Mass Digitization

Copyright owners probably thought they got off to a good start legislatively: they took bold steps 10 years ago to strengthen their rights, fearing an erosion of control in the digital environment. Unfortunately, they designed their two major pieces of legislation, the Digital Millennium Copyright Act and the Sonny Bono Copyright Term Extension Act, to protect them from the future, rather than to help them embrace it. Those who embraced a different vision of the future were slow to respond, perhaps slow to recognize the profound effects of the laws that were being proposed, but have managed in the wake of these two monumental enactments to prevent further deleterious modifications of the law. This is the stalemate of moneyed interests I referred to earlier where, for example, the device manufacturers counter content industries and defeat each other's initiatives.

On the other hand, efforts to roll back protection, even when it seems in a majority of the industries' best interests, can be stymied by even a single copyright industry segment that effectively persuades our lawmakers that catastrophe will result if the law passes. The 2006 failure of orphan works legislation that would have enabled use of works whose copyright owners cannot be identified or located is a good example of this phenomenon. It seems very likely that a second attempt this session to pass this legislation will be similarly doomed.

So in most cases we have legislative stalemate. I include myself among those who believe this is actually a good thing. Bill Patry, copyright treatise author and counsel to Goolge, remarked just this month at the "Who owns this image" panel discussion in New York City, "Don't look to Congress for any help. If they can solve orphan works at the 1% successful level, it will be a bonus. Any solutions will have to be private" (Tushnet, 2008). Congress' inaction allows individual actors in the market to shape the role of copyright, and it may not play the same role in the future that it had in the past!

It has been observed that the copyright industries' success in passing the Digital Millennium Copyright Act and Copyright Term Extension Act actually undermined their ability to adapt by giving them tools to resist the inevitable, if only for a time (von Lohmann, 2007; Lehman, 2007 as reported in Geist, 2007). But reality has a way of patiently waiting resistance out, and most content industries are beginning to accept the realities of the digital networked environment, and are now enjoying at least some success pursuing alternative business models that do not rely on stringent enforcement of copyrights in digital copies or digital rights management (DRM) (Arrington, 2007; Amazon.com MP3 Downloads, 2008; Levine, 2007; but see Edgecliffe-Johnson, 2008). It will become more difficult to make the case that copyright owners need this protection (though they will not likely stop pressing the case as the 2007 Intellectual Property Enforcement Act introduced in the Senate demonstrates), because the evidence increasingly will not support them.

Resistance to resistance. Steve Jobs had to work very hard to legitimize consumer demand for relatively freer access to and use of digital media (Taylor, 2003). The music giants may be chafing under the constraints of his preferred deal structure, and struggling to break his strong hold on business model evolution (Krazit, 2007; Lee, 2008), but they owe him much for getting them off the starting line, and for later challenging continued use of DRM (Jobs, 2007). Television networks seem determined not to make the same mistakes and have begun to offer their programming in relatively freer formats than they were willing to consider just a few months ago (McIntyre, 2008; Nakashima, 2008; Goodman, 2008). Full-length feature films are still heavily protected by DRM, but they seem to be in a class by themselves given the astronomical costs associated with production. At this point, however, it seems that anything is possible, even DRM-free feature-length films (4flix.net, 2008; Good Copy Bad Copy, 2007).

Positively paring down. The Creative Commons has made it possible for millions of copyright owners to pare down their own set of rights (Doctorow, 2007), for hundreds of millions of items of Web content, in the absence of Congressional recognition of the inappropriate scope of copyright protection for the overwhelming majority of works created today. Studies of how creators use Creative Commons licenses tell us much about the nature of expectations regarding a reasonable scope for copyright (Kim, 2005). Congress would do well to pay attention to this kind of evidence.

De facto efforts to establish orphan works status - the copyright evidence base. Several corporate entities (OCLC, Google and Internet Archive), the Copyright Clearance Center, and individual libraries (Proffitt, 2008; free the books, 2008; UT Austin's Watch, 2008; University of Michigan's public domain search and its public domain humanities resources, 2008), among others, are building tools that help people locate copyright owners to obtain permission for uses, and that ultimately facilitate individual assessments of the risks of placing some protected works online, even in the absence of clear rights to do so.

Orphan works are those books, records, images, compositions, manuscripts, movies, screenplays, paintings and drawings -- in short, any work protected by copyright -- whose owner cannot be determined or located, or who does not respond when contacted. There have always been orphan works, but a number of factors have converged to turn their existence into a significant lost opportunity. In the past, who really cared? Orphan status may have been unfortunate, but for the most part, it was just what happened over time because of the relatively short productive life of most published works (Report on Orphan Works by the Register of Copyrights, 2006). The need to exercise the rights of the copyright owner of an out of print or otherwise unavailable works, that is, to make copies, create derivative works, display, perform or distribute an older work publicly, beyond the rights authorized in the Copyright Act, typically did not arise. And, works eventually became part of the public domain. Not necessarily so today. Now, extremely long copyright terms frustrate our desires to digitize library, archive and museum holdings, and offer access to them to as broad an audience as possible. The result is that the orphans with no owner to speak up for them will suffer a fate equal to death -- the obscurity resulting from their inability to be "permissioned," digitized and displayed.

As has been widely documented and reported, the vast majority of books on our shelves are still protected by copyright, but their copyright owners are unreachable (Lessig, 2007). The cost of misidentifying an orphan is extremely high: a copyright owner who registered her copyright can ask a court to award up to $30,000 per innocent infringement ($150,000 for willful infringement). These extreme penalties are meant to deter infringement, but for orphans, they deter public uses that no one will object to, by definition (there's no one there to object). Nevertheless, a small chance of being wrong, and a high penalty if one is wrong, makes for a risky situation. So what's going to happen to all those orphans? The promise of new life on the Internet dangles temptingly for them, but only very confident collection stewards will take this risk and digitize and display them.

And, I believe, that's just what we have among today's librarians: a growing number who are cautiously confident about their abilities to assess and manage risk.

The photographers' successful campaign to destroy the chance for effective Congressional legislation regarding orphan works in 2006 (Urgent Call for Your Action on Orphan Works, 2006) drew the line in the sand. Either those who care whether a staggeringly large part of our cultural record ever makes it off the shelves, out of the boxes and onto the Internet, are willing to step over or they are not. Enormous numbers of works provide no signs of who the author is, who the publisher was or when the work was produced or distributed. If a work is published without proper notice (name of publisher and date) during certain time frames (1923 - 1989), it becomes a part of the public domain. If it is not published, or if it is published after 1989 without an indication of who its author is, its protection is automatic and lasts for the life of the author plus 70 years in the U.S. (and longer in some countries). But if you can't determine the author, the date of publication or the date of creation, you cannot know for sure when such works enter the public domain. The obscurity we consign these works to is not a short-term condition. It lasts until someone has the courage to just put the work online. The opportunity cost of doing nothing is now unacceptably high.

The heart of the 2006 proposed but failed orphan works legislation, and the current version before Congress, is a limit on the remedies available to copyright owners whose works are mistakenly characterized as orphans. Such limits would certainly encourage the use of orphan works because the current remedies are so extreme. The other critical aspect of the legislation is the effect of a reasonable search for the copyright owner. A reasonable search entitles a user to the limited remedies if an owner appears after the work had been used. But as a practical matter, assuming one does conduct a reasonable search, the chance of mistakenly classifying a work as orphan might be pretty slim (depending on the quality of your search). That's the theory, and as we might imagine, there is a lot of disagreement about what a reasonable search is and how much detail should be spelled out in the statute.

Librarians tend to be on the risk-averse side, in general, but in this case, as information specialists, they may have more confidence in their searches for a copyright owner than the average person. Nevertheless, the Register of Copyrights recommends and the industries support the idea that the industries should define reasonable searches for their respective types of works (cites). Though this sounds logical, corporations tend to think in terms of corporate resources that can and probably should be brought to bear on the question of whether there is a contactable copyright owner, in light of an objective motivated by profit (STM Assn., 2007). Applying the same criteria to nonprofit public access uses is not realistic. Nonprofits have neither the resources to make exhaustive searches, nor the profit motive to justify the expenditures of resources.

Until (if ever) we have legislation that reduces the enormous penalties for being wrong, there is considerable pressure to get the "reasonable search" for an owner right, so that the reasonable search becomes our insurance against catastrophe. This calls for collaboration. For-profit and nonprofit companies and libraries, working together, are beginning to build an evidence base, at first dedicated to identifying copyright owners, but then to freeing the books that are in the public domain, but not currently properly identified as such. Next in line are the orphans. The protocols we develop to determine whether a work is properly copyrighted (notice) and whether the copyright was renewed as required, and then who owns the copyright, merges imperceptibly into an inquiry into when an author died, whether and where he left heirs, whether a publisher is out of business, where the last place of business was, and so forth. Participants in this project will document sources of law and the factual information about each book, author and publisher, document results and publicly display the process and the findings so that others may build upon what has been learned. In the U.S., evidence about a particular publisher's disappearance identifies not just one orphan book but every book whose copyright that publisher owned. When copyrights are owned by individual authors, evidence that an author left no estate or heirs, or that they cannot be identified or located, identifies as orphaned every book she ever wrote. So it is important to publicly cumulate this evidence and build not only our ability to find copyright owners, but to rebuild our public domain, and identify our orphan works. Slowly but surely, collaborators will put together the evidence we need to feel confident that, even if we must face draconian penalties, we can reduce the risks to manageable levels.

There is additional evidence that other institutions are also beginning to wade into these waters. Brewster Kahle, long committed to bringing public domain works to a world audience, was reported in hangingtogether.org (OCLC) to have encouraged participants at Open Content Alliance's (OCA) annual meeting (2007) to take the next step with a pilot-project that will "start digitizing out-of-print/in copyright works, a departure from the strictly public domain digitization in the OCA to date" (Waibel, 2007).

Brewster could be relying on fair use for the digitization step, but if he plans to offer these works to a public audience, he would be going much further, perhaps claiming these works as orphans. If Brewster is considering making orphan works available, he would be grappling with the same issue described above -- establishing the contours of the reasonable search that establishes the likelihood of orphan status. Perhaps the project seeks permission, and in the failed attempt, determines that a work is an orphan. This will be a project to watch.
The copyright notices that accompany some digital collections provide additional evidence of risk assessment: it is not uncommon to see works included in digital collections accompanied by notes such as one on the Copyright Office's American Memory Website. The note explains that the use of images in the American Memory Digital Collection does not mean that the images are freely usable by anyone for any purpose. Visitors are cautioned to make their own copyright assessment before engaging in any but fair uses of the images. These notes suggest that the Library has conducted its own risk assessment and taken reasonable action based thereon.

Here again, however, these actions highlight the unfortunate consequence of Congressional paralysis: potential copyright irrelevance. The pronouncement by then Patent Commissioner, Bruce Lehman, that, "[w]ith no more than minor clarification and limited amendment, the Copyright Act will provide the necessary balance of protection of rights -- and limitations on those rights -- to promote the progress of science and the useful arts," turned out to be quite wrong (introduction to the 1995 NIITF White Paper, p. 17). In marked contrast, he confidently dismissed opponents' concerns that turned out to be deadly accurate, among others, that fighting technology would result in massive disrespect for the law and its ultimate irrelevance (p. 15) -- a high price to pay for codifying industry fears.

But people are moving forward: copyright owners are confronting their fears of financial loss; other copyright owners are trying trimmed down copyrights for their digital works; and libraries are confronting their fears of being sued for displaying digital images of works on their Websites -- all in the space created by Congressional inability to take any meaningful action to reign in the laws it passed a decade ago, whose lack of utility is becoming obvious to many, if not to those with the power to change them.