three primary functions that need building: a storefront, an assembly technology (which could be much simpler than SharedBook: what if you wanted to put five Dickens novels together and print them?), and actual printing and delivery. Do we assume that BRR is free to put these capabilities together however it likes? Could it grant this as a sublicensed monopoly to Amazon or Ingram or Barnes & Noble? 

9. We puzzle over the pricing of POD. May we assume that BRR would be free to pursue any model? We can see two immediately: one is that BRR gets a percentage of the book’s retail (or wholesale) price and the other is that BRR charges a flat rate for the book content and the packager-reseller then charges whatever they want for the resulting book. Is BRR free to make these deals as it likes?

At the core of the important discussion about the settlement which has not occurred is  the question “what happens to the money the orphan books earn?” If it is divided among all the opters-in, which seems at least as reasonable as letting BRR just keep it, then there is a huge potential windfall to the copyright holders who stay in this database. That has not been mentioned by anybody (as far as we know). By consensus, 5 million of the 7 million books that are going to earn many tens, if not hundreds, of millions of dollars annually are orphans so, by definition, they have no copyright owner to pay! Either BRR keeps the money or they give it to the contributors to the database.

Not to have discussed this strikes us as a startling omission. Somebody gets a windfall much larger than the one going to Google. Who is it?

This post is an intellectual joint effort with Michael Cairns, who did a very helpful editing job on the first draft as well.

Share this post