First, the final chip configuration represents an integration of a number of individual "drawings", and the final product may be different from each of the individual layouts in itself.

Second, the diagrams, sheets and masks which we will register are not exposed on the market. When unauthorized duplication occurs it is usually done from the finished chip and not from the layout, drawing, or mask.

Finally, the extent to which copyright in the diagram sheet or mask protects against duplication from the chip is uncertain under 113(b) of the Copyright Act.

Mr. Chairman, I don't think it's necessary to read the section. As far as legislative history, that appears on pages 3 and 4 of this statement.

Turning from the layout, masks and schematics to the actual configuration of the chips, or in the words of H.R. 1007, "the imprinted patterns on integrated circuit chips"-the Copyright Office will generally refuse registration. This practice-which apparently goes back to a much earlier generation of printed circuits-is based on essentially two related facts.

The first basis for our rejection is the definition of "pictorial, graphic, or sculptural works", which appears in section 101 of the new Copyright Act, and that section itself is a codification of an earlier regulation of the Copyright Office. That regulation in turn was a codification of the practices of the Office going back to 1909. Again, Mr. Chairman, the text of the section and your committee's own legislative explanation of its operation is set forth on page 5 and 6 of the prepared statement and I don't think it necessary to read them to you. Perhaps it might be best paraphrased by referring to your committee's report where we said the copyright under the bill will not extend to the design of a refrigerator or an automobile or an airplane, and that if we add the words "computer" or "minicomputer" to that, that in a simplistic way is the basis upon which we refuse registration of the configuration of the chip. The second factor related to the definition in the statute is case law. There have been a number of decisions which are cited in page 6 of the statement which raise a number of significant legal and policy questions about the copyrightability of elements of mechanical and scientific devices. These cases raised questions, and in fact the two cases cited refused protection to works which were essentially parts of the calculating device. The court was concerned that by extending copyright protection to that part it might be extending the patent monopoly given to the device as a whole. Third, with respect to computer programs "stored" in the chips, the Copyright Office will generally register claims to copyright in these works. Under our deposit regulations, a visually perceptible printout of the program, rather than the chip itself, must accompany the application.

Again, Mr. Chairman, the merger of hardware and software within the chip has raised questions within the office about whether the program stored in the chip is a program as we traditionally consider it; however, again, as with the drawings, we generally make registration.

It is our understanding the proponents of protection for chip patterns also usually support copyright protection of computer pro

grams; however, they apparently believe that copyright in programs does not offer adequate protection against duplication of the chips. The reasons for this are not always clearly expressed, or at least I have not seen them clearly expressed, but we would imagine they are one or more of the following:

First, the chips for which they seek protection include types that do not embody computer programs at the time they are exposed to duplication. For example, there are "blank" chips or unprogramed chips which are sold for programing by the customer and might be copied before the program is loaded onto the chip, or there are in fact chips which are not designed to hold programed material at all.

Second, unauthorized duplicators of even "programed" chips may, we understand, avoid duplication of that part of the chip which carries the program; and finally, the copyright owner in the program may not be the same as the claimant of rights in the chip pattern design.

For these reasons, Mr. Chairman, it's our understanding that the proponents of protection in H.R. 1007 would not be comfortable merely resting upon copyright protection for the computer program.

In adopting the practices I have described, most notably our exclusion of chip configurations, the Copyright Office believes it is following the language and spirit of the Copyright Act, and particularly the dividing line between artistic works and "industrial designs," which your committee referred to in it's 1976 report. As I have indicated, we do not mean to detract from the efforts of chip designers. We believe that these efforts represent substantial originality, ingenuity, creativity, and investment, and in fact deserve protection against the increased and improved techniques of unauthorized duplication which have led to the introduction of H.R. 1007. We do, however, have a number of questions pertaining to the accuracy of our assumptions and the nature and scope of protection. These questions may be divided into five areas. Specifically, we respectfully suggest:

1. That the subcommittee assure itself that within the constraints of chip purpose and size the designer's choice of a particular layout and the representation of the designer's labors in the "masks" and "patterns" is not dictated by the function to be performed by the chip and does represent the creative choice from among different possibilities. This standard is implicit in our assumption that the words to be protected are the results of "authorship." If this assumption is not the case, there would be a very clear danger that the protection could go beyond the purpose of copyright in protecting not only the "expression" of the designer's concept, but the concept itself, and the principle of operation. Concepts, principles, and the like, have never been considered to be a part of the copyright protection, and are not intended to be.

2. The subcommittee should explore the relationships among, first, schematic drawings, mylar sheets, photographic masks and related devices; second the "patterns" imprinted on the chips; third, computer programs, if any, stored in the chips; and finally, computer programs used in generating the finished chip or its intermediate stages. The subcommittee should consider whether

protection already to each of these elements under existing mechanical, process, design patent, or copyright law, offers sufficient protection against chip duplication. If they do not, you might consider whether any policy considerations which underlie this absence.

3. Perhaps most importantly, the subcommittee should consider whether, in light of the existing and anticipated industry structure and technology, copyright protection of the "masks" and "imprinted patterns" should be subject to specific limitations regarding the term of protection, the scope of rights, or the nature of remedies against infringement. The precedents for this type of limitation, is a limitation adapted for specific purposes to particular works, include a number that this subcommittee is familiar with:

(a) The Sound Recordings Amendment of 1971, when Congress first extended copyright protection to sound recordings, it limited the right in sound recordings to protection against "dubbing" and direct electronic duplication, and expressly excluded protection against imitation-so-called "sound-alike" records. Similarly, Congress did not accord a performance right in the sound recording to recording artists or record producers. These limitations demonstrated Congress intent to provide effective protection and remedies for the industrial problem of record piracy, without altering other legal relationships under copyright between music copyright owners, record producers and broadcasters. The new Copyright Act, as the chairman is aware, continued these limitations on the reproduction of performance rights which are usually accorded by copyright.

(b) The Report of the National Commission on New Technological Uses of Copyrighted Works. The 1978 report did recommend that the copyright law be amended to make it clear that copyright protection extends to computer programs. At the same time, the report suggested that the usual reproduction and adaptation rights of copyright owners be somewhat limited in order to assure that the rightful possessors of copies of programs use or adopt them for their use.

(c) Perhaps most clearly to the point, design legislation. For many years, bills have been introduced in Congress to provide copyright protection for industrial designs, or in the words of the bills, "original ornamental designs of useful articles." H.R. 2706, recently introduced by Mr. Railsback on March 7, is the most recent of these proposals. As this subcommittee will recall, a similar proposal appeared as title II of the copyright revision bill, but, at the urging of your committee, was reserved for additional congressional consideration. In a real sense, Mr. Chairman, the hearings being held today on H.R. 1007 may be considered to be the first step in the reconsideration of industrial design protection promised in your committee's report.

Each of the design bills would have accorded copyright protection for industrial designs. That is, protection based upon principle of originality rather than on the principle of novelty and invention embodied in the patent laws; however, in response to particular considerations of consumer preference, market structure and economic impact, the bills also included specific departures from ordinary copyright principles. These departures or limitations included

such devices as a shortened term of protection specifically related to the need for protection; the protection of innocent retailers from liability for infringement. The traditional copyright principle has been that seller of an infringing article is as liable as its maker. The design bills would remove the liability of the seller; and provisions for administrative cancellation procedures.

The subcommittee might consider whether limitations of this nature are appropriate in the area of chip designs. For example, copyright protection under the law that exists for 75 years from publication, or 100 years from creation of the chip, assuming they are works made for hire. The subcommittee may conclude that this is much too long protection for a device, although protection for some shortened period is indeed warranted.

Another precedent for specific limitations, Mr. Chairman, is the concept of typeface protection. The committee will recall that during the debates over copyright protection for typeface designs, it was suggested that because of the nature of the typographic industry and the possible reach of ordinary copyright remedies to those who produced or distributed books set in "infringing type," and potential first amendment implications, typeface protection should be subject to certain limits which included compulsory licensing of reproduction rights, and a limitation of remedies to those who directly duplicated fonts, and not extending to those who produced works distributed in infringing type. As with the case of design protection, your committee deferred the entire matter of typeface protection for more thorough consideration.

There are numerous other instances where Congress has concluded that the "usual" attributes of copyright protection, that is the absolute rights of reproduction, adaptation, performance and display, and the usual term of those rights, should be limited in application to particular works for particular uses. These are set forth in section 107 through 118 of the new law.

Significantly, and bringing the point home to chip design, some earlier discussions of chip piracy have suggested that certain forms of partial or reverse duplication should not be inhibited, in order to assure development of the art. There were specific suggestions to this effect in Mr. Edwards' statement in the Congressional Record of October 14, 1978.

Another area we suggest the subcommittee look into is the technical accuracy of certain terms employed in the proposed bill. We note that the title of the bill refers to "semiconductor chips," while its text identifies "integrated circuit chips," and with our limited knowledge we do understand that these different terms may have differing connotations, and we suggest that this be clarified.

The bill also refers to "imprinted patterns-incorporated in useful articles". We believe this suggests some clarification as to whether protection is to be limited to the surface appearance of the chip as implied in the use of the term "chip topography" in our prior experience with these words, or whether protection is to extend to the subsurface configuration of finished chips.

Finally, gentlemen, we believe the committee should consider expressly limiting the amendment, at least in terms of the "imprinted patterns", to those chips created after its effective date. This has been a common part of amendments in the past which

bring new subject matter or new rights, such as sound recordings, musical compositions, performance rights in dramas, and photographs, under the copyright law. In considering this question, the committee will, of course, want to explore the effect of such a limitation on the existing chips, but we believe you should also explore the expected useful life of the existing chips, the impact of retroactive protection upon uses of the chips, and indeed, earlier forms of printed circuits made before the effective date, and a general copyright policy against recapturing works from the public domain.

In concluding, I would point out that the issues I have discussed today involve the application of traditional principles of copyright law to works, and indeed to industries and markets, whose technology is still very new. But the sophistication and the newness of the subject matter should not conceal the familiarity of the basic issue: Whether existing law is adequate to offer protection against rapidly developing technologies of unauthorized reproduction, or whether it should be amended to do so.

Your committee has dealt with this specific type of issue twice in recent years, with differing results: In the early days of record piracy, although it has been a problem, it did not have the major impact on the industry it now has because to be a pirate you had to set up a record pressing plant with extensive record pressing equipment; however, with the development of consumer acceptance of tape, and the development of high speed tape duplicators, the need for protection became greater because it was easier to become a pirate; you bought a tape machine and set up operation in your garage. At that point the record industry pressed hard for copyright protection and your committee and the Congress responded affirmatively with the Sound Recordings Amendment of 1971. The opposite result was reached with respect to typeface. Again, in earlier days, typeface piracy was a problem, but assumed lesser proportions because the pirate of typeface had to go to considerable effort in reproducing the matrices; however, when typeface design became embodied in photographic form, to become a pirate all you needed was a camera, and you copied the photo. The rapid progress of the technology of duplication led the typeface interests to come to Congress for copyright protection. In this case, your committee responded by saying you wished to consider the matter further. We suggest this is the same type of issue, the type of issue faced in record piracy and the type of issue faced in typeface design, that your committee faces at this time.

Thank you for the opportunity to appear before you today, and on behalf of the Register and the Office, I'll be pleased to answer any questions you might have.

Mr. KASTENMEIER. Thank you, Mr. Baumgarten.

[The prepared statement of Mr. Baumgarten follows:]

STATEMENT OF JON BAUMGARTEN, GENERAL COUNSEL, U.S. COPYRIGHT OFFICE Mr. Chairman and members of the subcommittee, my name is Jon Baumgarten and I am General Counsel of the United States Copyright Office. I appear here today in support of the principle of protection "for imprinted design patterns on semiconductor chips" reflected in H.R. 1007; at the same time, I plan to suggest several issues we believe the subcommittee should consider before approving that bill in its present form.