Thursday, 26 April 2012
What do authors fear from "Orphan Works" licensing proposals?
Earlier this month I represented the National Writers Union at a fascinating high-level symposium at Berkeley on so-called "orphan works" -- written and other publications, the holders of certain reproduction rights to which cannot be identified and/or found by some or all of those who want to reproduce those works.
Slides of presentations and papers from the conference, including the white paper on Facts and Fallacies of Orphan Works (PDF) that I submitted with the endorsement of the NWU, have been posted online. Audio archives of some sessions have also been posted, and more will hopefully be made available eventually.
Many well-meaning librarians and other advocates for public access to information at the conference found it difficult to understand why writers would object to proposed schemes that would allow works deemed "orphaned" to be copied for certain purposes, by certain users, without those users having to get permission from the authors or other holders of rights to those works.
"What's the problem? What are you afraid of?"
It annoys me that so many librarians and others whose livelihoods depend on the work of writers in creating the books they use have so little understanding of our working lives that they need to ask this question. Nevertheless, the question reflects what is often sincere puzzlement, and deserves a serious answer. Here's a first attempt to provide one.
(While I have discussed these ideas with other writers, including fellow members of the NWU, I should make it even more clear than usual that I am speaking here solely as an individual author, in an effort to inform the discussion and answer some of the questions posed to me at the symposium. Unlike the white paper distributed at the Berkeley symposium, my comments here have not been endorsed by the NWU or any organization.)
Fundamentally, as discussed in more detail below, I think authors are afraid of two possible unfair outcomes of any scheme for identifying books as "orphaned" and/or "out of commerce" (many of the proposals, including those under consideration in the European Union, deal with both of these categories of works) and granting default licenses for their use.
First, authors fear that even facially neutral "orphan works" and/or "out of commerce" schemes are likely to incorporate defaults and procedures that are implicitly biased towards publishers and against authors, and which result in a de facto reallocation of rights and/or revenue share in works deemed "orphaned" and/or "out of commerce" from authors to publishers.
Second, authors fear that some of our own works from which we are, in fact, generating revenues, especially through online self-publication, are nevertheless likely to be classified as "orphaned" and/or "out of commerce" under the procedures being proposed, and that the uses by libraries and others which are proposed for such "orphaned" and/or "out of commerce" works would reduce or destroy authors' ability to continue to generate revenues from those works.
Are these fears well-founded? Would working authors' livelihoods be "collateral damage" -- perhaps unintentionally on the part of librarians and academics -- of schemes to create a "digital library" in which the librarians, the software architects, the builders, and everyone else would be paid -- except the authors whose work would comprise its contents? How would this happen?
1. Structural bias of "orphan works" and "out of commerce" schemes toward publishers vis-à-vis authors
Librarians and other advocates for the reading public certainly don't intend to favor publishers' interests over those of authors. In fact, the interests of both readers and writers are typically opposed to those of publishers and other intermediaries that technology is increasingly rendering obsolete and parasitic.
Direct, decentralized, disintermediated network distribution of written work from writers to readers can result in simultaneously lower prices to readers and higher revenues to writers that enable more writers to devote their time to writing and produce more work for readers.
That's obviously a win-win situation for readers and writers. The losers, equally obviously, are traditional publishers. Rather than accepting their obsolescence or inessentiality in a network distribution model, publishers are trying both to take rights to digital reproduction from writers -- by fair means or foul -- and to redefine the terms of debate to enable publishers to speak for authors, "negotiate" on our behalf, and sell out authors' interests for publishers' profits.
Many well-meaning, public-spirited librarians and other advocates for the reading public have unwittingly been taken in by these tactics, and have unintentionally become pawns in publishers' attempts at wholesale theft from writers of electronic rights to our work.
(One of the greatest ironies of the Berkeley symposium was seeing people who are normally advocates for a decentralized public informational commons and disintermediated peer-to-peer information sharing arguing in the case of "orphaned" and "out of commerce" works for enforced centralization and re-intermediation of licensing through registries and collecting societies.)
As discussed in the NWU white paper, much of the problem comes from the conflict of interests having been falsely framed as being "rightsholders versus readers" rather than "readers and writers versus the men in the middle".
The goal of copyright, after all, is to promote the interests of creators, so as to encourage them and enable them to devote more of their time to creativity. Any benefit of copyright for publishers or other intermediaries is purely incidental, and should be taken as evidence of friction and inefficiency in the creativity-promoting system of copyright, not of its success.
There are a wide variety of ways that organizations and procedures structured around the concept of "rightsholders" can incorporate provisions that favor publishers over authors. We saw some of these in the proposed Google Books settlement, in which even the US Department of Justice said unequivocally in its influential filing with the court that the proposed Book Rights Registry would have been controlled by publishers despite nominally equal representation for authors.
Many "rightsholder" groups that nominally represent both publishers and authors are in fact substantially biased toward publishers vis-à-vis authors, largely because of publishers' greater centralization and vastly greater financial resources. For example, IFRRO, the largest and most diverse international umbrella organization of "rightsholders", allocates votes according to member organizations' financial contributions to the organization, and has only a single token representative of authors' organizations on its Board of Directors.
Publishers and their trade organizations, not authors, are those with the resources for lobbyists and for attendance at public meetings and private consultations with governments at which theirs are the voices which purport to speak for "rightsholders". It's understandable that librarians or public-spirited politicians who don't hear from the (nonexistent) authors' lobby mistakenly believe that the publishers who claim to speak for "rightsholders" represent our views.
Even among authors, there is a diversity of interests. Those authors who are most successful under the current publishing system are, naturally, those with the least interest in changes that might risk their current livelihood under that system. As a result, those authors whose names are best known to the public, and who thus are most likely be believed when they claim to speak for all authors, are those whose interests are most closely aligned with traditional publishers and their current business models.
Authors are concerned that the defaults and the procedures for distributing licensing fees will end up favoring publishers over writers, thus making libraries and other users of "orphan" or "out of commerce" works unintentional accessories to publishers' longstanding attempts to steal electronic rights to backlist works they have published on paper.
A partial list of potential forms of favoritism for publishers over authors -- all of which have been included in one or another scheme for "orphan" and/or "out of commerce" works -- include:
- Default assumptions about whether electronic rights to work that was originally published on paper are owned by print publishers, authors, or both.
- Default assumptions about the formula for the division between the author and the original print publisher of revenues for usage or digital reproduction of "orphan" or "out of commerce" works.
- Decision-making and/or dispute resolution procedures (including evidentiary standards and burdens of proof) for allocating rights and revenues and resolving conflicts between print publishers' and authors' claims to rights. Even if the dispute resolution procedures are nominally equitable, allowing publishers to make blanket rights claims to their backlists by dumping catalog listings into a claims database, while requiring authors individually to dispute those claims to retain ownership or control of rights to our own work, can result in an unfair structural imbalance between publishers' and authors' interests.
- Default assumptions as to whether authors, publishers, or both hold the "right" to decide whether or not to participate, opt in, or opt out of collective licensing schemes. A scheme that "balances" authors' and publishers' interests by giving authors' and publishers' equal rights, when in fact they don't own equal shares of the underlying rights, is unfair.
- Payment schemes that channel all payments to authors through publishers, rather than paying authors our share directly (or paying all royalties to authors as the original copyright holders, and leaving it up to authors to pass on any revenue share we have assigned to publishers or other third parties). Authors do not trust publishers, and should not be required to rely on print publishers -- who may not hold, and may never have held, any rights to digital reproduction -- to pass on revenues to authors or decide what share of revenues to pass on to us. Failure by print publishers to pass on authors' contractually entitled percentage of e-book royalties is epidemic to the point of constituting the industry norm.
- Procedures for determining which works are "orphaned" or "out of commerce" that are more likely to mis-categorize works that are being exploited by writers (such as through digital self-publication) than works that are still being exploited by the original print publisher. Any process of "search" for the rightsholdings or the current status of exploitation based on records of print publication is inherently print-publisher-centric and likely to disproportionately mis-categorize work currently being exploited through commercial, revenue-generating digital self-publication.
- Searches for rightsholders conducted by parties with an interest in seeing those searches fail.
2. Misclassification as "orphaned" or "out of commerce" of works that authors are themselves commercially exploiting through digital self-publication
One of the most astute observations at the Berkeley sysmposium on orphan works came from Prof. James Grimmelmann of New York Law School, who commented almost in passing on the importance of considering who would bear the "error costs" of any "orphan works" scheme.
The more closely we look at what has actually been proposed, the more authors have become concerned that we will be the ones to bear a disproportionate share of these "error costs" of mis-categorization of our works as "orphaned" or "out of commerce" if they are no longer in print in the original paper format or from the original print publishers, even though we are earning revenues through digital self-publication or licensing of these works.
[One reader of this post coined the term "orphan snatching" to refer to this problem.]
A typical scenario is that people who want to buy a book that has been declared "out of print" by the publisher of the original edition on paper, or an article by a freelance writer from a back issue of a periodical, track down the author and ask, "How can I get a copy of your book or article?" The author may not expect her personal backlist to generate much revenue, but it requires only a modest one-time investment to convert the work to HTML and put it on the author's Web site (where it generates revenue through advertising) or publish it as a book or article-length e-book available through the author's Web site or other channels.
Rights to electronic reproduction for most older works were never assigned by authors to print publishers. Electronic rights to many newer works reverted to the authors when the works went out of print in the original paper editions. Writers who hold the electronic rights to their own work are under no obligation to tell (former) publishers of out-of-print paper editions about their new e-book or Web versions of the work. Publishers have no right to know, no reason to expect to know, and in fact have no idea how many of their out-of-print backlist titles have been reissued in digital (e-book or Web) form by the authors of those works.
Because the author typically receives only 5-10% of the price of a paper book issued by a traditional publisher, and 50-100% of the revenues from a self-published digital edition, the author may earn as much or more per copy from a $2.99 self-published e-book as from a publisher's $20 paper edition. And because demand for written works is highly price-elastic, writers are often surprised to find that unit licenses of inexpensive new editions far exceed the unit sales of the original print editions, and thus that their income from digital editions of their personal backlist exceeds the royalties they receive from publishers of their "in print" titles.
Incremental Web advertising and e-book revenues generated by personal backlists make a crucial and rapidly growing difference in enabling some long-struggling writers to make a living, quit our day jobs, survive financially after we are no longer writing ("self-employed" freelancers typically have no pensions), or simply continue to be able to afford to devote time to writing.
There is no database or single source that can be consulted to determine whether a particular work found in paper format in a library has been reissued in some digital format by the author, possibly in ways that are generating significant revenue for that author.
The nature of the freelance and self-publication business models for writers makes it likely that in many such cases, a database-driven and print-publication-centric search for "rightsholders" will determine that the work is out of print in the original paper edition, and fail to find that it is available in some form through the author's Web site or some other digital source or distributor.
Default "Extended Collective Licensing" (ECL) might work for a small, relatively homogenous country like Sweden with different publishing norms and labor laws. Perhaps most works published in Swedish are written by Swedish citizens and residents and published in one country, Sweden. Perhaps it's easy to deposit a copy of a Swedish work, even one published online, with the Swedish government for registration. Perhaps most Swedish writers belong to a small, well-defined group of Swedish writers' organizations and/or collecting societies. I don't know. I'm not a Swedish writer.
In the US context, however, all of these assumptions that underlie the EU and similar "orphan" and "out of commerce" and ECL schemes break down:
- Writers' self-exploitation of our work through self-published e-books, or even more through Web publication, is likely to be "under the radar" of publishers or database-driven searches. Limited exceptions to copyright through national laws, such as provisions for "fair use", use of "orphan works" or "out of commerce works", or other default "extended collective licensing" schemes, are permissible under Article 9 of the Berne Convention only if they do not "conflict with a normal exploitation of the work". But none of the major orphan works proposals to date, including the European Commission proposal, take any account of any form of self-publication in their definitions of "normal exploitation". On the contrary, they assume explicitly or implicitly that there is always both an author and the publisher, denying the reality of unmediated writer-to-reader distribution.
- Most digital publication, especially Web publication, is not reflected in copyright registrations in the US. I recently spent two and a half years trying to register copyright in this blog, only to be told this month by the examiner at the U.S. Copyright Office assigned to my application that the only way to register copyright in a blog is to register each day's articles separately, for a separate fee, which would cost $12,775/year in copyright registration fees alone for a blog that's updated daily (in addition to the time to process the registration claims). I asked the Register of Copyrights, Ms. Maria Pallante, about this at the Berkeley conference, and while she expressed sympathy for the problem, she was unable to offer any timeline for improvement in the possibility or process for registration of copyright in dynamic Web content. Copyright registration or "legal deposit" simply isn't a realistic option for most works first published online in the US today. FWIW, that's one of the major reasons this blog is hosted and "published" in Canada even though I live in the USA. But that's a poor workaround, not a solution to the problem of "legal deposit" for work published online, especially in the cloud.
- Most writers in the USA don't belong to any collecting society or organization of writers, and are listed in no published directory of writers. Works by writers who are citizens of, and live and work in, countries around the world, are routinely published in the US.
- And vice versa: Works by US writers are routinely published in other countries around the world, often simultaneously with their publication in the US. In particular, many books first published in the US are simultaneously first published in the UK and/or Canada, with the place of first publication listed on the title page as, e.g. "New York and London". If a search for the author is only conducted in one of the countries of first publication, such as the UK, many US authors will find their work deemed orphaned in the UK, and thus the EU, even though they are actively exploiting their own work commercially online, in the cloud, or in the US.
Would a specific requirement for a search for the author(s), independent of any search for the publisher(s) or other potential rightsholder(s), cure this defect in the European Union or similar proposals for "orphan" or "out of commerce" works?
Possibly -- I'm honestly not sure -- but certainly not unless:
- The search for the author would be individualized and not limited to publication records, copyright or "legal deposit" registries, collecting societies, or other databases.
- One of the resources required to be consulted in each such search would be a unitary global "opt out" and contact information registry of authors and pseudonyms. The International Standard Name Identifier (ISNI) database being developed by IFRRO and others will not, as currently envisioned, be such a registry. It's an identity management scheme and system of numeric identifiers, but one with no provision for authors to apply for or manage the identifiers assigned to us by third parties. Obviously, authors are those who know which pseudonyms we may have used, and which works we have written and which have been written by other people with the same name. But ISNI's would be assigned exclusively by libraries, publishers, licensing agencies, and other intermediaries. There's no requirement in the ISNI standard for the individual being "identified" to be consulted before an ISNI is assigned, or even notified after one is assigned. It's a recipe for authorial identity theft and misidentification of authorship. (Meaningful opportunity for authors to opt out of default licensing of their work, in advance of its being made available, is a necessary addition to, but not sufficient to redeem, the EU, French, or similar schemes.)
- The search for "commercial availability" of a work would be required to include a search for normal forms of commercial exploitation of works including digital self publication through authors' and third-party Web sites and self-published e-books, and such publication methods would be explicitly recognized as forms of "normal exploitation" as that term is used in Article 9 of the Berne Convention.
- The search for "commercial availability" of a work would be a search for availability of the work in any form, not limited to a search for a particular edition, format, or instantiation of the work.
- The search for the author would be conducted worldwide, not limited to any one country or by any determination of the country or countries of first publication of the work.
- The search would be conducted by an entity with an interest in the success of the search rather than by an entity with an interest in failing to find the author. (Libraries, despite the best intentions, would have a structural interest in not finding the author, since not finding the author would allow the imposition of default standardized licensing terms without the need for individualized negotiations with the author to acquire the desired rights.)
Like most writers, I want to see my work more widely available, and I'm doing what I can to make it so. But I don't want to see those writers who have done the most to revive the availability of works abandoned by their original publishers be the ones whose livelihoods are most undercut by new library licensing schemes.
I don't think that's what librarians want, either.
[Update: An op-ed by Prof. Pamela Samuelson published in the L.A. Times after the symposium exemplifies the problems I and the NWU have raised, and the lack of attention paid by proponents of "orphan works" schemes to the concerns of working writers.
In her op-ed, Prof. Samuelson cites a "broad consensus" on "orphan works", but fails to note the absence from any such "consensus" of working writers or groups like the NWU, who have not yet been included in the discussions or the "consensus" among those who want to use (but excluding those who created and who own the rights to) works which might be classified as "orphan works".
Prof. Samuelson's claim that, "These books aren't producing any revenue for copyright owners, and most of them are unlikely to be reprinted," ignores the realities of working writers' business models and the issues raised in the NWU white paper and the blog post above. In fact, many books are likely to be classified as "orphan works" in spite of the fact that they are generating revenues for their authors through digital self-publication as e-books or on authors' Web sites. And growing numbers of these books are being digitally "reprinted" and made available by author/rightsholders, often in ways not reflected in any of the bibliographic databases or registries being used to for searches for rightsholders or the status of exploitation of works.
Finally, Prof. Samuelson suggests that because similar laws have been enacted in France and are being considered in the European Union and elsewhere, the US should do likewise. But objections similar to those raised by the NWU in the US about how an "orphan works" scheme would favor publishers over authors have been raised by French authors in response to the new French law on orphaned and out-of-print books. Perhaps the US should learn from, rather than repeat, the mistakes made in drafting the French law and the current EU proposals without involving working writers or considering the impact of such schemes on our ability to afford to devote time to creating new written work.]