The Hairball Theory of Ownership

Buried in a University of Washington web site on information technology, one encounters this statement:

Except as noted by an agreement, a law, or a University policy (such as copyright policy), the UW owns all data and records, and all associated copyrights created by UW employees within the course of employment.

This statement is evilflawed on many different fronts. Let’s work through it. First, we can ask why such a policy statement is buried here, in IT policy, rather than in the university’s patent, invention, and copyright policy. In fact, there in EO 36, the copyright policy provides just the opposite:

University faculty, staff, and students retain all rights in copyrightable materials they create, including scholarly works, subject to the following exceptions and conditions…

The list of conditions is contract requirements, work for hire, commissioned by the university (where that does not include “normal duties” of faculty). Apparently under the IT policy, another such special circumstance is everything that isn’t stated by other policies, like, say, the copyright policy.

The IT policy uses wording that appears drawn from the Washington state public records act. That act establishes obligations for state agencies to make records publicly available with various exemptions:

(3) “Public record” includes any writing containing information relating to the conduct of government or the performance of any governmental or proprietary function prepared, owned, used, or retained by any state or local agency regardless of physical form or characteristics.

A writing is also defined in a desperate kind of way, coming down to “every means of recording any form of communication or representation.” Here it is, just for the fun of watching attorneys draft: a list, a summary of everything else, and then an “including but not limited to” list, with a combination thereof tacked on, the another other statement of document with yet another including.

(4) “Writing” means handwriting, typewriting, printing, photostating, photographing, and every other means of recording any form of communication or representation including, but not limited to, letters, words, pictures, sounds, or symbols, or combination thereof, and all papers, maps, magnetic or paper tapes, photographic films and prints, motion picture, film and video recordings, magnetic or punched cards, discs, drums, diskettes, sound recordings, and other documents including existing data compilations from which information may be obtained or translated.

The IT policy appears to use the language of the public records act, but it confuses the requirement to make records publicly available with ownership of the record. The standard of the public records act has to do with writings containing information relating to the conduct of governments, not whether or not those records are owned by the agency. Certainly there is no requirement that an agency acquire copyright in a record in order to make that record publicly available. Ownership of a record is merely one possible condition of a record subject to the state law.

But the IT policy statement is even more confused that that, as it mistakes ownership of a physical thing, a writing, with the copyright that might vest in the expression of that writing. The public records act concerns “writing containing information.” Original work of authorship, the domain of copyright, is not the concern–it is making the writing available, so the information of the writing can be disclosed.

The ownership of copyright is not an issue, though the possession of a work that is subject to copyright does raise an issue with regard to the permission the state has to make copies of records, or publicly display those records, that it uses or retains. To comply with the state law, an agency would have to obtain permission from the owner of a copyright work or refuse to “use or retain” the record, or rely on fair use, which can only get one so far. But the UW IT policy merely asserts ownership of records, rather than considering permission.

There’s yet another thing going on with the UW IT policy statement. For UW to come to own a record, there has to be a theory of ownership, and compliance with its requirements. As for physical things, like notebooks, the university might own such a thing because the university paid to acquire it, after which the notebook was written in, and thus the university owns the record containing the writing. But that ownership does not extend to copyright. For that, one has to get at how copyright arises. If the work is not work made for hire, then for ownership to pass from the author to the university, there has to be a written assignment signed by the author. No written assignment, no ownership by the university, despite the policy’s assertion. University officials might demand the right to receive assignment–but they still have to find a way to get assignment.

We can then look at the qualification “course of employment.” Doesn’t that create a work for hire situation? Here’s the definition:

A “work made for hire” is—

(1) a work prepared by an employee within the scope of his or her employment; or

(2) a work specially ordered or commissioned for use as a contribution to a collective work, as a part of a motion picture or other audiovisual work, as a translation, as a supplementary work, as a compilation, as an instructional text, as a test, as answer material for a test, or as an atlas, if the parties expressly agree in a written instrument signed by them that the work shall be considered a work made for hire.

There follows some explanatory text, but this is the essence of it. Note here that the law uses “scope of employment” while the IT policy uses “course of employment.” Are these the same thing? If so, why not use the same language? Michael W. Klein summarizes the way courts have approached scope of employment in copyright:

(1) whether the work is of the type that the employee is employed to perform; (2) whether the work occurs substantially within authorized work hours and space; and (3) whether the work’s purpose, at least in part, is to serve the employer.

That is, employment for copyright purposes is not merely receipt of payment, but rather whether the employer has control over the work and the work place, and expects the work to “serve the employer.” It is a long way, for much of faculty and student work, to stretch scholarship into work for hire, especially when the state ends up controlling that work. Of course, some universities try to do just this.  Here is Washington State University, on the other side of the state:

a.) The University shall own the rights to all patentable property and other tangible research and scholarship developed as a result of University employment, or when the equipment, supplies, facilities, employee time or proprietary information of the University are used.

The language here is not apparently “all patentable <<property and other tangible research and scholarship>>, which doesn’t make any sense, but rather “all patentable property” and <<other tangible research and scholarship>>. The policy qualifies the ownership by the expectable employment and resources claims, but the employment claim is “as a result of University employment.” That’s expansive. One could leave the University’s employment and produce scholarship that was the result of that employment, but not within the scope of that employment. The problem of “result” of employment is like that of “course” of employment. The issue for IP claims is precisely what the agreement is under which faculty, in particular, agree to work for the university as employer–that is, to produce their work under the direction, and for the benefit of, the university.

It’s difficult to expect that there’s much attention to detail in WSU’s IP policies, given that nearly two years after Stanford v Roche, WSU still has the following:

b.)  Under the federal patent and trademark legislation of 1980 (35 U.S.C. § 200 et seq.), the University has the right of first refusal to title in inventions made in the performance of federal grants and contracts.

[The language is still there, as of May 24, 2018–clearly WSU just doesn’t care what the Supreme Court ruled.] Of course, the university has no such right of first refusal. But WSU, like many other universities, hasn’t bothered to change its policy. It doesn’t care, apparently. Or maybe the state’s Attorney General Office is just thumbing its nose at the the Supreme Court.

Unlike “scope of employment,” which has significance for copyright, and “result of employment,” which appears to be an overreaching graspingness, “course of employment” is a broader concept. It is used, for instance, in considering the liability of an employer for actions of an employee, or for an injury sustained by an employee.  “Course” of employment involves a broader range of circumstances than does “scope” and concerns actions and consequences, not merely the agreed upon range of services provided by the employee to the employer. An accident, for instance, is not going to be addressed in the scope of employment, but may well happen in the course of employment. In many ways, the IT policy confusion treats a record as if it were an accident rather than a mutually agreed upon work product.

The IT policy claims, as well, not just records and copyrights, but “data.” Data in some forms can be subject to copyright, of course, but there is plenty of data that are simply measurements or numerical stuff (computer “data”), and that’s not copyright subject matter, as there is no original work of authorship. So what is the theory of ownership for such data? It would appear to have to be something akin to trade secret–that is how a company would do it. But here it’s a university, which doesn’t require all employees to sign a non-disclosure agreement as a way of claiming ownership of data. Data, in university research, is what’s supposed to be made available, published, deposited. There is no way that a university comes to own data created in faculty-led research, though the university may own the media on which the data are recorded, and the laboratories in which the media are stored. The claim that the university owns data lacks substance.

Finally, we might say a word or two about public policy. While it may be fine for companies to tie up data with ownership claims based on trade secrets and scopes of employment and non-compete clauses, and while it may be acceptable for private universities to assert control over faculty work in various circumstances, state universities are not mere companies or just any university: they are also instruments of the state, and state government. When a government claims ownership of faculty scholarship, then the state is setting up to control rights of disclosure, rights of use, rights of publication. What’s the point of publishing a discovery if a state official gets to decide who can practice what the publication teaches? For that matter, what’s the point of teaching the invention, if students have to ask a state official whether they can practice what they have learned? Worse and worse if the state official is only in it for the money, and ignores practice in favor of finding a speculative investor willing to buy out the state’s interest to force the practice community to pay up. In such a case, the state derives its income by denying the public the right to practice what is published and taught by university faculty. That’s more akin to technology ransom rather than technology transfer.

The purpose of public record law is not to force faculty scholarship or even email to be owned by the state. Compliance with public records law is not based in ownership but in whether a record is used in the activity of government. The IT policy at the University of Washington gets this all muddled up and asserts state control of everything, releasing only what law, contract, or other policy releases–even when each of these other instruments is generally directed at claiming rather than releasing work. Even UW IP policy is constructed on what the university claims, not on what it releases, having claimed everything. The IT policy assumes a “cough up the hairball” approach–eat everything, even that which is not to be eaten, until such a time as something forces a reversal. The IT policy may be an assertion of power by the state, but it is not a power that the state authorizes, nor is it consistent with the independence faculty should have from the control of the state over their scholarship, nor is it necessary for compliance with the state’s public records act, nor is such state ownership aligned with innovation.  In short, it’s an abuse of state power and creates an adverse environment for creative work.

[I will add that the policy statement regarding ownership in the IT policy is no isolated thing. A senior (vice provost level) administrator at UW in 2001 told me much the same thing, along the lines of “the university owns everything, just we don’t tell the faculty that.”]


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