Let’s come back around to Hal Plotkin’s question–who owns digital learning resources? Plotkin wants the answer to be: certainly not the university bureaucrats aiming to “commercialize” everything and therefore putting everything behind a paywall. That makes sense. Bureaucrats don’t have a great track record when they appropriate research results–and haven’t since 1968 when the NIH revived its Institutional Patent Agreement program, and haven’t under Bayh-Dole, which replaced the IPA program after it was shut down for being ineffective. But the alternative to “not university bureaucrats” is not the Department of Education requiring everyone to license openly final versions of digital learning resources.
The best roles for federal regulations pertaining to rights in grant deliverables (data, compounds, software, protocols, documentation, inventions, methods, know-how, whatever) made with federal support is
(i) to prevent administrators, speculators, corporations, and governments from forcing their bureaucratic thumbs into every new thing, and
(ii) to limit the incentives for those who do work with federal grants to bounce from one federal grant to the next, dropping the development and support of each thing created in order to gain the benefit of the next year’s federal funding.
For (i)–prevent university administrators from asserting ownership or control over work prepared by faculty, students, and staff unless the university administrators proposed the funded work and assigned the personnel and sought the grant. Limit the ability for companies to buy up all the rights to such works as well–they might get an exclusive right to sell a particular version of the work, but not a right to exclude all others from working with each work.
For (ii)–include what people have done to make prior federally supported work available in any review of proposals for further government funding. If more federal funding is easier to get than continuing to work on whatever was funded last year that was so very important that it beat out nine other applications, then the federal agencies are taming the bears and feeding them from cars, creating their own versions of Yogi Bear–more interested in navigating the federal funding landscape for picnic baskets than being faculty working on what they would otherwise work on.
It may well be, then, that the Department of Education “open” requirement is directed at university administrators who would otherwise claim ownership, withhold everything from distribution, and attempt to “commercialize” each new thing as if it were the next lucrative biotech drug–which is a failing proposition, even for biotech drugs–and that’s based on a history of university bureaucrats attempting to do so since 1968.
So who owns digital learning resources? The answer is not so clean as “no one should” or “no one does” or “the public” or “anyone may as long as there’s no ownership interest to speak of.” The answer is certainly not AUTM’s implied answer–“university bureaucrats should own everything,” and certainly not the Department of Education’s “softened” requirement to openly license everything except for all the exceptions.
Let’s start with basics. Authors own copyrights in original works of authorship. Inventors own rights to patent in patentable inventions. That’s basic public policy. Neither has to enforce their ownership positions and neither has to “work” their rights–they don’t have to publish and they don’t have to use the patent system and they don’t have to use their inventions. To the extent that “digital learning resources” are owned via copyright or patent, there’s the starting point–authors and inventors. We might ask, then, why basic public policy should change just because a federal agency makes a grant–a subvention, assistance for what individuals have proposed to do that is judged to be in the public interest if they do it. What’s so wrong, as a matter of public policy, with authors having a right to their original works of authorship or inventors to their new, useful, and non-obvious inventions? If the federal government has a royalty-free nonexclusive license to anything that’s created with federal support, what more does it need?
We might put it another way–why should the federal government even need a compulsory license when the funding is made as a grant and not as a procurement contract? In a grant, the government is not procuring anything–it is supporting what has been proposed because the doing of what has been proposed has been judged to be in the public interest.
Consider how authors and inventors might contract with others to give up their rights or constrain their exploitation of these rights. A would-be author might agree to work for hire, giving up authorial control to an employer or a commissioning party. In such cases, the author becomes the employer or commissioning party. If the commissioning party is the Department of Education, then the Department might contract to secure copyright as a condition of the grant. Of course, since there is no copyright in works of the federal government, the result is, hah–the equivalent of an open license to “wider stakeholders.”
Similarly, if a faculty author gives up the right of control to his or her university, agreeing to allow administrators to dictate the final form of the work, and that work then comes within the scope of the faculty member’s employment, then the university is the author outright of the work. The “grantee” in Department of Education terminology, which stands in lawyer talk for “any institution that receives Department funding, including universities receiving funding as trustees on behalf of faculty-proposed and -conducted projects in which the universities have no right to direct or control the work or any dissemination of the work.”
The federal regulations here at 2 CFR 315(a) are carefully constructed to make it clear that “vesting” is a matter of deciding how federal grant funds are treated once they come into the hands of an organization. When the organization uses federal grant funds to acquire property, does it acquire that property on behalf of the federal government, or for itself? The regulations provide that grantees (or “non-federal entities” in advanced lawyereez), if they acquire ownership, can keep that ownership–that is, title “vests” in the party that has done the acquiring. Gain the right to possess, gain title. Thus, even the federal regulations provide no special privilege for grantee organizations to assert ownership of original works of authorship–just as Bayh-Dole does not provide such a special privilege for patentable inventions.
Are you with me? Of course! Now consider–those doing the authoring of these digital learning resources, especially in universities, get to decide how others may get involved. It’s their agreements that establish who may be co-authors and who contributes works by permission and who does work that is not original or authorial and therefore does not create any ownership interest in the final work. That’s a matter of choices and practices and agreements, including those that are recorded in writing and those that are oral and those that are implied by conduct. That’s where ownership happens for works of authorship. There’s a similar situation for patentable inventions. One doesn’t know the inventors relative to any given patent application until one has written the claims–who has contributed to a claim of the claimed invention? What claims are allowed in the issued patent? There’s the ownership of the claimed invention.
University administrators for the most part hate such an outcome. It stinks of freedom and personal self-determination and fails to pay sufficient homage to the status of otherwise low-status administrators who think they should own and control anything of value made anywhere in a university. For them, it is an abuse of “public” resources for anyone not to cede control to university bureaucrats (so they can have generally better paying jobs than anyone actually creating digital learning resources of any value). It’s pretty easy to see through their rhetoric.
But it is also the case that unmanaged authoring situations can lead to a mess of trouble, especially in university environments where people think “fair use” is pretty much anything they want to do. Even if that were true, they end up producing work that cannot be distributed because it incorporates copyright elements from many sources that cannot be traced. Fair use for the university’s use, perhaps, but not for most anyone else’s use. It doesn’t much matter that the Department of Education requires open licensing for such works–because the Department then walks back precisely what it ought to be insisting be well managed (2 CFR 3474.20(e)):
The license set out in paragraph (b)(1) [the open license] of this section shall not extend to any copyrightable work incorporated in the grant deliverable that is owned by a party other than the grantee or subgrantee, unless the grantee or subgrantee has acquired the right to provide such a license in that work.
So, pull in any durn thing you want and incorporate it into your grant deliverable. Now that grant deliverable cannot be openly licensed because you simply don’t have the necessary permissions to do so. Publishers carefully manage this part: the work will be original. Anything that’s not original will be recorded and permissions obtained consistent with the publisher’s rights to publish established under contract. The author indemnifies the publisher for any breach of this obligation. These are the provisions that ought to be in any Department of Education grant intending to support the preparation of widely disseminated works. It doesn’t matter whether the work must be licensed “openly” unless the work consists of materials that can be licensed openly. By backing out such included materials, the Department of Education goes exactly the wrong direction–towards a useless ideology and away from the pragmatic requirements needed to produce distributable works, regardless of whether they are self-published or distributed by a publisher, whether a nonprofit or for-profit.
The same is true of the federal government’s license. It doesn’t operate to take rights from those that own copyrights in material incorporated into the grant deliverable. It just gives the government rights in the material that it can claim rights in–pieces of the muffin, but not most of the raisins. It’s not possible for the federal government to distribute such stuff, though state agencies (and other instruments of a state such as universities or school districts) might claim sovereign immunity and distribute anyway. No private educational organization could touch such stuff, however–at least not to modify or copy.
It may well be that the university grant environment encourages sloppy handling of other people’s work. Just throw things together! Fair use! Grant funding! If it’s on the internet, it’s free! Cruddy thoughts like that. No one conducting work that seriously ought to be widely disseminated should think that way. University “technology transfer” people are like the Spanish Inquisition in the Monty Python sketch–unexpected, ridiculous, and late to the party. When someone is reading to “disclose” an “invention” (defined in the ridiculous way that most universities by policy now define “invention”) of a digital learning resource, they are typically years after the choices should have been made about how to build and manage the work. Doesn’t matter how the work is to be disseminated: it generally can’t be. It wasn’t built to be disseminated. It was built to demonstrate how well the builders could build if they didn’t have to worry about dissemination.
Ownership of digital learning resources isn’t the right question, then, though it does a good job of starting a discussion. What are good management practices in developing a digital learning resource that can be widely disseminated? That’s one question. With good management practices–making clear who is authoring and who is contributing and who is providing editorial or data services or clerical services; logging in any existing works to be used and the permissions for those works; establishing contracts that provide for dissemination; and allowing the government to make the work available as well, for any government purpose.
There’s a second question: how does a disseminated work get maintained, updated, corrected, validated, localized, extended, installed, documented, trained for? It’s easy to put something up with a CC or GPL license. And it can sit there until the internet freezes over. Just because a project ends up on GitHub doesn’t mean anyone is going to jump in and lend a hand. If the Department of Education worries that despite local dissemination and conference presentations and professional organizations adopting something and commercial publishers selling copies some grant deliverables still aren’t getting the “widely” dissemination that the Department’s officers desire, then they need to look at how the mechanisms of adoption and support are handled. An open license does not solve this problem. In fact, it makes it worse. Luckily, the open license requirement is so easy to work around that in the end it involves creating much more bureaucratic baggage–writing stupid dissemination plans, justifying why an open license can’t be had–instead of managing new work from the get-go for wide dissemination.