Compulsory licensing is an idea that has been getting a lot of attention recently. All of the compulsory licensing schemes I've seen, if adopted, would have far reaching effects on our system of copyright. The issues at stake aren't simply how to raise and distribute funds fairly. Such major changes to copyright law will inevitably and significantly affect great swaths of our social, economic and legal institutions. Some of these effects will be positive, some will be negative, and some will simply be ... different. Even if we can satisfactorily address questions regarding the raising and distribution of funds, we must also consider these other issues and try to tease out the consequences of such proposed changes.
One question we must think about in some depth is what effect the proposed changes will have on the public domain. It would be wrong to simply assume that compulsory licensing will be wholly beneficial to the public domain. Although compulsory licensing will likely have many beneficial aspects with regard to the public domain, different systems will have different effects and unintended consequences. For example, it is possible that Harvard Law Professor Terry Fisher's draft compulsory licensing scheme (Chapter 6 of Promises to Keep: An Alternative Compensation System [PDF]) might create what I call a "public domain lottery" effect. This effect is not necessarily positive or negative, but it is ... different, something unusual that might occur when you make major changes in the current copyright regime.
Distribution and the Public Domain Today
One of the wonderful things about the copyright is it limits; the fact that it permits anyone to pretty much do whatever they want with works in the public domain, such as reproduce, modify and distribute them. Although people often focus on the modification aspect of the public domain, the ability to freely reproduce and distribute is very important, even if no changes to the underlying work are made. It may come as a surprise, if you accept the arguments of copyright maximalists, but it is possible to make a profit publishing and distributing works in the public domain. These profits can be substantial, as this New York Times article from Feb 2003 points out (Publishers Give Classics a Makeover [PDF] - thanks, Furdlog). Major publishers such as Penguin Classics, Modern Library and even Barnes and Noble get a significant amount of their business from the sale of works in the public domain.
This is wonderful. No one begrudges the profit Penguin Classics makes from publishing and distributing works in the public domain. Indeed, I think most would agree that Penguin should get some market-determined profit for going through the trouble of publishing and distributing the classics. And, as the NY Times article shows, there is vigorous competition in the classic books realm providing differentiated products to suit every price range and need. Everyone wins. The question is, what happens to this model were we to adopt Prof. Fisher's proposed compulsory licensing scheme?
Registering Public Domain Works
Now, I can't guarantee that what I'm about to describe will happen. Still, I think this scenario one likely possibility. Perhap Prof. Fisher's proposal can be tweaked if we think this outcome undesirable (which isn't necessarily obvious).
One could easily register a public domain work as one's own under Prof. Fisher's plan. Even if some modification were required to register such a work, it would be simple enough to add a brief introduction or afterword (as many classic editions of books do) or editorial comment to meet the requirement. Presumably, the individual who holds the registration for the registered version of the public domain work would then receive 100% of the share of funds allocated, based on how much the work is shared. One could imagine a distribution of funds in which modifications of public domain works received only a partial share of funds allocated, but such a system would have interesting consequences of its own.
So far, so good. The proposed system doesn't seem much different than our current system. Barnes and Noble can reprint "Huckleberry Finn" with no changes and receive 100% of the funds spent on the new edition. Few, if any, have a problem with this.
Only a Few Originators for Files
However, internet distribution is different, particularly in the P2P realm. Shared files have a tendency to overwhelmingly come from a few initial sources. For example, many thousands of people who participate in P2P MP3 sharing might actually own the physical CD for a particular track. Many of those may have even ripped and made the file available for sharing. Yet, it is highly likely that the vast majority of shared files of a particular track will have come from only a handful of the people who did the ripping.
This is primarily because the majority of people who share MP3 tracks aren't originators, but copiers who then go on to share those copies. Sharing wouldn't be nearly as widespread if most people were simply ripping their own collections and uploading them. People overwhelmingly join filesharing networks so that they can download files, not so they can upload them. What sharing is done is usually default. The defaults for filesharing programs generally make files that have been downloaded available for sharing, while content that already exists on the hard drive is much less likely to be shared as a default. If people are primarily downloading, and not uploading original content, it stands to reason that a single originated file will be copied and downloaded many times. Every time an original file is downloaded, it likely to be shared (the default), and thus it becomes more likely that it will be shared even more. It is a positive feedback loop. The more any particular file is shared, the more likely it is that particular file will be shared more, when people search for it. While P2P networks are not monocultures, popular files overwhelmingly come from just a few originating sources.
Of course, this says nothing about how any particular originating source becomes dominant. It is likely that being first to market is a significant advantage. The sooner a particular file begins to be shared, the sooner the number of hosts that carry that particular file will grow, generally leaving much less opportunity for late comers. Still, an early, poor-quality version of a file may eventually lose out to a later, better-quality versions of a file. Regardless, it is likely that only a handful of variations of a file will dominate sharing of a particular file.
The Public Domain Lottery
What this means is that those who distribute slightly modified, or even unmodified, versions of public domain works have the chance to reap substantial rewards, if their registered version of the public domain work becomes one of the commonly shared versions. I imagine that getting your work to be one of the commonly shared versions will be somewhat random, and the rewards likely to be disproportionate to the effort, thus a "lottery."
One possibility for how this might work is when an individual identifies a work in the public domain that is likely to become at least somewhat popular. For example, if Osama Bin Laden were to release another audio or videotape, it is likely that thousands of people would be interested in downloading and sharing it out of curiosity. Since Osama and his associates are unlikely to file a copyright registration form, it would be easy enough for someone else to add just a bit of editing or even a "bug" in the lower left hand corner of the video, and register the work. Everytime the file was shared, the holder of the registration would, presumably, receive an appropriate share of the revenue. For a file that might be downloaded tens of thousands of times, that reward could be fairly substantial, and would almost certainly be disproportionate to the value added by the early registrant.
What if what gets registered is the Daniel Pearl Video, which I am certain has been downloaded many thousands of times? (For more information on the video see, FBI Agents Intimidate Publishers of Daniel Pearl Video.) I am certain that there are plenty of people, similar to the notorious typosquatter John Zuccarini, who would be happy to profit (at taxpayer expense no less) off such videos.
Of course, not every public domain work would be subject to such "hijacking." Some, perhaps many or even most, public domain works might end up being widespread without being registered. Still, it is unlikely that all public domain works (especially given the incentive to capture rents) will spread unencumbered of registration.
A second, and even more profitable means for winning the public domain lottery would rest in the hands of the P2P search engines. If, as I've suggested previously (Compulsory Licensing - The Death of Gnutella and the Triumph of Google), there will be only a small number of P2P companies that dominate a completely legitimate P2P market, they will be in an excellent position to dominate the public domain market. The means will be very simple. In step one, the dominant P2P companies will have their own registrated version of every relatively popular public domain work (popularity will be easy to determine given the wonderful tracking and census systems being proposed).
Step two will be simply to ensure that the number one response to any searches for a public domain work will be the one registered by the P2P company. Looking for Mozart? The number one response will be the version of Mozart inexpensively commissioned (work for hire, of course) and registered by Grokster. This wouldn't even necessarily break the search function substantially, since the engine could easily distinguish when you are searching for a particular performance of Mozart (and returning results appropriately) and when you simply want some Mozart for background music (Grokster: ka-ching). Looking for some cool Space Shuttle footage? Most of the stuff shot by NASA is in the public domain. A few small changes, and suddenly, the number one result for Space Shuttle footage is Grokster's registered version.
Eventually, at some point (I hope), Steamboat Willie will enter the public domain. The very same day, every one of the remaining, successful P2P companies will immediately register their version of Steamboat Willie with the copyright office and begin offering their version as the number one search result for Mickey Mouse. Cosmic justice? Perhaps.
Good, Bad, Indifferent?
I personally think the scenario I paint would be a somewhat perverse result, but current copyright law also has some strange results. Others might not think the result perverse (or simply think my logic is flawed). However, even if the result is perverse, it might not be all that bad in the greater scheme of things. The benefits of a particular policy might outweigh its flaws. It might even be possible to rework the law (or social practice) so that such scenarios would be less likely. Still, it is the job of everyone in this debate over compulsory licensing to look deeply at the proposals and consider what the various ramifications will be, not simply how do we collect and distribute the money.
This is one in a series of posts looking at some of the issues regarding compulsory licensing that don't include collection and distribution of funds. For previous postings, see:
Compulsory Licensing - What is Noncommercial Use?
Some Questions and Concerns Regarding EFF's Filesharing Policy
What Should the RIAA Sue For?
Compulsory Licensing - More on Commercial/Noncommercial
Compulsory Licensing - The Death of Gnutella and the Triumph of Google
Compulsory Licensing - Where Are the Defenders of HTTP?