Orphaned works and unexploited copyright

While working on my next additions to the C Data Structure Library, today I'm going to talk about a more philosophical problem: resource underutilization due to wasted exclusive rights.

Although each country in the world has its own laws and rules, nearly every country in the world today adheres to the Berne convention, a uniform international policy governing copyright (with some exceptions in Africa, Southeast Asia, and the Middle-East). In the United States, the most notable change introduced by the Berne convention was that works need no longer be registered to receive copyright protection; as soon as they are written or stored in some fixed form, exclusive rights are instantly and automatically granted to the author. This is quite convenient for some works - for example, a thief couldn't legally steal the rough drafts written by the author of a book and publish them. But it can get pretty strange: even if the author is a 4-year-old child who knows nothing about copyright at all, they still get exclusive rights.

A serious practical problem arises if you find an abandoned work lying around, and want to create a derived work based on it; chances are, some author somewhere still has exclusive rights to it. If you don't contact the author to secure rights, and they later find out about your derived work (say because you make a lot of money off it), they can sue you. But if the author has abandoned it and vanished, it may be very difficult to determine who the author is, much less contact them. These works are called orphaned works.

Another serious practical problem arises with unexploited works: copyright is intended to protect the creative work of an author so that they can exploit it to make a living. But under the Berne convention, an author - perhaps one with no knowledge of copyright at all - can sit on top of a huge pile of work, most of which they have no intention of exploiting.

The cost here is to society at large: these abandoned and unexploited works aren't used and benefit no one at all, when they could be contributing value directly through use and indirectly through derived works. And this isn't just a problem with books and paintings; how many abandoned software projects languish in the vaults of corporations, never to see the light of day again? How many abandoned open-source projects are up on SourceForge under old licenses that are incompatible with modern ones? That could be code that you're rewriting right now.

The United States was always a nation that emphasized growing the public domain; all works of the federal government are automatically released into the public domain. When U.S. copyright was established in 1790, works gained copyright protection for 14 years when they were registered, and it could be renewed for another 14 years. Besides the much more limited term, this meant that authors only registered works that they considered to have potential value. This did mean that sometimes people who didn't know to register their work would have it stolen with no legal recourse, but overall it struck an effective balance between allowing authors to make money and making old and abandoned works available to society for reuse.

One obvious question to ask is, do people really create derived works? What is the value of derived works? In 1790 this may not have been obvious, but one look at the Internet, where copyright infringement is rampant and derived works are ubiquitous and, frequently, highly creative, proves their value. Indeed, these works could be said to be at the center of Internet culture. Usenet, mailing list, and forum posts are routinely copied, archived, and republished without acquiring permission from their participants, an exercise that would be prohibitively costly if it were enforced. Photoshopped images, YouTube montages and fansubs, YTMNDs, musical reinterpretations and parodies, much of the artwork on deviantART, and many popular Flash videos are derivatives of copyrighted works. It also includes code: although few software developers dare to blatantly infringe copyright, forked codebases by different developers like the X.Org fork of XFree86, the XEmacs fork of GNU Emacs, the OpenSSH fork of SSH, and forks of Linux and BSD distributions (including Mac OS X) are ubiquitous. Enabling derived works like this is part of the mission of open source software.

In 2003, a bill called the Public Domain Enhancement Act (PDEA) was proposed but never put to a vote. It would have, without changing the term of copyright, reinstituted copyright registration in the U.S. for domestic works at a dramatically reduced fee of $1 per work. This helps solve the problem of orphaned works: if the author registers it, they must provide contact information that can be used to contact them and license the work. If a work is abandoned or unexploited, the author will (most likely) not invest effort in registering it, and it will enter the public domain and become available for reuse.

A large part of getting an act like the PDEA right is in the implementation. I imagine a government website where you can go and fill out a web form to register a work. You create an account associated with your name and contact information. Each time you want to register a work, you go to a page where you supply the work's title and description, and an electronic copy of the work, if available. The payment would work similarly to an MP3 purchasing site: you enter several works, and the cost is summed up and billed to your credit card at the end (in order to mitigate point-of-sale credit card fees). In return you get a permanent URL that you can hand out to anyone proving that you own this work and supplying the next renewal date. If you think $1 per work is too much for authors with many works, this system could automatically offer a "volume discount," reducing the fee per work for authors with many works. For companies that routinely register many works, a web service interface could be used to automate the procedure; this could also be incorporated as a feature into popular software like blogs, wikis, photo galleries, and version control software. The fees would be more than sufficient to cover the bandwidth, hardware, and storage needed to drive this service.

What about the case where an author's unregistered drafts are stolen and registered by someone else? The key is to make the registering of unpublished works that you did not produce a punishable offense that also invalidates the claim of copyright. This gives authors the opportunity to develop a work privately and only register it immediately before publication.

What about authors who don't have access to the Internet? Most tax filing is electronic today, but the government retains paper forms for those who aren't able or choose not to use it; copyright registration would be no different.

In short, I believe this system is practical to implement and would be effective in growing the public domain with many works that have no value to their authors, but might be quite valuable to someone else. All we need is the legal framework to make it reality.

All content in this document is granted into the public domain. Where this is not legally possible, the copyright owner releases all rights. This notice may be modified or removed.