tions of our works. However, my reference to the word "copy" does raise an additional technical point, because this bill says the word "copy" applies to chips only for the purpose of certain sections of the law and these do not include our basic right to reproduce. As soon as we solve this "copy" problem, we'll be fairly confident that we'll be able to handle the issue.

Mr. KINDNESS. Thank you.

Mr. KASTENMEIER. I have a technological question or technical question. Am I to understand that these chips have a capacitystorage informational bits-which may or may not be the same? If you produce a given chip, whether it has 256 K capacity or something else, then you can introduce into that chip different pieces of information than you would another chip made precisely the same way, so that it becomes a more or less storage factor that intrinsically is not the book, but merely the device for accepting, retaining, and for discharging bits of information. Therefore, it's not an analog, as you suggest, you cannot say that these are the same, that one is the chip and one is the published work. They are totally different in that regard; is that the case?

Mr. BAUMGARTEN. Once you make the vehicle-

Mr. KASTENMEIER. We're talking about the vehicle. We haven't talked about a vehicle, because almost anything can be stored in here.

Mr. BAUMGARTEN. From our viewpoint, we do view the chip as a vehicle. That's the way we see it. Something that will carry our books and software. The chip industry doesn't see it as only a vehicle; they see it as a work in and of itself. But I think that in concept, notwithstanding this difference of viewpoint between the publishers, the chip industry and the members of this committee, no one intends to limit the protection now available to books and programs because of the limits of protection on the chips. And I think we have to make that clear in the legislation.

Mr. KASTENMEIER. I don't quarrel with that, but I want to know the theory we're talking about in comparative terms; that one is not the other. The chip may, indeed, contain proprietary works, may contain book A, but that chip or some other chip may contain totally different bits of information, even though their capacity may be the same or made by the same manufacturer.

Mr. BAUMGARTEN. I would agree with that description. It is a vehicle.

Mr. KASTENMEIER. It's like a library with shelves on it. So it's not necessary to compare the library to the book, but merely the library may contain a book or may not contain a book; and, of course, the question of whether librarians or chips will disseminate these books in an unauthorized way is another matter. I'll yield to Mr. Sawyer.

Mr. SAWYER. I haven't got a good enough grasp of this to ask any questions. To tell you the truth, I'm more interested in listening, until I think I understand it better than I do now.

Mr. KASTENMEIER. The gentleman from Illinois, Mr. Hyde.
Mr. Hyde. No questions.

Mr. KASTENMEIER. Let me rephrase what I tried to somewhat discern from you at the outset. You have two interests. One is, protect proprietors of published works, no matter what legislative device

may go through to give protection to this particular industry. And the second, which may be completely compatible with the first, is the larger point of what should an appropriate vehicle be, not merely should it have a disclaimer saying something like, "Copyrighted in a mask" or "mask work shown," then do or affect any other copyright of any other author for some other language.

The question really is, what should the vehicle be? Is this, in fact, a copyright in a normal sense, or should it be what we talk about and have talked about for 3 years of design debate. Even the authors of the legislation, concede that this is not typical copyright, because they've only asked for a term of 10 years. They could see that this was not normal copyrightable material, in terms of protection.

Would you be more comfortable, conceptually, if it were treated as design legislation, much as the old design proposals affecting lamps and other things in the Mazer case we used to talk about in the last two decades? Is that what you would like it to take the form of?

Mr. BAUMGARTEN. Yes, sir, but may I add, Mr. Chairman, that I don't think one would have to wait for the passage of design legislation and include chips in it. One could just as easily start design legislation with chips. With that qualification, yes, that's what we're talking about.

Mr. KASTENMEIER. I think the sponsors of the legislation fear, again unnecessarily, that that would impair movement of the legislation. I don't know if I speak for them, but if it were the case, that this could take the shape of design legislation and be opened up either by analogy or by actual inclusion to other forms of designsand we have talked about many forms, the ability designs and typeface and all forms of design which could either be included or not included, that nevertheless someone who's been in the Copyright Office and looks at the field more generally-would that be the way to go?

Mr. BAUMGARTEN. Understanding that I'm now representing AAP and not speaking as the General Counsel of the Copyright Office, the chairman has a very good memory. You mentioned typeface design with very good reason. The publishers had a very difficult time with copyright for typeface. I'm not unconcerned about this, that it might slow things down. But if someone desired to go forward with chip design legislation, I've worked with this committee and its staff long enough to know that it's hardly beyond their ability to accomplish well-fashioned and rapid passage.

Mr. KASTENMEIER. Yes; I think I tend to agree with that analysis. you have any feeling for it, what do you think the turnoff point would be?

If

Mr. BAUMGARTEN. I don't purport to be an expert in this area. I think the chip industry has settled on 10 years as an adequate length of time for chip, and I don't question that.

Mr. KASTENMEIER. What does the design legislation hold for turnoff?

Mr. BAUMGARTEN. I don't recall specifically, but we can supply that.

Mr. KASTENMEIER. I think it's something like 7 years.

Mr. BAUMGARTEN. Something like that.

Mr. KASTENMEIER. May I ask this. From your experience as General Counsel with the Copyright Office, is there anything under H.R. 1028 that would preclude or hinder that Office from satisfying its obligation of the bill?

Mr. BAUMGARTEN. I don't believe so.

Mr. KASTENMEIER. OK. Going back to the other question of designs, using a design concept for protection for semiconductor chips, are there dissimilarities between design protection for semiconductor chips and other utilitarian under the Mazer case?

Mr. BAUMGARTEN. It's somewhat inaccurate to refer to the Mazer case or some of the other illustrations that were given in Congressman Mineta's statement. In the Mazer case, the work was held copyrightable on the grounds that it was clearly a work of art, though it happened to be embodied in a statute.

Mr. KASTENMEIER. I should not have referred to the Mazer case. I guess I should have referred to the design legislation that followed out of that.

Mr. BAUMGARTEN. I believe that there may be some differences that would require legislative adjustment. There were references, for example, in the design bill to commonplace symbols and the like, which may not be particularly appropriate to the configuration of the chip, since some may be designed as a series of rectangles, even though they go from the surface to the subsurface. I believe there may be points of language in the design bill that may require changes, but I believe the overall objective of the bill, to give protection without the long and costly experience of patent prosecution, may be appropriate to chip protection.

Mr. KASTENMEIER. One of the reasons I raise these questions is because the gentleman from California, Mr. Moorhead, has introduced a bill on general design protection and this may be relevant in the same terms.

Mr. BAUMGARTEN. I believe we referred to that in our statements.

Mr. KASTENMEIER. I don't have any further questions. You have been very helpful, and I thank you very much.

Mr. BAUMGARTEN. Thank you.

Mr. KASTENMEIER. Our final witness today is F. Thomas Dunlap, Jr., who is secretary of the Intel Corp. Mr. Dunlap worked in the industry for 9 years as both an engineer and an attorney, and we welcome Mr. Dunlap. You may proceed as you wish. You might identify your colleague.

TESTIMONY OF F. THOMAS DUNLAP, JR., CORPORATE COUNSEL AND SECRETARY OF INTEL CORP., SEMICONDUCTOR INDUSTRY ASSOCIATION, ACCOMPANIED BY RICHARD STERN, COPYRIGHT COUNSEL

Mr. DUNLAP. Yes, I am accompanied by Mr. Stern, who is our copyright counsel here in Washington. [Complete statement follows:]

TESTIMONY OF F. THOMAS DUNLAP, JR., CORPORATE COUNSEL AND SECRETARY OF

INTEL CORP.

I represent Intel Corporation, a manufacturer of semiconductor chips and the Semiconductor Industry Association (SIA) an industry association comprised of chip

manufacturers and users. I appreciate the opportunity to appear before this committee and explain the technology which the Semiconductor Chip Protection Act of 1983 ("the act") is intended to protect from piracy.

CHIP TECHNOLOGY

The Semiconductor Chip Protection Act of 1983 gives copyright protection against pirates copying semiconductor chips (also known as integrated circuits). These chips are collections of transistors formed on single structure which work together to perform a particular electronic function. The latest generation of chips on the market today contain upward of 250,000 transistors which are compacted on a quarter inch square area of a silicon wafer. These chips have more computing power, compute faster, are more reliable, consume far less power, and sell at a fraction of the cost of the mainframe computers built in the 1970's.

The most advanced semiconductor chips can be broadly classified into two categories: microprocessors and memories. The microprocessor is often referred to as a "computer on a chip" because it has logic circuits capable of electronically performing various information processing functions. It serves as the "brains" of much of today's electronic equipment. A memory, on the other hand, is a semiconductor chip who's function is simply to remember certain data. This data could be the input to the microprocessor. That is, it could be data upon which the microprocessor will operate. It could also be the output of the microprocessor, i.e., data which the microprocessor has already operated on and which needs to be saved for future computations. Of course, the functions of a microprocessor and a memory can be integrated on the same semiconductor chip.

A typical use of a semiconductor chip could be to control the flow rate of fuel into a automobile carburetor. The semiconductor chip would be programmed to maintain a particular flow rate. A sensing device would measure the actual flow rate and provide data to the semiconductor chip which would compare the actual flow rate to the desired flow rate. The semiconductor chip would control the opening or closing of a valve to adjust the actual flow rate to make it equal to the desired flow rate. These types of semiconductor chips are used today in various electronic equipment such as automobile fuel and emission control systems, robotics, minicomputers, mainframe computers, calculators, telecommunication equipment, electronic games, medical equipment, wordprocessing equipment and computer aided design/computer aided manufacturing equipment (CAD/CAM), and of course, the personal computer.

TECHNOLOGY

The basic building block of a semiconductor chip is a transistor. A transistor is an electronic device which is capable of amplifying electrical signals and acting as an electrical switch. These transistors are then connected (integrated) to form a particular circuit which performs the electronic function desired by the chip designer. The transistor is fabricated on a material known as a semiconductor. Semiconductors can act as electrical insulators or electrical conductors depending on the electrical state of the semiconductor. Since a transistor can conduct or not conduct, and the properties of the semiconductor can be adjusted by "doping" the semiconductor with certain impurities, it is referred to as a semiconductor.

PRODUCTION OF A CHIP

Transistors and chips are formed on a thin semiconductor substrate (typically silicon) which is known as a "wafer". Typically, it is a 5" diameter disk approximately .025 inches thick. Approximately 100-200 chips will be made at one time by processing a wafer. The wafer will be subjected to certain chemical, photographic, and heat treatments. Figure la-le shows a cross section of a typical transistor. The fabrication of a simple transistor would be as follows:

(a) Grow a thin oxide over the entire surface of the wafer (see Figure la).

(b) Next a thin layer of photoresistive material ("Resist") is deposited on top of the oxide. It will now be necessary to selectively remove certain portions of this resist as well as the underlying oxide so that the silicon surface will be exposed (see Figure 1b). This is done by imprinting a pattern on the resist to develop certain areas of the film while leaving other areas undeveloped. The entire wafer is then dipped in chemical baths and the undeveloped resistor and the underlying oxide can be etched away but the developed resistor will not be etched away and the underlying oxide will be protected. It is these patterns that allow a layout designer to connect 250,000 transistors in the appropriate manner on a single chip. It is these patterns that the Semiconductor Chip Protection Act of 1983 is intended to protect.

The 3 dimensional set of patterns which appear on the actual chip are called "mask works" in the ACT. When the single patterns (or portions) are embodied in other forms which are necessary to manufacture the chip, they are called "masks" in the act.

(c) Portions of the silicon substrate are now exposed and certain impurities can be deposited onto the substrate or directly implanted into the substrate (see Figure 1c). These impurities (typically boron, phosphorus or arsenic) will change the properties of the silicon substrate.

(d) Now a layer of conducting material such as polysilicon or metal is again deposited over the entire surface of the wafer (see Figure 1d).

(e) The polysilicon is then selectively etched away similar to the manner that the oxide was etched away. We are left with a basic metal, oxide-semiconductor (MOS) transistor (see Figure 1e).

The actual production of a chip will require many additional iterations of this selective etching process to allow connection between the transistors and to the customers system.