from the incorporation-by-reference dept
For many years, we've written about Carl Malamud and his non-profit organization Public.Resource.org, which goes to great lengths to make sure that the law and other government documents are widely available to the public. While he's gotten lots of attention for battling states over their claims to hold a copyright in the law, perhaps his biggest fight has been over the question of whether or not private standards that are "incorporated by reference" into the law, are still covered by copyright. And, unfortunately, the federal district court in Washington DC has just ruled against him, and effectively said it's okay to lock up some important elements of the law with copyright. This is bad news.
Some background: as you probably know, there are tons of standards bodies out there who create various standards. Most techies are quite familiar with various technology standards, developed by various groups. But standards obviously go way beyond just the tech industry. Think: building codes for plumbers and electricians. These are often developed by independent, private bodies. Of course, you may also realize that some of these standards are in the law as well. These are generally known as "incorporated by reference." That's just a fancy way of saying that a private group created a standard and then lawmakers put into the law "this thing we're regulating needs to meet those standards." So, for example, fire codes may be developed by a private body, but then governments say that any building has to meet those standards. Voila: those standards are "incorporated (into the law) by reference."
The question, though, is how accessible are these standards? Many of the standards bodies that create those standards like to sell them. That's often how they make their money. But that seems to be in fairly dire conflict with the idea that the law should be publicly accessible. It's fairly difficult to argue that the rule of law is paramount when you can't even see the law without having to buy a bunch of expensive standards. To deal with this, many regulators and standards bodies have come up with awful hedges -- which basically say that any such standard incorporated by reference must be "available to the public," but they allow that availability to be insanely limited. So, for example, the EPA basically says, sure, you can see all of the standards, if you trek to DC and go to a special reading room (or a few other limited places):
Certain material is incorporated by reference into this part with the approval of the Director of the Federal Register under 5 U.S.C. 552(a) and 11 CFR part 51. To enforce any edition other than that specified in this section, the EPA must publish notice of change in the Federal Register and the material must be available to the public. All approved material is available for inspection at the EPA Docket Center, Public Reading Room, EPA WJC West, Room 3334, 1301 Constitution Ave. NW., Washington, DC, telephone number 202-566-1744, and is available from the sources listed below. It is also available for inspection at the National Archives and Records Administration (NARA). For information on the availability of this material at NARA, call (202) 741-6030 or go to http://www.archives.gov/federal_register/
And, of course, even then, there are serious restrictions on what you can do. Many of the reading rooms are "read only." You can read them, but you can't print or download electronic versions. As you can imagine, that can be fairly useless. Imagine you're trying to build a house, and every time you want to check if something is up to code, you have to go to DC to a special reading room, find the standard, check the details, but you can't print it out or download an electronic version.
You'd probably think that's a pretty silly way to have laws.
At least, that seems to be what Malamud thought, so he bought copies of a whole bunch of these standards, scanned them, and put them online, arguing that once they're incorporated by reference into the law, they are a part of the law and thus copyright shouldn't apply. The big standards bodies, including the American Society for Testing and Materials (ASTM), the National Fire Protection Association (NFPA), the American Society of Heating, Refrigerating, and Air-Conditioning Engineers (ASHRAE), the American Educational Research Association (AERA), the American Psychological Association (APA), and the National Council on Measurement in Education (NCME) sued Public.Resource.org for copyright infringement (and for some of them, trademark infringement). The plaintiffs were lumped into two groups -- the ASTM group (covering the first three groups listed above, which filed for copyright and trademark infringement) and the AERA group (covering the last three groups, which filed for just copyright infringement).
Public Resource made a number of arguments for why the standards bodies shouldn't win, but the court doesn't buy them. It's not impressed by the idea that since many people worked on these standards, the bodies don't really hold the copyrights. Malamud also argued that there's no valid copyright in standards, because there's nothing copyrightable in a bunch of standards, which are nothing more than "methods or systems" -- comparing it to the famous Feist case that said a telephone book isn't copyrightable, because it's just a collection of facts. And, of course, "methods and systems" are not copyrightable material under Section 102(b) of copyright law (this is the same issue that was debated in the Oracle/Google case over whether or not APIs were covered by copyright).
Unfortunately, the court doesn't see it that way for standards. It notes that even under Feist, standards meet the "extremely low" bar of creativity to be covered by copyright. And, it completely disregards the 102(b) argument as a misreading of the law, saying that it only bars trying to copyright the system or method itself, "not the written work explaining or describing that method." I can see that argument, though I'm still left wondering who actually thinks that copyright is the necessary incentive to create a building code. But, unfortunately, that's not quite how copyright law works these days.
The bigger issue is over what happens to the copyright once it's incorporated by reference into the law. Malamud's lawyers argued that it should be considered public domain. The court says... sorry, nope. As you surely know, copyright law says that works made by the federal government are not subject to copyright and are in the public domain. But that only applies to works actually created by the government. It has long been recognized that works created by others, and then assigned to the US government allow the government to retain the copyright. We can argue that this is dumb and bad policy (because it is), but it's pretty well-recognized. Here, Malamud's lawyers argued that even so, Congress intended things directly referenced into the law to be in the public domain. But the court basically says that Congress had a chance to weigh in on this and didn't... so, no dice:
Congress was well aware of the potential copyright issue posed by materials incorporated by reference when it crafted Section 105 in 1976. Ten years earlier, Congress had extended to federal agencies the authority to incorporate private works by reference into federal regulation.... However, in the Copyright Act of 1976, Congress made no mention of these incorporated works in § 105 (no copyright for “any work of the United States Government”) or any other section. As the House Report quoted above indicates, Congress already carefully weighed the competing policy goals of making incorporated works publicly available while also preserving the incentives and protections granted by copyright, and it weighed in favor of preserving the copyright system....
However, recognizing the importance of public access to works incorporated by reference into federal regulations, Congress still requires that such works be “reasonably available.” ... Under current federal regulations issued by the Office of the Federal Register in 1982, a privately authored work may be incorporated by reference into an agency’s regulation if it is “reasonably available,” including availability in hard copy at the OFR and/or the incorporating agency.... Thirteen years later, Congress passed the National Technology Transfer and Advancement Act of 1995 (“NTTAA”) which directed all federal agencies to use privately developed technical voluntary consensus standards.... Thus, Congress initially authorized agencies to incorporate works by reference, then excluded these incorporated works from § 105 of the Copyright Act, and, nearly twenty years later, specifically directed agencies to incorporate private works by reference. From 1966 through the present, Congress has remained silent on the question of whether privately authored standards and other works would lose copyright protection upon incorporation by reference. If Congress intended to revoke the copyrights of such standards when it passed the NTTAA, or any time before or since, it surely would have done so expressly.
Again, based on the legislative history, perhaps that's a reasonable, if unfortunate, reading of the law. And it's something that Congress should fix, but almost certainly won't.
But, taking a step back from all of this, there's still the problem that it's crazy. The idea that the law itself should be locked up under copyright is immensely problematic for a whole host of other reasons. And Malamud and his lawyers pointed this out, arguing that there's a due process issue here, where using copyright to block people from accessing the law deprives them of their due process. But, the court isn't impressed, basically saying other court opinions haven't bought this kind of argument, so it won't either. And, further, the court seems to feel that those limited reading rooms or fee-based copies are enough access:
... there is no evidence here that anyone has been denied access to the standards by the ASTM Plaintiffs or AERA Plaintiffs. Instead, Defendant simply argues that the public should be granted more expansive access.
Another argument is that once incorporated by reference, under the merger doctrine, the "law" and the creative expression are so tightly intertwined (i.e., "merged") that the law is now factual and thus not subject to copyright law. The court says "eh, we're not going to look at that, because it doesn't matter either way."
The court declines to resolve this merger doctrine issue, since under either approach, the standards maintain copyright protection.
After all that, we finally get to the fair use argument. Public.Resource argues that what it's done isn't copyright infringement, even if the work is covered by copyright, thanks to fair use. The court does a four factor analysis... and says "no fair use." On the big question of whether or not it's transformative, Public.Resource tried to argue that making the works more accessible (including turning them into digital versions that could be searched) was transformative. The ruling in the Google Book scanning case would seem to support that, but the court says no, because it doesn't see turning a written work into a searchable HTML file as enough of a transformation.
While it appears Defendant may enable blind individuals, like all other individuals, to access the standards at no cost, they still may have to take additional steps like OCR processing or converting to a different file type, as well as using additional screen reader programs in order to access the standards. There is no evidence that this would not be possible with Plaintiffs’ PDFs or by scanning Plaintiffs’ hard copy standards. In Defendant’s view, taking the first step or two towards making the standards entirely accessible to those with visual impairments is enough to have transformed the standards. This attempts to stretch logic, and certainly the doctrine of fair use, too far. Defendant has not offered a sufficiently new purpose to render the use transformative, and this weighs against a finding of fair use.
Within this analysis, I'd argue that the court goes way too far in dismissing the comparisons to the Google Books ruling and the ruling in the Swatch case that said that posting full transcripts of analyst calls can be fair use. The court here seems overly nitpicky, saying that Google Books doesn't count because it doesn't show full works, and Swatch (which does show full works) doesn't count because that work wasn't available in any other way (though I'm not sure what that really has to do with the fair use analysis).
The court also says that "the nature" of the work goes against Malamud, which surprises me. You'd think that the fact that these works are a part of the law would push it the other way. But, instead, the court misreads the nature of the Constitution to misunderstand what the framers meant by the word "science." This is unfortunate:
Defendant argues that Plaintiffs’ standards are “factual,” both because they are highly technical and because they are “the law.” However, the Constitution explicitly states that copyright exists to “advance the progress of science and the useful arts.” ... That Plaintiffs’ works involve technical scientific concepts and guidelines does not push it away from the core of intended copyright protection, but actually brings it closer. Plaintiffs’ standards are vital to the advancement of scientific progress in the U.S. and exactly the type of expressive work that warrants full protection under the Constitution and the Copyright Act.
This is just wrong. At the time the Constitution was written "science" had an entirely different meaning. It meant learning. That's entirely different than something that is mandated by law.
On the third factor of the "amount" of the work copied, obviously here it's all of it, and the court says that there is not "a single case" that supports the Defendant's view that because the entire standards were referenced into law, it's fair use. But that's also wrong. Again, the Google book scanning and the Hathitrust rulings both made it clear that using the whole work was fine because "it is literally necessary to achieve" the purpose they were looking to achieve.
While I found the court's reasoning on the copyrightability of the standards more persuasive, the fair use analysis seems incredibly weak -- and hopefully the appeals court will overturn it.
The only "win" for Malamud was that the argument for contributory infringement failed, but for fairly weak reasons, leaving it open to the possibility that a more complete attempt to make that claim could work.
Finally, the court also says that for the ASME plaintiffs, what Malamud did is also trademark infringement. I won't even bother getting into why, but this result was more or less expected given the results of the copyright part of the case.
The court then grants an injunction, basically ordering Malamud to delete all these standards from the internet. It includes a fairly bizarre determination of whether or not the public will be harmed by this:
Additionally, the public must not be disserved by the issuance of an injunction. Here, the public interest is served by the policy interests that underlie the Copyright Act itself, namely the protection of financial incentives for the continued creation of valuable works, and the continued value in maintaining the public-private system in place in the U.S. to ensure continued development of technical standards.
Did you get that ridiculous sleight of hand? The public is served by no longer having access to the law because it's better for some private organizations to get rich off of the standards that are a part of the law, or else such standards might not be developed. Huh?
I'm guessing that this case will be appealed, and hopefully the appeals court is more receptive to the fair use arguments. In the meantime, though, this seems like a pretty big loss for those who believe not just in the rule of law, but the idea that the law ought to be accessible to the public if it's to be respected.