Sooner or later, Oklahoma IP counsel might get it right about inventions

[expanded to include further discussion of the OU policy claim to inventions]

In a talk last fall (archived here), the IP counsel for the University of Oklahoma, presented this slide on the university’s patent policy:


It appears to be a quote from the university’s patent policy, and as far as it goes, it is. We have the usual messes–a policy on ownership that claims to own unpatentable inventions, the odd use of “made or conceived” which appears to conflate “discoveries” (which are “made”) and inventions (which are also “made” but, to be patentable, with two steps–“conceived” and “reduced to practice”). Of course if an invention is unpatentable, then who knows what is meant by “invention”? And what does it mean to “conceive” of an unpatentable invention? An unpatentable invention could be anything–an idea, an insight, a gizmo, a piece of software–and it doesn’t have to be new, or useful, or non-obvious. It just has to be called an invention, apparently. And how does anything get “conceived” through the use of university facilities or funds?

The claim made by the policy is indifferently vague, intended to be understood, perhaps, as authoritatively general. It would be much more direct to focus a university patent policy on “inventions which are or may be patentable” and get on with matters of ownership of those. But this is the usual mess of garbled policy. Next to impossible to read for an objective meaning. It does not have to be this way, but it is. Let’s press on.

There’s more to the Oklahoma policy than the part the IP counsel quotes in her slide. Here’s the rest of the sentence–the part she doesn’t include:

and the member will assign and does hereby assign all of member’s rights in and to the discoveries and inventions to the Board of Regents of the University of Oklahoma except as described below.

The policy states what “shall be” and then makes the usual awful effort to turn policy into patent agreement by asserting what a “member” has done (“and does hereby assign”) rather than authorize the university to obtain an agreement from the member. That’s nonsense. The policy statement is not a patent agreement, and not even a contract (though it might be incorporated into a contract–but that’s another matter).

But that’s not the part I want to focus on. Consider the “except as described below.” The patent policy’s approach is to claim everything except what’s not claimed. That’s not quite what the IP counsel puts on her slide.

Here’s the operative exception:

In the event faculty or staff make discoveries or inventions outside the course of and/or scope of employment and using no University facilities, equipment, or supplies, or if using such reimburses the University for this use in accordance with a prior written agreement with the University and in accordance with University policy, title to such discoveries or inventions shall remain in the inventor, provided the University Vice President for Technology Development determines that the discovery or invention was made under these conditions.

We have the usual ambiguity (at least the policy is drafted with consistent, uniform garble throughout) with regard to ” the course and/or scope of employment.” Here, the policy can’t quite tell whether there’s a difference between the “course” of employment (what someone does to accomplish a task for which one is employed (directed, compensated) by one’s employer and “scope” of employment (the range of duties for which one is responsible to one’s employer). For “course,” think driving to get a pizza for the provost. There may be nothing in the scope of employment about driving, but if the provost needs a pizza, then it’s reasonable for an employee in the course of employment to drive to pick it up. For “scope,” think of those things set out in an employment contract as duties an employee owes to an employer in exchange for compensation. In the case of faculty, one may have a duty to teach courses as assigned, but research tasks and discovery and inventing typically are typically not assigned, and are not obviously within either the scope or course of employment. Research activities may well be expectable, desired, encouraged, and supported. But there’s nothing employable about them for faculty until an employer assigns or directs them (or could, in theory–but I’d like to see a dean attempt to assign a tenured faculty member a research task).

We’ve been through all this a number of times at Research Enterprise. The main point is that not everything a faculty member does is within either the course of employment or the scope of employment. Thus, though the Oklahoma policy statement is drafted to appear broad, the requirements on the indifferently tossed course/scope language are actually narrower than it would appear. It is as if those drafting the policy wanted to persuade inventors that there was no hope–that the university claimed everything.

But we see that even with the use of facilities and funds, there are two conditions that mitigate a claim of ownership. The discovery or invention must have been made “substantially through the use of facilities or funds provided by or through the University.” The word that matters is “substantially.” Anything that’s not “substantially” supported is outside the scope of the University’s policy claim. Of course, this gets walked back, as the Vice President for Technology Development has to determine the meaning of “substantially.” Second, if the inventor has agreed in advance to reimburse the university for the use of facilities or funds, then the invention is outside the scope of the policy’s claims. Notice, too, how “facilities and funds” is ignored in the policy, and in its place we find “facilities, equipment, or supplies.” Clearly, “facilities” cannot also mean “equipment” and “supplies” or there would be no need to list equipment and supplies in addition to facilities. Yet the policy does not make a claim to inventions made using equipment and supplies–just “facilities and funds.” Go figure. Perhaps it is just sloppy drafting. But what is it that the university intends by its claim? Just how broad is its claim? As broad as it suits any administrator whenever? If so, just write that out and see how it flies as policy.

This second condition is remarkable. It suggests that the test for any invention that might be claimed by the university is what the university can show that would put a money value on the use of resources directed at making a given invention. If the university cannot show any actual cost–such as standing in a lab drinking coffee and having a realization, as distinct from using $3,287 worth of reagents–then how could the university show that an invention was made “substantially through the use” of any university resources? For faculty, who do not keep time cards, one would apparently have to show that a faculty member was busy inventing and not, say, meeting scheduled classes or attending assigned committee meetings. For extramurally sponsored work, there’s a similar problem–the university is compensated for such work by the sponsor, so there’s nothing for an inventor to “reimburse” the university for, unless the inventive work used grant money but wasn’t within the scope of the grant. Perhaps you see how the clause must operate.

The reimbursement condition shifts the burden to the university to show that an inventor has used university facilities and funds in a manner that would require reimbursement. What’s the point of making a written agreement to reimburse the university for nothing at all? We might think, then, that this reimbursement condition creates an expectation that an university claim to inventions based on the use of facilities or funds be documentable in a form that could be reimbursed. Furthermore, the university’s claim to an invention depends not on whether a reimbursement is unpaid, but whether the use is “substantial” with regard to the making of the invention. The reimbursement required for the university to waive its claim to an invention is the documentable, billable substantial portion of the use.

Thus, before any claim can be made by the university to an inventions, the invention has to be within course/scope or must be substantially made with university facilities (buildings, typically) or funds without a prior agreement to reimburse the university–an agreement only necessary if there is some documentable use of facilities, funds, or supplies that would constitute a substantial use for which the university requires reimbursement. “Substantially” is directed at the proportion of the “making” that is supported by use of university facilities and funds, not the significance or value of those facilities or funds in general. An inventor could use a great deal of university facilities and funding to discover that something doesn’t work, and then come to a realization with regard to what might work, done entirely differently, and use no resources whatsoever to make this invention. To what extent have the university’s facilities and funds contributed to this realization? One might argue not at all–the university work was in the wrong direction, a dead end. Thus, “substantially” is a strange measure for a claim on inventions. It works when an invention is intended and funds and space are allocated to make the invention, and lo that’s what happens. For most anything else, however, this use of “substantially” works to restrict university claims, but for a determination by the Vice President for Technology Development.

The policy then comes back around to repeat itself (continuing its effort to be consistent in its being a mess):

All rights in and to discoveries and inventions described in this policy shall be disclosed to and inventor will assign and hereby does assign all rights in and to said discoveries and inventions to the Board of Regents of the University of Oklahoma as a specific condition of employment with the University and admission to and/or attendance at the University.

The construction of this sentence is at best awkward. Perhaps it is intended to read that inventions will be reported to the Board of Regents, and inventors will assign to the Board of Regents those inventions in which the university asserts a claim, and that the requirement to disclose and to assign is a condition of employment for employees and a condition of admission to the university for students. We’ll skip the question of whether a present assignment can possibly operate in a policy statement. Look at the first clause:

“All rights in and to discoveries and inventions … shall be disclosed”–that makes no sense. Discoveries and inventions are to be disclosed, not rights in them. Perhaps “all rights” are to be assigned to the university, but what about “title” and “interest”? There’s an argument that “interest” is broader than either “title” or “rights”–in which case, “rights” does not necessarily get at what the university might want. What “right” does an inventor have to a discovery? Perhaps the university intends “the right to apply for a patent.” What other rights might there be? The right to be named (or not named) as an inventor on a patent application? The right to decline to file a patent application (which also constitutes a publication)? But for “non-patentable” inventions, what rights are there at all? Are we talking here about the right to publish the discovery? The right to claim priority as the first to discover? The right to be named (or not named) as an inventor on a patent application? The right to decline to file a patent application (which also constitutes a publication)? Who can say what is intended by “all rights”? Some of these “all rights,” however, would appear to be ones that a patent policy has no business requiring, especially of university faculty.

Notice, too, that the claim that is made here is restricted to those inventions “described in this policy.” Now, what inventions are those? Are they the inventions that the university claims–made and conceived within scope of employment or substantially with university facilities or funds that could otherwise be reimbursed but there’s no agreement to do so? Or does the university want all inventions, including non-patentable ones, disclosed, regardless of the university’s claim to own them? Not clear. A reasonable university patent policy might start with a requirement that inventors report any invention that is or may be patentable that has been made in assigned work or that has used significant resources made available under a written agreement. If there’s a matter of doubt, the university can reserve the right to review inventions after a patent application has been published to confirm that the circumstances do not come within the university’s policy claim. If an inventor wants to push such a review up, he or she can request the review at any time. All this would be straightforward to draft. But university folks would rather draft vaguely general claims and set up a process by which administrators can later decide what the words mean to suit their needs. That might work for policy, but it’s not a workable way to handle employment contracts or patent agreements–in these settings, this approach amounts to an agreement to agree, and does not create an enforceable contractual obligation, other than that a university typically has more money to spend on attorneys to force a situation in the direction administrators want.

Notice that the assignment requirement is asserted to be a condition of employment. That’s interesting. Does an assertion buried in a policy document really create a condition of employment? Or is the policy here stating that if an inventor does not report or assign an invention, university administrators are authorized to fire the inventor–they don’t get to own the invention, but they can get their revenge? If so, that’s surely different than making an agreement to assign future inventions in exchange for some other consideration, such as diligence in pursuing licensing of the invention and payment of a share of royalties.

The effect of the policy at this point is to say: if the invention is all that valuable, then leave with it and do what you want. Otherwise, if you stay, you have assigned it to the university. But the policy could be interpreted exactly the opposite way: if an inventor doesn’t assign and the university doesn’t fire the inventor, then the university has waived its claim. Which is it?

The present assignment language adds an additional complication, as if there is not enough vagary as it is. The “hereby assigns” makes it appear that something is assigned. Administrators have added that language because they were told this plugs a hole in policy, where an inventor might promise to assign and then assign to someone else instead. The present assignment language doesn’t actually plug the policy hole–instead it changes the hole to a question of what inventions actually were intended by both parties to come within the scope of the assignment. For a company, this scope is much easier to ascertain–the present and foreseeable future business interests of the company. But a university does not have such present and foreseeable business interests, other than trying to profit from anything it can acquire from its faculty and others. The inventions that are assignable are the ones that the university can establish meet the conditions it expressly states, not just any inventions administrators envy, or any inventions that happen to be related to an inventor’s area of expertise, or that might have some relationship to research work an inventor has undertaken. The policy sets out specific requirements. That’s the limit of its scope, with no added “you know what we mean” or “we really meant something else” or “you have agreed that we get to decide what you agreed to at a later time.”

It’s even possible that the university policy, construed as a unilateral contract, would not pass the test of conscionable public policy. Students pay to attend the school, but if they invent anything by “substantially” using university “facilities or funds,” the university claims it? That’s a bit broad to start with, and given the claim is buried in a garbled mess of poorly formatted policy further buried in a faculty handbook, how is any student supposed to know of this requirement at the time the student is sent a happy offer of admission? The way the university gets away with such a thing is that students tend not to have the $100,000 or more to hire an attorney to beat back such claims.

Oklahoma’s IP counsel also includes a slide on Bayh-Dole:


Here we have a selection of points about Bayh-Dole. That’s fine. But look more closely. The first point uses the statutory “elect to retain title” wording but directs that to innovations. Bayh-Dole deals only with subject inventions–patentable inventions owned by a contractor. That’s a much narrower focus than the university’s own claims. An innovation is something almost but not entirely different. And we have the ambiguity of what it means to “elect to retain title.” The Supreme Court ruled that the wording meant “elect to retain title that one has otherwise legally obtained.”

The second point builds on the ambiguity by using “elect to own.” Again, Bayh-Dole does not give anyone a right to own inventions by “electing” to own them. The election in Bayh-Dole is the election of an option granted by the federal government of the form: if you come to own an invention made with federal support, then you have the option to keep that ownership, subject to various requirements. If you choose to retain ownership and comply with the requirements, then elect the option to retain title. But here, we get “elect to own” as if the university, by notifying inventors of its choice, comes to own inventions by operation of federal law. And this is what the Supreme Court said was absolutely not the case.

The third point is accurate in its way, though it does not indicate that there is no penalty for not giving preference in licensing to small businesses other than that a federal agency may require a university to alter its policy on the matter for future transactions. And, really, of all the strange things one could point out about Bayh-Dole–U.S. manufacture for exclusive U.S. licenses, no assignment except to an organization that primarily manages inventions, a requirement to promote use of inventions rather than troll users of inventions, share royalties with inventors–why pick the licensing preference for small businesses?

The fourth point is inaccurate, because the government does not “retain” a license: the law requires the standard patent rights clause to include a requirement that the contractor grant a license to the government. The contractor (including the inventor-as-contractor) must convey the rights, must establish the government’s rights in inventions. The point here is deeply wrong, not merely superficially wrong. The government has no rights in any invention made by anyone, other than when the government contracts for those rights or slaps a secrecy order on them. Thus, even in sponsoring research at a university, the government does not have any right to inventions that might arise unless it expressly contracts for those rights. There is no presumption that the federal government has rights in anything. The presumption–that of the Kennedy and Nixon patent policies–is that the federal government will contract to obtain the rights it needs.

That’s why the (f)(2) written agreement clause is so important in the standard patent rights clause. It requires universities to connect the federal government interest directly to the inventors. It is inventors who can establish the government’s rights in inventions, because inventors own their inventions when they make them. It may be that an inventor agrees to assign title to an invention to her employer. But the government’s rights in the invention are not somehow “retained” by the government as a license. No, the assignee must execute a license for the license to exist and the government to have its rights.

The government’s license is not, as the slide contends, to “practice the patent”: the right is to “practice and have practiced” the invention, not the patent (37 CFR 401.14(a)(b)):

nonexclusive, nontransferable, irrevocable, paid-up license to practice or have practiced for or on behalf of the United States the subject invention throughout the world

“Practice” as used in the Kennedy patent policy and in the IPA program on which Bay-Dole is based means “to make, use, and sell.” The point here is that there’s a license, it is granted by the owner of the invention, and that license establishes the government’s right to practice the invention. So, another instance of not being clear about the law, being rather sloppy about it, and slopping in the general direction of advantages for the university’s interest in inventions, to the disadvantage of inventors.

What’s odd, then, is the Oklahoma patent policy feels the need to add a little something with regard to ownership of inventions:

The University agrees to act in good faith with respect to the determination of ownership.

If the policy forms a contract between the inventor and the university, then the university already would have a good faith obligation with regard to its duties under the contract. Good faith performance is a necessary element of any contract. But if the policy calls out this one duty as requiring a commitment of good faith, the policy suggests that it is not a contract at all, but rather a statement of the university’s operations and claims quite apart from any contract. That is–the university policy states what it is going to do with regard to inventions–it will claim them–and provides that in claiming them, it will act “in good faith.” Outside of a contract’s implied good faith, just what sort of “good faith” does the patent policy commit itself to? Clearly, the policy does not delegate authority to the Vice President for Technology Development to determine what “good faith” means in addition to “substantially.” You might read up on “good faith” here and see if you can make sense of the usage. The part about “honesty” and “intention not to defraud” are clear enough in general. But what exactly does this mean for a determination of ownership?

And so we come back around to Oklahoma IP counsel’s slides. If the university agrees to act in good faith in determining ownership, what are we to make of slides that omit the key exceptions to the university’s patent policy claims and make it appear that Bayh-Dole permits a university to “elect title” when the Supreme Court said the law does not do that? We are not talking about just anyone giving a talk–this is the university’s IP counsel. Perhaps the IP counsel explained all this in her talk and the slides by themselves are misleading (in which case, why are they posted without a disclaimer?). Otherwise, can such a talk be said to be in good faith on ownership, in compliance with the Oklahoma patent policy?

This entry was posted in Policy and tagged , , , , . Bookmark the permalink.

One Response to Sooner or later, Oklahoma IP counsel might get it right about inventions

  1. Pingback: Vice presidents for research beg for Directive 10-289. | Research Enterprise

Comments are closed.