Articles on Law, Science, and Engineering

Custodians of Information

by Charles Walter and Edward P. Richards, III, 15 IEEE Engineering in Medicine and Biology Magazine #5, pg. 138 (Sep/Oct 1996)

The control of valuable technical information determines its beneficiaries. The "owner" may exploit the information for a profit or share it with colleagues and the general public. When used for a profit, benefits focus on monetary considerations and the narrow interests of the "owner" of the information. When information is in the public domain, benefits extend to everyone.

Information, of course, exists whether there is an "owner" or not. Until discovered, information "belongs" to everyone but can be used by no one. After discovery, the ownership of the information does not change, but who can use it does. Thus, one who discovers information becomes its first custodian, but not its owner.

In his thoughtful "GNU Manifesto,"[1] Professor Richard Stallman, a well-known computer scientist, points out that the desire to be rewarded for one's creativity does not justify depriving the world of that creativity, and that creativity is a social contribution only insofar as society is free to use the results. Indeed, if the initial custodians of valuable information deserve to be rewarded for their creativity, by the same token they deserve to be punished if they restrict the use of it.

Nevertheless, most legal systems recognize at least three methods for the control of information. Trade secret law recognizes the right of the first custodian of information to keep the information secret and be protected from misappropriation by a subsequent custodian of the same information. The first custodian of information comprising a significant advance in a useful art is permitted to use the information as embodied in the specific advancement exclusively for a period of time in return for disclosing it promptly to the public in a patent. And first custodian of newly-created forms of information set in a tangible form is permitted to use the creation exclusively for a period of time as a copyright.

Underpinning the justification for protecting trade secrets, patents and copyrights is the tenuous assumption that the discovery of unknown information depends strongly upon economic incentives. Professor Stallman suggests that this assumption, even if valid, does not justify monopoly reward systems such as software copyrights. He points out that, if users of innovative technology (software) would rather pay for the technology and service than get it free without the service, then a company ought to be profitable which provides service to people who have obtained the technology free. Hence, innovators who have created the technology would still do so and profit by selling services to users who want to pay for them, rather than by collecting from everyone through the present system and its "tremendous cost to society."1

Nevertheless, most legal systems accept this assumption and vest at least some control of newly discovered information with the entity that paid for its discovery regardless of costs to society. The most important instance where the control of information is transferred is where the first custodian of the information is an individual employed by a business. Trade secrets are usually first known to an employee, but their control vests in the employer. Inventions are made by individuals, often employees, but "shop right" control vests in the employer, and often, the employer obtains full control of employee-inventions through assignments imposed as a condition of employment. The employer is actually deemed to be the "author" of copyrighted works for hire even though the works were actually created by employees.

Thus, employers control trade secrets, as well as most patents and copyrights, even though the initial custodian of the underlying information was an individual employee. The rationale for vesting control in the employer rather than the individual who actually exhibits the genius necessary for the discovery of the information is that if employers control valuable information, they will continue to pay employees to advance the useful arts and science, and the public will ultimately benefit. In what follows, we refer to this assumption as the "hired genius theory."

Trade secrets are scientific or technical information that has value and that the discoverer has taken measures to prevent from becoming available to others. In the context of the teaching profession, trade secrets may include designs, teaching procedures, tests, handouts, and the like. In the context of the research profession, trade secrets may include data, experimental protocols, calculations, and the like.

Employers control trade secret information by keeping it secret. An employee who frustrates this goal may face serious legal consequences, even though the employee does not benefit from his action. For example, in Texas and many other jurisdictions, merely making a copy of a document representing a trade secret is a felony, regardless of the value of the information copied.

There can be no doubt that the law vests control of teaching and research trade secret information discovered or developed by faculties and other employees to the schools that employ them. Unnecessary competition due to outrageously redundant federally-funded research, especially in biomedical areas, has changed the way many researchers share data with other scientists. The increasing paranoia that another scientist will pilfer data and/or ideas has led to policies wherein measures are taken to prevent at least a portion of the data from becoming available to others. Under trade secret law, such information is clearly the property of the school and cannot be misappropriated by others. This includes the employee-custodian of the information. If the faculty member who discovered the information knowingly communicates or transmits it to another without the effective consent of the school, the faculty member may be prosecuted for a felony, regardless of whether harm results.

For most scholars, the concept of keeping valuable information secret is repugnant to their mission. For example, rather than worry about having their data pilfered, the goal of most researchers is to disclose their newly-discovered information as soon as possible after they become its custodian.

According to the provisions of basic patent law, the hired genius theory need not apply to patents. In fact, the employee-inventor must appear as the applicant for a patent on his or her invention. Absent a contractual obligation to assign the patent to the employer, the employer owns no rights to the patent. The exception to this general rule is that when the employee uses the employer's facilities to make the invention, the employer may exercise the right to practice the invention without paying a royalty to the inventor. An exception might also arise where the employee was hired specifically to invent the invention.

Most universities where significant research is conducted use contract law to change this default rule for patent ownership. Usually this is done by requiring an assignment of future patentable discoveries to the university as a condition of employment. Since the inventor still must be the applicant for the patent, there is also usually a requirement that the inventor cooperate with the university in drafting and submitting the application. There may also be provisions purporting to prevent the inventor from making certain types of disclosures or taking other steps that would jeopardize the patentability of the invention. Thus, even if it is not a breach of the applicable trade secret statute, it may be a breach of his employment contract for a faculty member to publish his research results, theories, etc. without the consent of the school that employs him.

Usually, assignment clauses are effective to transfer control of teaching and research discoveries from their inventor-faculties to the employer-school. However, where such clauses are imposed unlawfully, they may be ignored by the courts and not enforced. For example, a tenured research professor, with existing rights to his invention under the default rules of patent ownership, may not be required to agree to an assignment as a condition of continuing his tenured status with the school.

For most scholars, the concept of giving anyone, even one's employer, the right to prevent or delay others from practicing an art is repugnant to their mission. One basis for this reluctance is that extracting money from users of inventions to reward innovation may reduce the amount of wealth that humanity derives from the technology. Scholarship is about enriching the wealth of humanity, not diminishing it. Thus, principled individuals will exercise their option to decline contractual arrangements that are professionally or societially unsatisfactory, thereby freeing themselves to disclose their newly-discovered information as soon as possible after they become its custodian.

In contrast to patents, the hired genius theory applies to copyrights. In fact, according to the copyright statute, the employer is deemed the "author" of copyrightable creations made as a work for hire. The courts have construed the phrase "work for hire" as just about any work created by an employee in the course and scope of his or her employment. Absent a contractual obligation to re-assign the copyright back to the employee, the employer owns all rights to the copyright.

Due principally to a California case styled Williams v. Weissner, 273 Cal.App.2d 726, 78 Cal. Rptr. 542 (1969), there is a "teacher exception" to the statutory work-for-hire rule. The case dealt with copyright ownership of a teacher's lectures, but was based in part on the proposition that, to deny a professor the rights in his lectures because they were a work for hire would have the undesirable consequence of restraining his mobility in what is a peripatetic profession.

Does the "teacher exception" extend to materials that the teacher has prepared for publication? Williams had prepared lecture notes for his own personal use in delivering his lecture, but he did not attempt to publish them. The notes in the case had been made during one of Williams' classes by an individual who heard the oral presentation. It is possible that the "teacher exception" would not have applied if Williams had reduced his lecture notes to manuscript form and submitted it for publication.

Does the "teacher exception" apply to copyrightable materials created during the course of academic research? To deny a research professor the rights in his experimental protocols, data bases, computer programs, etc. because they are works for hire would have the undesirable consequence of restraining his mobility in what is still a somewhat peripatetic profession.

For most scholars it would also be repugnant to give anyone the right to prevent others from using copyrightable research materials. For example, Professor Stallman believes that computer programs should be shared free of charge. In his "GNU Manifesto" he states, "Copying all or part of a program is as natural to a programmer as breathing, and as productive. It ought to be free."1 He explains that sharing of computer programs is essential to their development, and that, instead of wasteful duplication of programming efforts that occur under the current system, the efforts could go instead into advancing the state of the art.

Should the hired genius theory be applied to academic employers and faculty employees?

The short answer is "no." Schools exist to serve the public interest and are accountable on that basis to the public. Therefore, vesting control of information developed by faculty to the employer school is logical only to the extent that it is necessary to carry out the school's goal of serving the public interest. To the extent that vesting control of information to schools results in the misuse of the information for the school's independent support, the policy supports the undesirable end of making the school less accountable to the public, and is therefore counterproductive.

Does this mean that faculty should control the information they discover in their classrooms and laboratories?

The short answer is again, "no." Faculty employees are, by definition, agents of their school's goals, including the primary goal of serving the public interest. Therefore, vesting control of information developed by school employees to the employee is logical only to the extent that it is necessary to carry out the school's goal of serving the public interest. Given that the goal of every school employment contract is service, teaching and research, there is ample incentive in the ordinary employee relationship with the school to foster creativity, inventiveness and scholarship.

To the extent that there is a trend toward less accountability for public service, teaching and research, both the employer schools and their faculties are responding to a deteriorating environment sponsoring less loyalty and dedication to the underpinning values of public service, and greater exploitation of opportunities to extract for personal gain whatever happens to be available at the moment from the public trough. The process underlying this trend is auto-catalytic. It feeds relentlessly on itself, and will eventually end catastrophically.

However, no scientist can believe science is served by scientists who keep data secret or who use public support for science to enrich themselves. No educator can believe education is served by teachers who keep teaching methods secret, or who use public support to enrich themselves at the expense of other teaching environments. There is a long-standing tradition amongst educators, research scientists and other scholars supporting the proposition that standing tall enough amongst one's piers to enable another to stand on your shoulders above the rest is ample reward for creativity and scholarship. Policies supporting the full revival of that tradition are far more important than policies designed to make academicians rich at the expense of the public that has traditionally supported teaching and research. We believe that such self-serving policies will neither achieve the goal of advancing the useful arts and science, nor survive public scrutiny.

Endnotes

1. R. Stallman, GNU EMACS Manual 175-84 (1st Ed. 1985).[return to text]