In Part 1, we considered the 1946 University of Texas patent policy–clear, simple, smart–and the 1977 revision that grew more complicated but retained a focus on patents and the rights of inventors to decide whether to seek patents or just to publish. Now we consider how Texas turned its policies from providing resources to enable innovation to becoming an institutional predator on creative work of all kinds.
The 1988 Texas Patent Policy
The patent policy undergoes another revision in 1988. The form of the preamble is retained, but the policy is re-titled “Intellectual Property Policy” and “inventor” is replaced with “creator”; “invention” with “creation.” Immediately, we are in a different world. Inventor has a well defined usage, and to own an invention means to have rights to secure a patent on it. What one owns is a patent right. In the 1977 policy, this all makes sense: inventors decide whether they want to own (that is, patent) their inventions, and if so, then they work out the deal with the patent committee based on the circumstances. Otherwise, they publish and everyone is done with it.
But in 1988, things are different. The changes make nonsense of the preamble:
While the discovery of patentable processes or inventions and the creation of other intellectual property is not the primary objective of the System, for any such discoveries or creations, it is the objective of the Board to provide an intellectual property policy which will encourage the development of inventions and other intellectual creations for the best interest of the public, the creator, and the research sponsor, if any, and that will permit the timely protection and disclosure of such intellectual property whether by development and commercialization after securing available protection for the creation, by publication or both.
Take a look at a file compare between the 1977 preamble and its 1988 doppleganger:
It’s one thing to say that the System does not aim to create patentable inventions, as the 1977 preamble has it. That makes sense. It’s entirely another for the System to disclaim creating any “intellectual property,” especially when the policy goes off and defines intellectual property:
This policy shall apply to intellectual property of all types (including any invention, discovery, trade secret, technology, scientific or technological development, or computer software) regardless of whether subject to protection under the patent, trademark or copyright laws.
That’s “all types” and not the merely the statutory forms (patent, copyright, and trademark) but the folk forms–whatever people call “intellectual property” when they mean “intangible asset” (if they even knew such terms). The policy definition makes clear that it does not mean statutory intellectual property nor anything that might be owned–“regardless of whether subject to protection under the patent, trademark or copyright laws.” That’s what “protection” means–ownership of the right to exclude. The preamble uses “intellectual creations”–unable to grasp even conventional wording such as “intangible assets.” The preamble here reads, in essence, that it is not the primary objective of the System to create anything intellectual whatsoever. One might ask, what the heck is a primary objective of the System, if not to encourage the work of the mind in all its forms–precisely what the preamble here disclaims, and in doing so denigrates.
The opening move in the 1988 policy, then, is to abstract from something definite and bounded by law–what constitutes a patentable invention–to something bespoke and entirely a creature of policy itself–a specially defined term “intellectual property” which a casual reader might mistake for an otherwise, mostly well behaved standard term. The policy then argues that this general thing, this special form of “intellectual property,” which now encompasses most anything meaningful that folks at a university might do, is not a primary objective of the university, and so, apparently, whatever claims the policy might place on such stuff is incidental to the great purposes of the universities involved, whatever those unnamed purposes might be.
The preamble has been shifted in other significant ways beyond the merely nonsensical. Now that “creating” anything intangible regardless of whether there are any laws regarding ownership of such stuff is not a primary objective, the Board takes on itself the responsibility for encouraging the creation of such stuff anyway. Encouraging the development of patentable inventions means, in the 1977 policy, doing those things the get inventions published and used by others, whether or not owned, whether or not patented, whether or not resulting in profits for the university. That is “development” has a specific meaning–not the creation of inventions, but their use beyond the university. But in the 1988 policy, the scope is broadened to all things intangible and development appears to have no meaning at all. Later in the same sentence, we read that the policy will
encourage the development of … intellectual creations … and that will permit the timely protection and disclosure [of these creations]… whether by development and commercialization after securing available protection for the creation, by publication or both.
I’ve removed intervening words to make the fundamental grammar visible. In short, there isn’t any:
Encourage the development of creations and
permit their timely protection and disclosure
by development and commercialization
after securing available protection
or by publication.
This is, in technical terms, garble. There is no policy need to encourage people to use their minds in the abstract. That is teaching birds to fly, or here, making it appear that they don’t generally fly (lazy birds) but for an encouraging policy statement.
“Disclosure” is left hanging and no longer refers to the means of disclosing by patenting or publishing. One does not disclose by development or commercialization. The idea of patent as publication is lost entirely.
Furthermore, one does not “protect” creations by “development and commercialization” or by “publication” (other than in some strange sense of priority, perhaps). The sentence is now nonsensical. One might develop an invention by commercialization (company efforts to make and sell a product) or by other means–say, networked, non-market collaboration, as is often the case with open source software and with industry standards, which often never have a commercial product form. No one buys and sells TCP/IP, for instance. It’s a standard. It gets developed (the standard may change). It gets used. It’s not “commercialized.” It’s not “protected”–it’s negotiated, it’s stable, it’s valuable not as a source of rents based on threats of exclusion, but because of its widespread adoption as a foundation for shared opportunity. Grief on grief not to get this and have the urge to revise a patent policy into an intellectual creations policy labeled an intellectual property policy but really a statement of administrative power by people who are clearly clueless in the face of the task.
It is not at all clear why university creations require “protection” as a matter of policy. It is not even clear why inventions do. A patent does not “protect” anything. A patent grants a right to exclude for a limited time in exchange for full, open publication. If anything, a patent makes an invention vulnerable–to work-arounds, to blocking improvements, to legal challenges, to abuse by patent owners. In a company setting, one might use “protect” with an invention as a shorthand for “exclude others”–competitors, specifically, who might sell their own version of the invention without having the luck or spending the money to work it up into a decent product first. Lacking that, one might use “protect” in the sense of “build up a portfolio to impress investors, or to trade with others with desired technology, or to countersue nasty folks and little folks who might mess with us or mess up our plans.” But in a university, these uses of “protect” are empty.
The point of development–in the 1977 policy, as “progress” in the U.S. Constitution–is for the public to take up and use and improve the invention–spread, diffusion, development, progress, public benefit. To “protect” an invention (regardless of whether there are laws that give one the right to do so)–and worse, to “protect” any “creation”–in a university is to claim that before anyone might use or improve the creation, it first has to be subjected to a right to exclude, to be kept from everyone else. That, too, is nonsense.
Given that the preamble to the 1988 policy is garble, we might expect the rest is, too. Let’s see. The policy defines intellectual property broadly, claims this broad scope as its area of application, and then becomes a tangled mess trying to exclude stuff it ought not to have claimed. From a drafting perspective, this is the work of an uncertain, incompetent hand–claim everything (that which is possible and just to be safe also that which is not possible) and then fumble the exclusions so that any ambiguity can be argued remains within the scope of the general claim. “If we didn’t exclude it clearly from the scope, it’s not excluded.” That’s a mindset of folks who don’t know how to draft a clear scope for what the university should have some reason to take an interest in. Here’s the tangle:
This policy shall not apply to faculty authored written or
visual work, except computer software, produced in the author’s professional field, which property shall be owned
by the creator; nor shall the policy apply to intellectual property produced as a work for hire in the performance of a contract with the System or as a part of an employee’s assigned work responsibilities, which property shall be owned by the Board.
Funny, the policy still applies to faculty-authored musical works, sculptural works (are sculptural works “visual”?), choreographical works, database works, and anything else that’s not “written” or “visual.” Computer software is not excluded, even though it is often faculty-authored written work and is included in the category of “literary works” by the Copyright Office. As for “intellectual property produced as a work for hire” the policy here means “copyright” and merely restates the effect of federal law–the employer or commissioning party owns a work made for hire. What the policy does not do is explain when, if an “employee” is a member of the faculty, a work is “part of an employee’s assigned work responsibilities.” Does the policy claim that any work produced within the area of a faculty member’s appointment (as, say, a professor of history) is “within the scope of employment” because it is an “assigned work responsibility”? Even if it is not a primary objective of the System that anyone create anything (and so, it’s hard to gather how anyone could be much assigned to create anything, except incidentally, or by accident). We don’t get an answer on the point that a policy might clarify–such as, “Faculty-authored works are not within the scope of their university employment unless the work is expressly assigned by the university in writing, identified as a work made for hire, the commission is voluntarily accepted by the faculty member, and the work is done in exchange for acceptable consideration beyond the regular pay and benefits attending to the faculty member’s appointment.” That would be useful in a policy. Mais non.
We will leave little disturbed the drafting irregularity by which a policy both asserts that the policy does not apply to certain works and at the same time declares as a matter of this same policy that the Board owns those works. It cannot be both, except in the fantasy world of university administrators playing at writing policy.
We now turn to the effects elsewhere in the policy of replacing “invention” with “creation” and “patenting” with “development and commercialization after securing available protection for the creation.” Again, “creation” is randomly broad, where “invention” was specific, limited, and stable. “Patenting” pertains to a specific, limited form of publication by which a personal property right is granted. “Development and commercialization” are a class of activities that might operate on a “creation”–but they are not parallel with publication in the same way as patenting is. In the 1977 policy, the pairing has to do with publication without patent ownership or publication with that ownership. In the 1988 policy, the pairing is”development and commercialization” with ownership or publication. What’s been lost? The idea that a university might publish with ownership without demanding commercialization. The policy has replaced such options with a mandate to commercialize (and without a definition for what commercialization might mean in a policy given to strange definitions).
By the time we get to the restatement of the System’s “basic policy,” it is clear that the substitution in terms has wreaked a huge change:
It is a basic policy of the System that intellectual property be developed primarily to serve the public interest. This objective usually will require development and commercialization ….
In the 1977 policy, this statement applied to patents, which were to be developed in the public interest. That is, if an ownership position is to be taken via patents, the public interest must be served–even if whatever the public interest is and whoever it is who has standing to represent that public interest is left unstated. But in the 1988 version, the claim is expanded to “intellectual property” (of all types, whether ownable by law or not, “intellectual creations”–ideas, works, inventions, what-not) and yet the purpose moves from “requiring patent development by non-exclusive licensing” [with the prospect of other forms of licensing as needed] to “development and commercialization by [a list of licensing possibilities]. That is, developing patents by providing broad access (which permits uses of all sorts) has been replaced by a restricted mandate (“development and commercialization” of the intellectual property–meaning the subject matter, not the laws by which the subject matter might be owned. In short: “whatever you make, if the System claims it, will be offered for commercialization, not merely for public use.”
In this effort, the creators of 1988, even though the claim to their work is broader and more invasive, have less control by policy than the inventors of 1977. In 1977, the policy stipulated that
any person affected by this policy who…makes a discovery … should retain the ultimate right to decide how it is to be made public–by publication, by patenting, or both.
Here’s the 1988 version:
Any person affected by this policy who … creates intellectual property … should have a major role in the ultimate determination of how it is to be made public–by publication, by development and commercialization after securing available protection for the creation, or both.
Another sea change. Rather than the academic freedom that the 1977 policy recognized, the 1988 policy, uses deceptively similar wording to change an “ultimate right to decide” into a “major role in an ultimate decision”–meaning that those creating work no longer have the right to decide, just a role. And we are talking here about all “intellectual creation”–not merely patentable inventions. Before: inventors decided whether to patent or not. Now: inventors get to play a role. Before: inventors were “staff”; now they are “academic or non-academic employees.”
Now let’s look at the circumstances laid out by the 1988 policy to indicate the degree of the System’s interest in “intellectual property.” Remember the four categories from the 1977 policy–
(1) independent (own time, no System resources);
(2) state support–inventor’s choice to patent; if so, review for System interest; if none, independent;
(3) grant support–conditions of award take precedence over policy;
(4) industry support–exceptions to policy require prior approval, including the System president.
Now for the 1988 policy. The four categories are reduced to three, with industry-supported inventions merged with government grants. And rather than tracking the forms of support, the new circumstances track whether work is “related” to one’s “employment responsibilities.” Clearly, the idea of employment (work for the benefit of an employer) has taken precedence over the idea of appointment (work for the benefit of the public). This, too is a sea change. Here is a paraphrase of the new list of circumstances, with notes on what has changed. IP in the summary means “intellectual property” as defined by the 1988 policy.
- Unrelated. Adds IP “unrelated” to the “employment responsibility” plus independent; adds that unrelated IP is the “exclusive property of the creator”; no System ownership claim or financial interest.
- Related. IP “related” to “employment responsibility” or on System time, or with State funds, or using System facilities. Before IP is “disclosed to the public” or “for commercial purposes” or published, creator must “submit” IP to the patent committee for a determination of System interest. Long statement that component universities can make up their own categories and procedures. Mandatory for all employees to assign rights to intellectual property and patents to the Board. Royalties shared 50-50 unless a component establishes a different schedule.
- Grant support. Adds support from “for profit nongovernmental entities.” “Subject to ownership by the Board.” Contract terms take precedence over policy. Employees shall “Make such assignment of such creations as is necessary in each case in order that the System may discharge its obligation, expressed or implied, under the particular agreement.”
The first condition establishes a fundamental change from own time/own resources to whether a creation (“intellectual property of all types”) is “related” or not to “responsibilities.” Related is a broad, undefined term. How would anyone know whether something is related or not? If something is “within the scope of employment” then there is some document or pattern of dealing that makes clear that scope. But related carries the sense of “associated,” and “similar to,” and “having its origins in”–all broader than within the scope of employment, within what the employer has directed the employee to do, and the employee has promised to do. To use unrelated then means to take the complement of related–everything that cannot be so related.
The wording puts in doubt what an electrical engineering faculty member can do, even working on her own time and with her own resources. She could start an electrical engineering hobby at home–building circuit boards to control home appliances remotely–and that could be related to her “employment responsibilities” because it’s electrical engineering. The effect is to dramatically reduce the scope of what one’s “own time” is. In 1977, it was whatever one decided to do while not working on university work. It was a personal choice to make the distinction. In 1988, there’s no choice within one’s area of “responsibility”–which could be as broad as field of expertise, or area of appointment, or departmental domain, or teaching assignments. The policy is silent on just what is meant–which is to say it is ambiguous in introducing the change without explanation.
The second category introduces onerous obligations that run counter to academic freedom and reverse entirely the tradition of policy going back to 1946. In 1977, for state-supported work the inventor had the right to decide whether to seek a patent and when and how to publish. State-supported work included work “on System time” and with “System facilities.” This provision was consistent with academic freedom and with the idea that the System existed to make resources available to faculty, a matter of subvention, not employment (or procurement).
Only when an inventor chose to patent was there an obligation to submit the idea for review and determine the System’s interest. But eleven years later, no one can disclose or use or publish any intellectual creations made with any form of state support or which are related to “employment responsibilities” without first submitting them for System review. One observation: no one realistically could comply with this policy requirement. It would bring the university to a halt. Every email, every idea, every bit of dance step, every discovery has to be pre-screened, and if the System decides it wants to own that creation, then the creator is required to assign the creation to the Board.
When a policy is drafted so badly it cannot possibly be followed, and its requirements are therefore only selectively enforced, one may well wonder if the policy itself has any authority in areas of doubt or inconsistent practice. If the policy stated something of the form “the System shall own anything it decides to own” it would not be far from the effect created by the 1988 policy language. But putting it in such words would also create questions as to how the System has such authority to compel assignment of what amounts to federally established personal property–in works of authorship and patentable inventions. And that’s perhaps the point. By using obscure, almost impossible to follow language, the policy works to suppress the questions that a plain statement would raise regarding its authority.
Finally, we come to the third category–in the 1988 policy, all creations resulting from grants and contracts and gifts of any kind, from any source, are “subject to the ownership by the Board.” A new office of “Asset Management” gets added to the “Intellectual Property Office” (formerly “System Patent Office”) to approve efforts to gain “favorable treatment for the creator and the System” in such funding agreements. A third office–the Office of General Counsel also is introduced as being involved in the preparation of model agreements–more bureaucracy layered on each possible thing that might involve creations made in sponsored settings. But no such office is proposed to advocate for favorable treatment of creators with regard to the System. So, as between the Board and the creators, when the System comes to negotiate with research sponsors, one might expect that the System negotiates for itself, because it asserts that it has the right to own whatever creators may happen to create.
The remainder of the 1988 policy is taken up with issues of holding equity in companies taking licenses from the System. It’s noteworthy for strange, tangled language of its own. One example: “employees of the System who conceive, create, discover, invent or develop intellectual property may hold an equity interest in a business entity that has an agreement relating to the research, development, licensing or exploitation of that intellectual property.” Why is the employee right to hold equity limited to this string of five verbs? What about creators who, um, author or build or improve or collect or solve? Apparently the System forbids such creators from holding equity. And why limit agreements with the System to four nouns? Why not also agreements that, say, lease space or that manage services related to the intellectual property (“creations”)? And what about equity in “business entities” that aim to design around the System-owned intellectual property? Are such holdings forbidden by policy? Or not? No one could possibly know.
Having completed a careful reading of the 1988 policy, one might be led to read in a new light the next section of Texas policy, having to do with the use of alcoholic beverages and advocating the overthrow of the government.
Let’s review. The 1946 policy focuses on patentable inventions and establishes that any patents are not the System’s, but that inventors owe the System a modest share of any net income they make.
The 1977 policy continues to focus on patentable inventions, distinguishes four categories of circumstances for review, and maintains that inventors have the “ultimate right” to decide whether to publish, to patent, or both. Only if inventors choose to patent and they have used state resources, then they must submit their ideas for a determination of System interest. Federal and nonprofit grant work follows the conditions of the awards and may alter policy. Industry work that would alter policy must be pre-approved by the patent officer and the president.
The 1988 policy changes things substantially. Patentable inventions give way to a wildly broad and indefinite definition of intellectual property of all types that expressly disclaims patent and copyright as limitations on the definition. Certain “faculty authored” “written and visual” works (but not if software) are excluded from the policy altogether. Works made for hire are both excluded from the policy and asserted by the policy to be owned by the System. System interest in this bespoke intellectual property or “intellectual creation” is determined by the “relatedness” of the creation to an employee’s “responsibilities” along with any use of facilities or resources or “time.” The System asserts the right to own any such creations, presumably under whatever theory of ownership administrators can dream up–apparently not merely ownership of patents and copyrights. Creators of such creations have only a right to a role in “ultimate determinations” regarding publication and commercialization, and for all “related” creations, are forbidden from disclosure or publication prior to submission to the intellectual property office for review of System ownership interest.
One can begin to see how a university IP policy can morph from one that limits university claims to own patentable inventions but expects a modest financial interest in those inventions that are patented to one that aims to preserve academic freedom but provides for university ownership if an inventor chooses to patent and has used university resources. And from there, one can see how administrators might expand a policy’s scope from patentable inventions to everything, and claims to own everything except what the policy is allowed to spit back upfront, and everyone doing anything creative at all related to their university work must report their work first to the administration and assign ownership of whatever it is that the administration wants, retaining only the right of a role in whatever the university decides to do, and a share of any royalties. As for royalties, the university’s share in 1946 maxed out at 20%. In 1977, the university’s share maxed out at 75%, but only for things inventors chose to patent and used System resources and the like. In 1988, the university share fell back to 50%, but the university claimed most anything that someone might create at the university.
In 1946, there were “faculty personnel and other employees.” By 1977, these were “staff.” By 1988, these were “academic and non-academic employees.” Trustee became self-interested. Governance became management. Financial interest became ownership. And ownership expanded the scope of what was owned and who decided what to do with it. One can see the rationalizations that went into the policy change that turned Texas from supporter to suppressor of individual initiative.
One might say that by 1988, Texas had gone to the dark side, but it took the changes in 2002, where the Board simply asserts ownership over everything except for the exceptions, except where there are exceptions to those exceptions. By 2007, creators have only the right “to give reasonable input”–so much for retaining the right to the “ultimate decision” of whether to patent, or even having “a major role” in determinations of patenting. By 2012, the Board had become brazen enough to claim that it “automatically” owns intellectual property and asserts that it, the policy itself, is an assignment instrument that acts (as with power of attorney) to make the assignment outright on behalf of each employee and before any creations have been created. And by 2012, everyone is an employee, whether faculty or staff or visiting scholars. Students are included if they use university resources, which is hard not to do. Lesson for students in Texas universities–don’t let any good idea show up in a Texas lab or the Board may want it. Do you feel hunted by administrators now?
In part 3, we will look again at the preambles to the 1977 and 1988 policies, and compare them with developments in the 2007 version, watching inventor rights slip away and the stated purpose of the policy change from getting stuff developed and published to helping administrators prop up their failing fixation on the linear model of innovation.
Pingback: From provider to predator: University of Texas patent policy, Part 3 | Research Enterprise