Now let’s deal with “digital” educational works in the context of university intellectual property claims. This is something I’ve spent a couple of decades dealing with.
The Department of Education published its final rule in January 2017, requiring open licensing of final versions of works prepared in the Department’s competitive grant programs, with some exceptions and an opportunity for appeals. The implementation of the final rule is at 2 CFR 3474.20.
There are a couple of big observations to make at the outset. First, if someone doesn’t like the idea of open licensing of final work prepared with the support of the Department of Education, then he or she (or it, in the case of organizations) can seek elsewhere for funding. No one is forced to take Department of Education funding. Just move on. In my experience working with university-hosted research of digital media and other works of authorship, most of the effective dissemination programs I have seen have been ones that have not received federal funding.
Federal funding counter indicates “technology transfer.” Nothing in the federal regulations appears to matter one way or another. Folks that seek out federal funding are not the folks that are primed to do much of anything else but seek out more federal funding. The same is true of the SBIR and STTR programs. The National Science Foundation cannot figure out why its SBIR grantee companies don’t create commercial product. Well–it’s easier to get another SBIR grant! And SBIR grant proposals only have to look better than other SBIR grant proposals. If one really did have a great product idea, then being happy to spend nine months waiting for $100,000 with a bunch of audit strings attached is often a bad sign. Just saying.
Second, the Department of Education open licensing requirements are simple to work around. Prepare a final version of a work under grant, release it as required via an open license, and then prepare a new, revised version not using Department of Education funding–add a user interface, correct bugs, add useful extensions, give the new version a distinctive name and put it out on a web site or distribute through a major publisher. The new version will swamp out the open version. It would take a determined group to develop the open version in competition with a revised, corrected, and extended version. Department of Education open licensing requirements make this workaround express. Here’s the Department’s comment on the issue:
For copyrightable works that are not funded by the Department, we have similarly left the terms under which any derivative works may be licensed to the discretion of the owner of the derivative work (e.g., if a grantee created a deliverable with grant funds and then creates a derivative work with other funding, the grantee would have the flexibility to choose how to license the derivative work, such as through commercial channels).
Since a licensee can create derivative works not funded by the Department of Education and ought to be able to license those works any way it chooses, then so can the author of the original work. What the Department of Education cannot grasp is that the author of the original work can license that original work multiple ways–open and not–just as one might choose to pay for food in a restaurant so someone was there to wait tables and wash up afterwards, not to mention supply tables and chairs and a generally nice ambiance for a meal–even if the same food was available out back of the kitchen in the parking lot, if you brought your own dishes and hauled the food away somewhere else to eat.
“If you want the open version, here it is, without support and without anyone prepared to help you use it or correct any defects. Have at it! If you want support, and the current version, then license or buy our non-federal version. Oh–by paying for licensed rights or purchasing a copy, you support updates to the work, which otherwise would become obsolete now that the Department of Education funding has run out.”
And of course, the Department of Education can waive its own rule on a case-by-case basis, so even if the rule is simple to work around, the Department can help out those too inept or unmotivated to do the work-around themselves:
A grantee or subgrantee that has received an exception from the Secretary under 2 CFR 3474.5 and 2 CFR 200.102 (e.g., where the Secretary has determined that the grantee’s dissemination plan would likely achieve meaningful dissemination equivalent to or greater than the dissemination likely to be achieved through compliance with paragraph (a) or (b) of this section, or compliance with paragraph (a) or (b) of this section would impede the grantee’s ability to form the required partnerships necessary to carry out the purpose of the grant)
Translation: the Department may waive this rule where a grantee has a plan for non-open dissemination that sounds good or the grantee wants to do something different that also sounds good. Sigh.
In other words, this open licensing regulation is mostly an empty administrative gesture. The requirement is a form of compulsory license that creates, in general, orphan versions of grant-funded “deliverables.” Perhaps some orphan version will get picked up by a powerful organization who then makes changes and rushes that version into commercial distribution under a proprietary deal. The result? Any powerful organization that creates a canonical version gets the revenue from sales and the original creators of the materials–forced to use an “open” license, get nothing. Unless a community organizes what amounts to a common monopoly interest in producing the canonical version of a work and maintaining that version with updates and support, and anyone making other changes must “share alike” with the common monopoly, “open” pretty much means “never getting paid for what others can exploit and get paid for.”
The Department of Education final rule allows a copyright owner to exclude from the open license uses for “commercial” purposes (see 2 CFR 3474.20)–thus, the distribution might be said to be “open” or “somewhat open” but certainly discriminatory–companies can be denied the right to use these educational materials for companies purposes–which might include, say, provision of training services. As well, anyone–companies and nonprofits and individuals–can be denied the right to use the materials in commerce. Not really open at all. Just an expanded treatment of fair use.
The Department of Education regulations also exclude works with proprietary materials in them that do not permit open licensing. Thus, if one doesn’t want to go open, then one merely includes materials that don’t permit open. It is really that easy. Same for “peer-reviewed” scholarly publication. Only those that want to go open, will go open. Same for any other “rights” or “obligations.” The rule does not apply to (2 CFR 3474.20(d)(viii)):
Grantees or subgrantees for which compliance with these requirements would conflict with, or materially undermine the ability to protect or enforce, other intellectual property rights or obligations of the grantee or subgrantee, in existence or under development, including those provided under 15 U.S.C. 1051, et seq.[trademark], 18 U.S.C. 1831- 1839 U.S.C. 200, et seq. [foreign espionage–a weird version of trade secret].
That means, if an owner of a deliverable work claims some stuff in the work is a trade secret, then so much for the open licensing. Answers to test questions comes to mind, as do algorithms and methods. The Department leaves off from the final regulation “and 35 U.S.C. 200, et seq.“—which is there in the Department’s commentary on its final rule.
Several commenters asserted that the proposed regulations conflict with the Patent and Trademark Law Amendments Act, also known as the Bayh-Dole Act (Pub. L. 96-517, 35 U.S.C. 200 et seq., which covers the intellectual property rights for patentable inventions resulting from Federal funding, as well as E.O. 12591.
For what it’s worth, Bayh-Dole deals only in patent and plant variety protection rights, not “intellectual property rights.”
Commenters citing these conflicts note specifically that computer software source code can be both patentable and copyrightable and that under the Bayh-Dole Act, inventors, rather than the Federal government, are entitled to the title of the patents. These commenters suggested that further clarification of rights is necessary in order to avoid both confusion and litigation.
Too stupid for words. AUTM was one of these commenters. Here’s their summary—
The Department of Education issued long-awaited rules regarding the licensing of copyrighted software using Department of Education funding. The ruling sets a potentially damaging precedent by requiring that software inventors using Ed research funding to open-license their inventions. Its impact has yet to be assessed. Higher-ed groups, including AUTM, were able to soften this language.
What a crock. Let’s start with Bayh-Dole’s “exceptional circumstances” (35 USC 202(a)). Any federal agency can determine that exceptional circumstances exist–essentially any circumstances at all–and modify Bayh-Dole’s basic gesture of permitting contractors that acquire inventions made in work with federal support to keep those inventions. That’s what the Department of Education does–or ought to have done–with its open licensing rule. There’s no “damaging precedent” here. The Department of Labor created an open licensing requirement in 2016. See 2 CFR 2900.13.
In addition to the guidance set forth in 2 CFR 200.315(d) [regarding data], the Department of Labor requires intellectual property developed under a competitive Federal award process to be licensed under a Creative Commons Attribution license. This license allows subsequent users to copy, distribute, transmit and adapt the copyrighted work and requires such users to attribute the work in the manner specified by the recipient.
So the Department of Education does not set a precedent here. More: if AUTM is correct in its assertion, then it is “damaging” for federal agencies to use the apparatus Bayh-Dole sets up to be used to assert any public interest as a federal agency objective. Any private owner of an invention, according to AUTM’s–um–logic, should be able to preempt any public purpose. Preemption of all other public purposes is in this twisted account, the public purpose of federal funding. Anything else is “damaging.” AUTM and others didn’t “soften” the language–they made the Department of Education’s rule useless for its purpose.
Back to the Department of Education’s account of Bayh-Dole:
Many of these commenters questioned whether the Department was aware that 35 U.S.C. 212 provides to institutions the rights for copyrightable intellectual property or whether the Department has the legal authority to require an open license under the provisions of that section.
The commenters here are nutso. 35 USC 212 has to do with educational grants. Here is 212:
No scholarship, fellowship, training grant, or other funding agreement made by a Federal agency primarily to an awardee for educational purposes will contain any provision giving the Federal agency any rights to inventions made by the awardee.
The “any rights” are the rights to “inventions” under title 35–federal patent law. Not other rights–such as rights in data, say, or rights in copyright. The Federal Acquisition Regulations make this clear. So does 2 CFR 200.315. But the commenters persist–they cannot separate patentable invention from original work of authorship, and don’t want to.
Commenters citing these conflicts note specifically that computer software source code can be both patentable and copyrightable
These commenters are happily and authoritatively clueless. Source code cannot be “patented”–at least not now. Perhaps back when the idea of interpreted or compiled languages were being developed there was an opportunity to patent the concept of source code. Source code is a considered a “literary” work in copyright law. In patent law, source code is at best a “best mode” of practicing an invention–which has to do with what the executable software does, not how that executable software is represented. The method, the application of algorithm–that’s the patentable invention. These folks cannot sort out a claimed invention from a best mode of practicing the invention, and yet here they are being authorities. Why would anyone allow them to get anywhere close to inventions made in work with federal support, let alone original works of authorship?
And here the commenters mistakenly get Bayh-Dole almost right:
and that under the Bayh-Dole Act, inventors, rather than the Federal government, are entitled to the title of the patents.
Under Bayh-Dole, nothing changes for inventors. Inventors are “entitled” to “title” as a matter of federal patent law. Bayh-Dole does not disturb inventors’ “title” to inventions. And Bayh-Dole does not give universities any special rights in those inventions. If one follows the–um–logic of Bayh-Dole and its implementation, inventors own their inventions, universities are required to agree to keep their mitts off those inventions unless they have assigned and controlled the work (and therefore have equitable title), and inventors are covered by a patent rights clause specific to them that doesn’t even require them to use the patent system. But no matter. No one expects Bayh-Dole to have logic, other than the Supreme Court and perhaps a few intrepid readers of Research Enterprise.