Orphan works, fair use and best practices July 30, 2009
Posted by Kevin Smith in : Copyright Issues and Legislation, Fair Use, Open Access and Institutional Repositories , add a commentAll of the above are recurrent themes in copyright and scholarly communications these days, but a recent publication from the Society of American Archivists has put a little different spin, I think, on an ongoing conversation.
The SAA released a revised version of their Statement of Best Practices on Orphan Works on June 17. In the statement about the purpose of the report, the SAA makes specific reference to the two bills that were considered by Congress in 2008 as attempts to solve the orphan works problem (I blogged about those bills here and here). The revised statement of best practices is an explicit attempt to define a term that was used in those bills — a “reasonably diligent search” for a copyright holder. It would be only after such a search that a remission of the damages for a user of an orphan work would be available under these bills, and the SAA is trying to suggests standards and practices that define what is reasonable and diligent in the real world of archival materials.
It is important to realize that there are two different approaches to using orphan works. The bills proposed in Congress take a remedies-based approach, offering a substantial reduction of the possible penalties for users of orphan works if they first undertake a reasonably diligent search and, subsequently, a rights holder surfaces and demands compensation. The SAA statement of best practices is directly related to this approach and undertakes to define the steps necessary if such a search requirement is enacted. But the statement of best practices also recognizes another option, reliance on fair use. The statement says “Fair use may be a better rationale for creating a copy or publishing a copy of a document,” but it does not make an explicit connection between fair use and the best practices outlined in the remainder of the statement.
Fair use is an exception to copyright’s monopoly that already exists and is currently available to potential users of orphan works. The value of the “reasonably diligent search” in the fair use context is that it would have, I believe, a profound effect on the fourth fair use factor, the impact on potential markets for the work. If a search such as is suggested in the SAA statement is carried out and no rights holder can be located, that would go a long way toward showing that no market is being harmed by the use (especially if the use itself is educational and non-profit). In this situation, it is hard to imagine a court actually rejecting a fair use defense, and even if such a defense did fail, archivists and other employees of non-profit institutions could still fall back on the partial remission of damages that is provided in section 504(c)(2) of the Copyright Act. As the SAA notes, a reasonable belief in fair use, even when a court disagrees in the end, “is sufficient to protect the archivist from statutory damages.” Such protection is not as complete as would be provided by an orphan works bill, but it is is nonetheless substantial. In the end, it really might make more sense for educational users to rely on fair use when contemplating a use of an orphan work, after employing some or all of the strategies in the SAA statement of best practices to try and find a rights holder. Waiting for Congressional action may be both impractical and unnecessary.
Whether orphan works legislation proves useful or not will depend in large part on the details of any final bill. There were strong hints last time that in order to gain approval, a bill would become so burdensome and expensive that the library and archives community would be better off without new legislation, simply relying on fair use. No doubt that debate will be revived if any orphan works bills are re-introduced. But the SAA has made an important contribution from either perspective that one takes. In regard to potential legislation, they have offered a standard that legislators should consider as they draft a bill, as well as one that those concerned about the burden created by legislation can look at to measure the depth of the problem. In regard to those who would rely on fair use, the statement of best practices provides a set of guidelines that can help give users confidence that they are truly making a good faith fair use effort.
Books in the cloud July 26, 2009
Posted by Kevin Smith in : Copyright Issues and Legislation, Technologies , 2commentsThere have been lots of reports flying around recently about the decision by Amazon to delete copies of two works by George Orwell (ironic, that) from the Kindle devices of folks who thought they had bought those books for once and for all time. There on reports and comments about this here, here and here. Technologically naive as I am, my first response to the story was “wow, I didn’t know they could do that.” My second response was to reconsider my growing inclination to buy a Kindle.
The really meaty issues, of course, are not whether I buy a Kindle or not, but relate to copyright issues and privacy. We are used to the idea that a copy of a book is mine to do with as I please once I have purchased a legally-made copy of it. The same is true of a CD or a DVD; I can rent, donate, destroy, lend or resell the single physical copy that embodies intellectual property because of the provision in our law called the doctrine of first sale. We know that first sale does not necessarily apply in the same way to digital files, but the Kindle cases really pushes the issue. Everything about the purchase of an e-book from Amazon looks like a sale, and consumers can easily be forgiven for thinking that they own something at the end of the process. It is probably time for a legislative look at how first sale applies in the digital world. While there have been suggestions of a “forward and delete” model for digital first sale that would allow a consumer to transfer a digital file to someone else as long as copies were not multiplied, this situation raises a more fundamental question. When can a transaction that looks like a sale be treated as a mere license, and when will consumer protection concerns step in to enforce the privileges that go with a purchase?
It is important to note that, in the case of “1984″ and “Animal Farm,” there was no question of preventing consumers from making unauthorized copies, which is the usual reason given for the assertion that first sale does not apply to the digital environment. Here, it was Amazon that was selling the unauthorized copies, and consumers were deprived of ownership by remote action, even after they had purchased the books. Why this could not be remedied by having Amazon pay the rights holder for the infringement, which would have been the solution if a publisher distributed print copies of a book without authorization from the copyright holder, is not clear to me.
One thing that several of the stories about this contretemps have in common is reference to Jonathan Zittrain’s must-read book “The Future of the Internet… and how to stop it.” It is always good for an author when he correctly predicts a technological trend, and Zittrain got this one dead-on. His warning that the Internet is moving away from the programmable devices that fostered so much innovation toward tightly-control, “tethered” appliances proves eerily prophetic when Amazon starts deleting books from consumers’ devices. It makes reading Zittrain’s discussion of all of the implications of this development that much more important.
Zittrain had an excellent op-ed piece in the New York Times on July 20, called “Lost in the Cloud,” that discusses some of the privacy and censorship issues that are inherent in the development of these Internet appliances and makes brief reference to the Kindle issue. I am happy to be able to report that Zittrain will be coming to Duke during the upcoming academic year as part of our Provost’s Lecture Series; I cannot imagine a more important discussion to have than one about the issues he is raising.
The elements of an open access quiz July 22, 2009
Posted by Kevin Smith in : Open Access and Institutional Repositories, Scholarly Publishing , 2commentsWhen I was a first-year law student, my professor for Torts used to threaten to call us up at 3 am and demand that, before we were fully awake, we be able to recite the elements of a negligence claim – duty, breach, causation and harm (thanks, Prof. Darling). I was reminded of this demand by a small part of a recent news story, and the thought of three “elements” about open access that I would like to see every member of university promotion and tenure committees remember, even if quizzed in their sleep.
The news story, from the Chronicle of Higher Education, reports on an unusual tenure process at the College of New Jersey. The Dean and faculty panel recommended against granting tenure to Professor Nagesh Rao, but the Provost and Board of Trustee disregarded that recommendation and granted tenure, after considerable internal and external protest. My interest in the story is focused on one small comment, where Professor Rao is describing the reasons he thinks the faculty panel recommended against tenure. In addition to mentioning that his subfield may be subject to some bias, he says that one of the principal places where he is published, an open access online journal called “Postcolonial Text,” may have been “arbitrarily devalued” due to its business model.
For the record, “Postcolonial Text” is a peer-reviewed journal published on the Open Journal Systems (OJS) platform. I recently published (shameless plug alert) an article on open access for theological studies in an OJS journal, and can testify that the peer-review process — which is determined by the editors, not by the publication medium — was as rigorous as any traditional publication I have experienced. We have reached the point, I think, where the notion that online or open access is somehow not as scholarly as print, toll-access publication is no longer a reflection on open access itself, but is an indication that some academics have simply failed to pay attention to radical changes in the environment for scholarship. If what Professor Rao says is true, it is shows an embarrassing ignorance on the part of the panel that evaluated him.
So what are the “elements” of open access I want everyone who is responsible for evaluating scholarship to be able to recite, even when awakened in the dead of the night? They are as follows:
1. Online open access journals are as likely to be peer-reviewed as are traditional print publications. The medium cannot be used as a surrogate for investigation into the editorial practices and personnel of a given forum.
2. Open access based on an author fee is not a form of vanity publishing, and these arrangements, which are usually traditional journals with an open access option added on, are peer-reviewed in precisely the same way as traditional publications in the same journal. They should be weight in an evaluation process in exactly the same way.
3. Many, perhaps most, works which an author self-archives in an institutional or other repository are also published in peer-reviewed forums. P&T committees should not dismiss works just because they can be found in an open access repository, and authors should be responsible for ensuring that sufficient metadata accompanies the article to tell anyone who finds it about its peer-review and publication history.
That pesky checklist July 19, 2009
Posted by Kevin Smith in : Copyright in the Classroom, Fair Use , 5commentsThe recent flurry of activity in the copyright infringement lawsuit brought by publishers against Georgia State University has focused attention – mine, at least – on the “Fair Use Checklist” that has been adopted for use in quite a number of college and university copyright policies. As part of the mini-controversy over the naming of Dr. Kenneth Crews from Columbia University as an expert witness for the trial, the plaintiffs have objected that Dr. Crews, as a co-author of the checklist that is part of GSU’s new policy (see a previous post on this topic here) cannot be an impartial witness. In one sense this seems an odd objection, since experts are hired by each side in a lawsuit precisely because thy favor the position taken by the party that hires them, but it also offers a chance to reflect on the use and misuse of the fair use checklist and to begin to explain publishers’ ambivalent attitude toward it.
There are two obvious problems with the checklist, it seems to me. First, it can encourages a falsely mechanical view of fair use, where a “score” of seven pro versus six con, for example, means something is definitely fair use, while a one-digit reversal means it is not. That, of course, is not how fair use really works, and no score card can actually predict the results of a judicial evaluation of the fair use factors. Second, the checklist would be pretty easy to manipulate so that it tends toward the result someone is seeking. There has been some discussion, for example, about whether or not there needs to be an equal number of check boxes on each side (favoring fair use v. disfavoring fair use) in order for the checklist to itself be fair. Although this seems plausible, it is important to remember that courts have not necessarily articulated an equal number of circumstances to be considered on each side of the argument, and the checklist seeks to guide its user through the considerations that are actually in play, not some artificial list created without regard to case law for the sake of balance.
Against these two problems, both of which can be quite real, there are also a couple of sound reasons for using the checklist. First, the very mechanical nature that makes it an imperfect tool also makes it one that can be used quickly and without an entire course in copyright law by staff and faculty. These are the major groups that need to make fair use decisions day in and day out; the checklist is a way to at least be sure that they think about all of the factors that are relevant. There are many people on college campuses that seem to believe that any educational use is a fair use, and the checklist helps counter that simplistic belief and remind all of its users of the full-range of necessary considerations. Second, the checklist provides documentary evidence that a full fair use analysis was undertaken. Since part of the “remedies” section of the copyright act gives college and university employees partial protection from damages for infringement when they make a good faith fair use decision, even if they turn out to be wrong, evidence of detailed analysis helps protect the institution from potential liability.
These two arguments in support of using a checklist may help explain the ambivalence that the publishers have shown toward its use. The Association of American Publishers has announced support for several university policies that include the checklist, including Cornell’s and Syracuse’s, but they have lately seemed more hostile towards it. It is easy to see why, really. On the one hand, it is in publisher’s interest to have university employees get beyond a simplistic view of fair use, which is usually too generous, and look more closely at the full range of considerations that need to be taken into account (this explains, I think, the use of a version of the checklist by the Copyright Clearance Center as well). On the other hand, that deeper consideration will, itself, make universities less attractive targets for litigation, which seems to be the chosen weapon in the battle to narrow educational fair use.
I have to admit that I too feel a good deal of ambivalence toward the checklist, albeit for somewhat different reasons. I would like every staff and faculty member who must make fair use decisions to have a complete and nuanced view of the doctrine they are applying. But I recognize how impossible that is. Until our campuses are populated entirely by IP lawyers (may that day never come!), I will continue to believe that the fair use checklist is a highly imperfect, but even more highly necessary, tool for navigating the traitorous waters of contemporary fair use.
A sequel on Salinger July 15, 2009
Posted by Kevin Smith in : Copyright Issues and Legislation, Fair Use , 3commentsIn a comment about my previous post concerning the decision to grant a preliminary injunction in the Salinger case, effectively banning US publication of a story about Holden Caulfield 6o years later, the writer took me to task, quite harshly but with some justification, for my cursory and sweeping denunciation of the decision. My post implied, although I admitted at the time that I had not read the entire ruling, that the Judge’s decision was uninformed. Now that I have read the opinion, I must admit that it is not uninformed, but wrong for several other, more subtle, reasons.
First, as I said originally, Judge Batts does not adequately address the idea/expression dichotomy built into copyright law in section 102(b). She acknowledges the concept early on in the ruling, even citing it as one of the elements that reconciles copyright with the First Amendment, but then does not mention it again. The result of this omission is that she never tries, in her analysis of the alleged similarities between “Catcher in the Rye” and “Forty Years Later” to distinguish protectable from non-protectable elements before making the comparison. This is standard practice in the substantial similarity analysis, and its absence leads the Judge to cite some amazing things as evidence of copyright infringement, such as the fact that both characters are liars, both complain a lot, and both travel to NYC.
It is certainly true that one can infringe in ways other than verbatim copying, but one must still show substantial similarity with protected expression; it is not enough to say that the protagonist in Catcher is delineated in words and therefore any word picture that is similar, even merely because it uses similar ideas, must infringe. I am reminded of the case brought against Dan Brown for infringing the copyright in “Holy Blood, Holy Grail” with his “The Da Vinci Code.” The basic idea in the two books is fundamentally the same, but the case was dismissed (in the UK) because the court separated out unprotectable ideas and then found no substantial similarity. That should have been the procedure, and the outcome, in the Salinger case as well.
I think a good deal of the problem is that the Judge spends too much time on the discussion of whether or not the new work is a parody of the old. Even when she turns to other transformative purposes in her analysis, the Judge never really moves past the parody issue, since she focuses entirely on the (critical) use of J.D. Salinger as a character in the new novel. But there are many kinds of transformative uses beyond these two types (which are really not very different); the Judge needed to consider the broad range of purposes listed in section 107 as examples of fair use in order to evaluate the full scope of possible transformation.
A clue that there is a problem with the Salinger ruling is that Judge Batts finds pretty explicitly that no unauthorized sequels would ever be non-infringing, but she offers no guidance at all about where the line is between impermissible sequels and other types of work, such as reference works that describe and analyze characters, settings and plots. Essentially, the exclusive right over derivative works has been allowed to swallow the idea/expression distinction that is really vital to American copyright law, and the Judge offers us no way to mitigate or reconcile that result. In fact, she tries rather feebly to make something positive out of her error here by suggesting that the right to prevent a sequel might actually give an incentive to an author to write a novel, secure in the knowledge that they will have the last word about every thought or idea they express. Even if one assumes lots of authors as unsocial and controlling as Salinger, this is not a recipe for promoting the arts, whatever the Judge tries to argue.
This severe limitation on future creativity points up the last problem I want to cite in the Salinger ruling, and in many ways it is the most damaging. Even though it is part of the required findings for issuing a preliminary injunction, Judge Batts never considers the balance of hardships between the parties if she grants the injunction, nor does she weigh the public implications of this prior restraint of speech, except with this odd argument about incentivizing authors who cannot tolerate followers. These considerations are not optional when ruling on a preliminary injunction, and their absence should lead to reversal at the appellate level. But even more important than following the letter of the Federal Rules of Civil Procedure is recognizing that it is absolutely vital that the public interest be considered whenever a judge is considering a request to ban a book, lest he or she run roughshod over the Constitution they are sworn to uphold.
The real problem with this ruling is not that it is uninformed, but that it misses the forest (including the public interest) due to its nearly exclusive focus on one tree.
