the act in 1909, the result was that, for a period of 70 years, phonograph records were not considered copies of musical compositions. That, of course, has been changed by the current act. But I mention this just to show the fundamental importance of this concept. The bill provides that the term "copy," as used to the statute, includes chip products with respect to only 9 of the 36 sections in the copyright law where the term is used. The bill would thus create a distinction between those sections of the law in which chips are treated as the equivalent of copies and those in which they are not. Of crucial significance, we think, is the point that section 101, which contains the definitions of the act, is not included in the nine sections specified. The consequences of this seem grave. We question whether a mask work can be published under the act, since the term "copies" in the definition of publication does not include chip products. If the object embodying the copyrighted work is not capable of publication, the notice requirements of the act apparently do not apply. We also question whether chip products would constitute copies of computer programs and other works embodied therein because of the bill. We realize this result is unintended. In the Senate subcommittee version, an effort has been made to address this question, but we believe that there is a fundamental problem concerning the concept of copies and the way in which chip products would be treated under the bill. As to the subject matter question, we do note that this would be the first time that Congress would protect a useful article under the Copyright Act. We don't think that this is a constitutional problem, per se, but it is a matter of policy. We do note that the bill attempts to establish a constitutional basis for the semiconductor product in its definition of semiconductor product, but there is no attempt to specify the constitutional basis for the subject matter of protection which is the mask work. This is perhaps a technical matter that could be addressed, but it does exist in the bill now. Another point of concern with respect to the pending bill is the use right. For the first time there would be an exclusive right to use a copyrighted work. This is unprecedented in American copyright law, and in any copyright law that we are aware of. It is a right that is found, I believe, in the patent law, but it is alien to copyright law. We would recommend its elimination. The use right has been eliminated in the Senate subcommittee version. With respect to the question of retroactive protection, the initial impression of the Copyright Office when we testified in the Senate on the companion bill, S. 1201, was that neither H.R. 1028 nor S. 1201 was intended to protect mask works retroactively. We now realize that because of the drafting of the effective date provision, it would appear that the bill does protect mask works retroactively. There are limitations on the protection with respect to particular material objects embodying the mask works, depending on when the infringement occurred, but the mask work itself would be protected, apparently, irrespective of when it was created or had been initially distributed commercially. The Copyright Office believes that retroactivity clauses generally should be scrutinized for basic fairness and appropriateness, and their inclusion in a statute almost always requires safeguard clauses and other provisions to avoid unfairness and actual damage to those whose rights or privileges are curtailed by the law. Unless they are very carefully crafted, retroactivity clauses invite litigation. The Copyright Office is not persuaded that the semiconductor industry has sustained the burden of showing the necessity for retroactive protection of mask works. To date, our investigation suggests that this is not a constitutional problem, per se, but it is a matter of fairness and policy as to whether protection should be made retroactive to subject matter that was created and commercially distributed years before the law was enacted to protect them. In view of the problems that we have noted with the pending bill, and our questions concerning protection of useful articles under the Copyright Act, the Copyright Office considered alternative approaches to protection for semiconductors. In our Senate testimony we mentioned the possibility of misappropriation protection, and we also discussed a possible design approach in general. Following that testimony, we were asked to prepare a draft bill that might accord protection for chip designs under design principles, and it is included in our statement as appendix C of the statement. The Copyright Office is not precisely endorsing this particular approach. We were asked to prepare it by a Member of Congress. We think in general it is better to turn to sui generis legislation than to try to craft the Copyright Act to fit semiconductor chips. At the same time we don't entirely abandon the copyright approach. If it is the wish of Congress, after reviewing the whole matter and considering the appropriate balance to be achieved, that there should be protection under the Copyright Act, the Copyright Office certainly would be pleased to work with the subcommittee in correcting the technical problems with the bill and in fashioning a suitable administration for the protection if the Copyright Office is designated as the administering agency. Under a design approach, which we tend to prefer, Congress would establish a new form of legal protection specifically for the design of semiconductors. It could be in a separate, independent chapter of title 17 of the United States Code, but separate from chapters 1 through 8, which now represent the Copyright Act. Protection would be accorded under a standard of originality similar to that of the Copyright Act, but you would not have the whole body of copyright precedent for 150 or 200 years, that would have to be gaged, assessed, and considered in determining whether particular mask works did constitute original works of authorship-the statutory standard for determining whether you have a protectable work under the Copyright Act. The design approach would directly confront the utilitarian nature of semiconductor chips, and would simply accord protection on the basis of the skill, labor, and investment employed in developing original designs. The constitutional basis, although unstated, would be the patentcopyright clause of the Constitution, probably more the writing aspect, but perhaps an amalgam of writings and discoveries. The exclusive rights conferred would be: To make, have made or import for sale or use in trade, and to sell or distribute for sale or for use in trade, the chip product embodying the protected design. Although the terminology is different, we think these are basically similar to the rights proposed for mask works in H.R. 1028— except for the use right. Protection would also endure for 10 years from the date the design was registered or first made public, whichever occurs first. Protection would be accorded only prospectively under the draft design bill. The registration system is somewhat simplified, and the certificate of registration would constitute prima facie evidence of the facts stated, but would not be prima facie evidence of the validity of the protected design, per se, as is the case under the Copyright Act, where registration within 5 years of publication means that the certificate is prima facia evidence of validity. There would be no examination of the prior art under the draft design bill following copyright-like principles, but a defendant in an infringement action could cite the prior art and shift the burden of proof back to the plaintiff. The remedies are similar to those under the Copyright Act, but no provision has been made for criminal infringement. That could be added, of course, if that were the wish of the subcommittee. There are penalties for false marking, fraud, and false representation. Finally, I would mention some of the advantages of a design approach. It establishes an especially tailored scheme of protection solely for the design of a semiconductor chips, avoiding the problems of fitting traditional copyright policies and principles to the previously unprotected useful article. It would avoid the technical, substantive problems that we have discussed, such as the problem of what is a copy and the relationship between the copyright in a computer program fixed on semiconductor chip material and copyright in the mask work. There would be a specific definition of design of a semiconductor chip product which we think is somewhat clearer than the concepts of mask work and images of a mask work as used in the pending bill. In conclusion, Mr. Chairman, we fully support the general proposal to protect the innovators of semiconductor chips, but we tend to believe a better system of protection could be achieved under a design approach. Nevertheless, we would be happy to work with you, if you elect to follow a copyright approach. Thank you very much. That ends my formal statement. If you have any questions, I will try to respond. [The statement of Ms. Schrader follows:] STATEMENT OF DOROTHY SCHRADER ASSOCIATE REGISTER OF COPYRIGHTS Before the Subcommittee on Courts, Civil Liberties Mr. Chairman and members of the Subcommittee, I am Dorothy Schrader, Associate Register of Copyrights for Legal Affairs and General Counsel of the Copyright Office. I thank you and the Subcommittee staff for giving me the opportunity to appear before you on H.R. 1028, a bill to protect semiconductor chips and masks against unauthorized duplication. The Copyright Office supports the principle of protection for original semiconductor chips and masks (hereafter generally referred to as "chips"). Same form of protection is just and necessary. The Office is, however, not certain that the Copyright Act is the best answer to this need for protection, and we have doubts in any event about some features of the proposed bill. Other features of the bill do represent an improvement in comparison with the approach considered by Congress in 1979,1/ which I will discuss later. 1. H.R. 1007, 96th Congress, First Session (1979) would have simply provided that the "photographic masks" and "imprinted patterns" on integrated circuit chips were copyrightable as "pictorial, graphic, and sculptural works." |