Every week, if not every day, representatives from various industries meet by themselves somewhere in the United States to develop voluntary consensus standards. Such standards may cover technical or compatibility protocols, ethical rules, standards of competence in a particular field, or other areas.
In nearly every case, the standards development organization (“SDO”) that arranges for these representatives to meet takes final credit for the resulting standard, including claiming copyright ownership. This practice, however, is usually predicated on a misunderstanding or ignorance of copyright law. Thus, the copyright claims of such sponsoring organizations are often highly suspect, though there are steps that can be taken to resolve the problem.
Copyright ownership in standards is a complex area of law. The problem is compounded when international standards are considered, since half the battle in such cases is figuring out what law applies. This article will focus on U.S. law only, which U.S. courts often apply in any event to cases with an international element to them.
Standards development involves three key aspects of copyright law: (1) joint authorship; (2) works made for hire; and (3) the transferability of copyright. How these three aspects interrelate is critical to understanding copyright in standards. Since many organizations reap profits by controlling the publication of standards and collect royalties on their sale, copyright ownership is a major — and potentially explosive — issue as to how these organizations conduct their business. Authors and SDOs alike who automatically assume that SDOs own the copyright in standards the “SDO” develops may be in for a surprise.
Copyright protection exists from the moment a work is first fixed in a tangible medium of expression. It is not necessary to register the copyright in order to have copyright protection (there are of course some advantages to registering). In general, the original author of the work initially owns the copyright.
The copyright law recognizes that a work can have more than one author. The resulting “joint work” is defined in Section 101 of the Copyright Act as a work “prepared by two or more authors with the intention that their contributions be merged into inseparable or interdependent parts of a unitary whole.” Joint authors need not work together physically, nor have an express collaboration agreement between them, nor contribute equally (though some minimal contribution is required). They do need, however, to share a preconcerted common design. Joint authors each own an undivided interest in the resulting joint work, and while each may grant a nonexclusive license of the joint work to others or otherwise profit from it, an accounting to the other joint authors may be required.
In the standards development context, each committee member who contributes to drafting or revising a standard is likely a joint author. Those who attend the meetings, but who do not contribute anything, are arguably excluded from authorship, though keeping complete records during the give-and-take of a drafting session of whose opinions or suggested edits to a standard are ultimately adopted is a well-nigh impossible task, and thus in practice each member of the committee will likely have a claim as a joint author. Where a committee member submits a draft standard that is adopted virtually verbatim, and there is no contribution of any other authors and hence no joint authorship in the final standard, the original author would remain the sole copyright owner of the standard.
Courts have recognized the joint authorship in standards developed by committee. For example, in American Dental Association v. Delta Dental Plans Association, 44 USPQ2d 1296 (7th Cir. 1997), the Seventh Circuit overturned a bizarre ruling of a judge from the Northern District of Illinois who had held that the ADA’s “Code on Dental Procedures and Nomenclature,” used to facilitate billing, was uncopyrightable because it could not possibly be sufficiently “creative” since it was written by a committee. The Seventh Circuit expressly found that committees are capable of original authorship eligible for copyright protection. As the court noted, “[b]lood is shed in the ADA’s committee about which [billing] description is preferable.”
Similarly, Victor Herbert v. The United States, 36 Fed. Cl. 299 (Fed. Cl. 1996), involved the development of the 10th Recommended Dietary Allowances (“RDA”) by a committee under the auspices of the National Academy of Sciences. The committee met regularly and reviewed various authors’ drafts “line-by-line.” Authors did not necessarily have the final say on whether proposed changes were to be made. The court found the committee as a whole, not the individual authors, had control over the draft manuscript, and that therefore the 10th RDA was a joint work and each committee member a joint author.
Works made for hire
A “work made for hire” includes a work created by an employee, as determined by the general common law of agency, acting within the scope of his employment. Some non-exhaustive factors to consider in determining whether an author is an employee for copyright purposes are: control over how, where, and with what equipment the work is done; control over the author’s schedule and other assignments; who has the right to hire assistants; and whether the employer is in the business to produce such works, provides the author with benefits, or withholds tax from the author’s pay. See Community for Creative Non-violence v. Reid, 490 U.S. 730, 109 S. Ct. 2166 (1989).
There are inevitably situations where an individual participates in the standards development process within the scope of his employment. For example, IBM might send an engineer to represent the company at an SDO meeting developing computer protocol standards. The employee’s contribution in such a case is a “work made for hire,” IBM is the “author” of the employee’s contributions, and is likely a joint author of the resulting standard (assuming there are other joint authors of the standard).
If a member of a committee drafting a standard is an employee of the SDO, the SDO would have a claim to copyright ownership as a joint author in the resulting standard through the employee (though it would still not own the entire copyright to the work). However, while an SDO employee might be treated as an “employee” for copyright purposes, it is highly doubtful that a volunteer member of a standards writing committee — who comprise the vast majority if not all of the typical committee membership — would be considered the “employee” of the SDO for copyright purposes.
A “work made for hire” can also include certain types of specially commissioned works defined by statute (e.g. tests or atlases) if agreed to in writing by the parties as being a “work made for hire.” The only one of the nine statutorily defined categories of a specially commissioned work made for hire that is arguably relevant in the standards context is a “contribution to a collective work.” Typical examples of “collective works” include periodicals, anthologies, and encyclopedias. A contribution to a collective work must be a separate and independent work in itself. In most cases, however, standards are not simply presented as a completed work to be incorporated as an independent portion of a larger collective work, but rather are submitted as drafts with the expectation that they will be modified in committee. Moreover, for the authorship in a contribution to a collective work to vest in the SDO, the parties must agree in writing that the work is a specially commissioned work made for hire. Thus, it is virtually never the case that a contribution to a standard qualifies as a specially commissioned work made for hire.
Many SDOs do not have their own employees participate actively in the adoption of standards. Such organizations may serve a secretarial function, recording the contributions of others, but their employees often do not contribute anything of substance to the standard. In such cases, the SDO is not itself an “author,” and therefore has no valid claim to copyright ownership. The only way for it to acquire copyright, therefore, is through either a license or an assignment.
Transfer Of Copyright
Under U.S. law, conveyance of a copyright generally must be in writing and signed by the author. The sole exception is for nonexclusive licenses, which may be oral or implied by conduct. Under this exception, an argument might be made that contributions to the development of a standard constitutes an implied nonexclusive license to the SDO to use the contribution in the final standard. Even so, such a nonexclusive license is not a complete assignment of the copyright, and does not result in a surrender of the author’s own copyright. Thus, while an SDO might have an implied nonexclusive license if it does not have a written instrument signed by the author conveying more, such a license does not give the SDO the sole, exclusive ownership of the author’s copyright in his contribution.
From the SDO’s perspective, it is clearly preferable to obtain an assignment of rights rather than rely on an implied nonexclusive license. An assignment eliminates any cloud on the SDO’s ability to publish the work without compensation to the contributing author. It also allows the SDO to freely resell or sublicense any rights to third parties, which the holder of an implied nonexclusive license could not lawfully do unless expressly authorized. Moreover, a nonexclusive license is revocable in the absence of consideration, jeopardizing the SDO’s rights to use the work in perpetuity. In addition, for registration purposes, the “copyright claimant” generally must be the author or assignee, not a mere nonexclusive licensee. Finally, the proper party to list in a copyright notice is the owner of the copyright, not a licensee, though an improper notice will not jeopardize the validity of the copyright.
SDOs’ Suspect Claims To Copyright Ownership
Many SDOs seem to be aware of the questionable nature of their claim to copyright ownership. They typically deal with this uncertainty in one of two ways. First, some SDOs simply assert their claim of copyright ownership. Others require participants to submit written agreements acknowledging that all copyrights in the resulting standard shall belong to the organization. In both cases, however, their efforts are usually legally insufficient.
The first strategy is followed by “SDO #1,” whose copyright policy states:
The text exploitation rights pertaining to any SDO #1 documents belong to SDO #1. The practice of using an indication of copyright in SDO #1 publications and stating that reproduction in any form requires permission in writing from the organization shall continue.
SDO #2, which often references SDO #1 publications, echoes this policy, stating that “SDO #1, as a matter of policy, holds the copyright on all SDO #1 standards.”
These assertions of copyright ownership are weak, however, which even the SDOs themselves implicitly recognize. For example, the preamble to SDO #1’s copyright policy admits that “no general answer can be given as to whether or not voluntary consensus standards would be eligible for copyright protection.” Similarly, SDO #2’s policy states that “[w]here SDO #1 does not hold the copyright, SDO #1 obtains exploitation rights that it also grants its member bodies.” This statement is a recognition that there are indeed some situations where SDO #1 does not hold the original copyright, despite its general policy, but instead must somehow “obtain” such rights. Indeed, SDO #2’s position expressly states that SDO #1, as a matter of policy — i.e. not necessarily as a matter of law — holds the copyright on the standard.
As an example of the second strategy, SDO #3 takes the reasonable precaution of requiring all contributors to sign an assignment agreement transferring the copyright from the author to SDO #3. But even SDO #3’s attempt to cover itself falls short. The document states that the person signing the form “represents that he/she has the power to make and execute this assignment.” The problem with this language is that in many instances the author will be participating within the scope of his employment, and thus the employer rather than the employee owns the copyright. The signature of the employee in such a case is virtually meaningless, unless he has actual (or perhaps apparent) authority to sign away the employer’s copyright.
Apparent authority typically exists where an employer knowingly or negligently allows an employee to assume such authority, or where third parties would assume the employee had authority by reason of the conduct of the employer. It is questionable whether such actual or apparent authority exists where an employee simply shows up at a standards development meeting and is required to sign a form before the meeting proceeds. A mere statement by the employee that he has the power to assign the employer’s copyright is likely not enough to create the requisite level of apparent authority to bind the employer. Moreover, it has been held that the apparent authority doctrine is not available as a defense to a copyright infringer (though an implied license from the author may still exist). SeePinkham V. Sara Lee Corp., 983 F.2d 824 (8th Cir. 1992).
SDO #4’s copyright policy is another example of an attempt to obtain express rights that doesn’t quite solve the problem. SDO #4 requires a contributor to include language on any written contribution that grants a license permitting SDO #4 (1) to incorporate text of the contribution in the creation of an SDO #4 standards publication, (2) to copyright in SDO #4’s name any SDO #4 standards publication that includes portions of the contribution, and (3) to allow others to reproduce in whole or in part the resulting SDO #4 publication. SDO #4’s policy makes no reference to requiring that the author sign the document licensing the work.
There are at least four problems with SDO #4’s approach. First, SDO #4’s license purports to allow SDO #4 to sublicense to others the right to reproduce in whole or in part the resulting SDO #4 standards publication. However, a nonexclusive licensee (which is the only type of licensee SDO #4 could be in the absence of a writing signed by the author, which SDO #4 does not expressly require) is not permitted to sublicense the work. Thus, while SDO #4 could unilaterally establish a policy that gives it ammunition to argue that authors have given it a nonexclusive license, it cannot take that license and sublicense others to use the work.
Second, the policy does not cover the verbal contributions of members of a standards writing body, which may comprise a significant amount of the work that goes into a final standard.
Third, the policy may not adequately cover the rather common employer-as-author situation that arises where the author’s contribution was written (or his presence and verbal comments were made) within the scope of his employment.
Fourth, the license allows SDO #4 “to copyright in SDO #4’s name” any resulting standard. Under the copyright law, copyright exists from the moment a work is first fixed in a tangible medium of expression. Such rights come into being as a matter of law. Perhaps the SDO #4 language is designed to allow SDO #4 to register a copyright in SDO #4’s name, but that is essentially a matter of administrative convenience and does not itself create a copyright. In any event, a nonexclusive licensee like SDO #4 would be in the absence of a writing signed by the author cannot legally register the licensed work.
Unquestionably, SDOs perform a valuable service by enabling members of industry to come together and develop common ground rules or technical standards. That work should be encouraged to continue. However, operating with a blind eye towards potential copyright problems with the finished standards may be to court disaster. It is better to recognize the issues and to take affirmative steps to ensure that an author or joint author of a standard does not attempt to hijack the standard or demand royalties later, thereby diminishing the usefulness of the standard. From the authors’ perspective, they should be made aware that they do have rights in their creative ingenuity in assisting in the development of standards, and while the assignment of those rights perhaps should be freely given for the greater good, authors should at least know what they are giving.
Keith Barritt is Of Counsel in the Washington, DC office of Fish & Richardson P.C. His practice is primarily focused on trademark law before the U.S. Patent and Trademark Office, medical device regulation before the U.S. Food and Drug Administration, and communications law before the Federal Communications Commission.