By Gregory J. Wrenn ^*

October 15, 1989

Copyright 1989 Gregory J. Wrenn

TABLE OF CONTENTS

INTRODUCTION

I. THE LOOK AND FEEL OF COMPUTER SOFTWARE AUDIOVISUAL DISPLAYS

A. The Technology

B. The Market

II. THE LANHAM ACT OF 1946

A. Secondary Meaning

B. Likelihood of Confusion

C. Doctrine of Functionality

III. THE COPYRIGHT ACT OF 1976

A. The Idea/Expression Dichotomy

B. Refocusing the Idea/Expression Analysis

IV. THE PATENT ACT OF 1952

A. Utility Patents

1. Process Claims

2. Judicially Determined Exceptions

a. Mathematical algorithms

b. Printed matter

c. Mental steps

B. Design Patents

1. Article of Manufacture

2. Nonfunctional Ornamental Design

3. Refocusing the Functional/Nonfunctional Analysis

CONCLUSION

APPENDIX

INTRODUCTION

The judiciary should brace itself for a surge of computer software-related litigation in the coming years. The Patent Office, once hostile toward claims involving software, ¹ now commonly issues patents for software inventions. ² As a result,

This imminent litigation is apt to involve the "look and feel" ⁴ of computer software audiovisual displays, ⁵ as well as legal theories of protection in addition to patent law. The means and scope of protection available for look and feel remain uncertain, yet the "stakes of this debate are enormous." ⁶

This comment reviews the means and scope of protection for computer software look and feel available from federal statutory sources: trade dress protection under Section 43(a) of the Lanham Act of 1946, ⁷ the Copyright Act of 1976, ⁸ and the Patent Act of 1952. ⁹ In this context we see the full presence and force of competing interests in the structure of the nation's industrial policy. As a result, the analysis of one body of law helps answer questions in another.

Specifically, this comment proposes that the focus of the doctrine of functionality under Section 43(a) of the Lanham Act be adopted for analyzing the idea/expression dichotomy in copyright law, as well as the functional/nonfunctional distinction of design patent protection. Some courts applying the doctrine of functionality in trade dress cases treat the label "functional" as a legal conclusion, not a metaphysical fact. If trade dress protection for a product feature would unduly hinder competition, the feature is deemed functional and not protected. This form of analysis proves useful in the areas of copyrights and design patents. Further, it is consistent with the underlying rationale for such distinctions in each body of law: enough protection should be provided to reward innovation, but in no event should product protection become a market monopoly.

Section I of this comment describes computer software audiovisual look and feel, and examines the market for the technology. Section II explores the availability of trade dress protection for computer software look and feel under Section 43(a) of the Lanham Act. Section III discusses copyright protection and develops a proposal for refocusing the idea/expression analysis under the Copyright Act based on the doctrine of functionality under the Lanham Act. Section IV considers the validity of utility and design patents for computer software look and feel, and advances a proposal, based on the doctrine of functionality under the Lanham Act, for refocusing the functional/nonfunctional distinction of design patents. This comment is followed by an appendix which provides an illustrated guide to the technology discussed herein.

I. THE LOOK AND FEEL OF COMPUTER SOFTWARE AUDIOVISUAL DISPLAYS

Computer programs manifest themselves in any number of ways. ¹⁰ Similarly, the term "look and feel," also known as "total concept and feel," has been used by copyright law in a number of contexts. ¹¹ Applied to computer software, it has been used in reference to the look and feel of written program instructions ¹² as well as the look and feel of a program's audiovisual displays. ¹³ This comment is concerned only with the latter. This section will describe the technology at issue and consider it within its market context.

A computer program normally generates audiovisual displays as one means of communicating with the user, a part of the user interface. The user interface is the means by which the software and user interact to achieve the desired result. ¹⁴ Normally one uses a typewriter-like keyboard or a pointing device (such as a "mouse") to issue commands to the program or enter other information, including data. All or part of these commands may be visible on screen and available for selection from a "menu." Selecting a command may execute a function or produce a list or "submenu" of additional choices. Executing a command usually results in visual feedback to the user on the computer's display. (Further information may be found infra, in the appendix, which provides an illustrated guide to computer software look and feel; it describes four products that are, or have been, the subject of look and feel litigation.)

Look and feel includes individual audiovisual displays, as well as dynamic elements in the user interface. ¹⁵ Unlike a motion picture, a program's screens are usually designed to follow a functional, not chronological, order. ¹⁶ Look and feel is "the sequence of the screens and the choices presented, the layout of the screens, and the method of feedback to the user ...." ¹⁷ One manufacturer described the look and feel of its computer's user interface this way: "Just listing the elements ... doesn't do it justice ... it's the way they all work together -- the gestalt." ¹⁸

Listing the elements of the interface doesn't do it justice because copyright owners are not seeking protection for just the elements. Were that the case, "a plagiarist would escape by immaterial variations." ¹⁹ Rather, developers seek protection against those competitors who use the same or similar audiovisual displays, presented in the same or similar structure, sequence, and organization. This explains the nebulous character of the term "look and feel," and the difficulty courts have in defining the scope of protection. "[A]s soon as literal appropriation ceases to be the test, the whole matter is necessarily at large ...." ²⁰

Software has been a significant and growing element of the economy; by 1988, the worldwide market for software reached an estimated $50 billion. ²¹ Software is produced by large and small firms, and there is a trend toward mass-marketed programs. ²² In spite of somewhat uncertain intellectual property protection, the market seems to have fared well. For example, while U.S. electronics manufacturers have suffered at the hands of offshore competition, U.S. software developers continue to dominate the world market. ²³

In the market, the look and feel of software may be of tremendous importance to the developer and is often of greater commercial value than the programming code that implements it. ²⁴ While traditionally it has been an item of commercial value, microcomputer markets are moving toward "standard user interfaces" among programs, ²⁵ i.e., software developers apply the same principles and techniques of a single user interface to the various programs developed for a computer. In effect, standardized interfaces dramatically reduce the time it takes to learn a new application program. Similar functions and commands are executed similarly, so users do not have to learn a new means of interacting with every new program. The result is that, in markets with standard user interfaces, programs are expected to look and feel the same.

Given this understanding of the technology and its market, the following sections examine available federal protection.

II. THE LANHAM ACT OF 1946

The Lanham Act ²⁷ codifies the federal law of trademarks in the United States. Section 43(a) of the Lanham Act ²⁸ prohibits the false designation of a product's origin or a false description of a product's contents. ²⁹ One form of false designation of origin is trade dress misappropriation, i.e., copying a competitor's trade dress ³⁰ in order to capitalize on the reputation of the competitor.

Section 43(a) was intended to establish a uniform federal law of unfair competition. ³¹ Although there appears to be no case in which a Section 43(a) claim has been litigated in the context of computer software look and feel, one commentator predicts trade dress protection under Section 43(a) may soon be extended to such cases. ³² In light of that prediction, this section analyzes the scope of trade dress protection in the context of computer software look and feel.

A plaintiff seeking relief under a theory of trade dress misappropriation must establish that (1) the trade dress of their product has acquired a "secondary meaning" in the marketplace; ³³ (2) there exists a likelihood of confusion on the part of consumers as to the source of the product; and (3) the appropriated elements are "nonfunctional" in nature. ³⁴

Trade dress has secondary meaning if in the minds of the public, the primary significance of a product feature or term is to identify the source of the product rather than the product itself. ³⁵

There is no telling how long it might take for a producer to become identified with the trade dress, although courts have noted that the velocity with which reputations may be acquired or lost differs among industries. ³⁶

It is conceivable that consumers of computer software might associate the look and feel of certain software products with a particular developer as the source of that software. One commentator claims this is the case with the Macintosh interface, "[which is] clearly closely associated with Apple Computer." ³⁷ This claim is untenable, however. Since virtually all software developers for the Macintosh have adopted Apple's standard user interface for Macintosh software products, the look and feel of such products tells consumers nothing as to who is the source of the product. The developer might be Apple Computer, Microsoft, or any of hundreds of different software developers who market products for the Macintosh computer. Consumers may associate the Macintosh interface with Apple Computer in some manner, but they will not assume that Apple Computer is the source of every product using the interface.

Secondary meaning is more likely to attach in computer markets such as that of the IBM PC, where there has not yet developed a de facto standard for user interfaces. ³⁸ Perhaps the best example would be the distinctive interface introduced by Lotus Development Corporation for its spreadsheet product, 1-2-3. ³⁹ It was frequently referred to in the market as the "Lotus" interface, suggesting the required identification between the look and feel of the product and its developer for a finding of secondary meaning. ⁴⁰

Actual confusion, however, is not the test. ⁴² What actions a competitor must take to avoid confusion depend on the circumstances; clear labeling of the source may not be enough. ⁴³

It would seem that consumers must be exposed to the look and feel of software prior to purchase for the requisite likelihood of confusion to exist. ⁴⁴ Stated another way, no confusion is likely to occur if consumers never have the opportunity to see, and subsequently be confused by, a software product's audiovisual displays. In the case of mass-marketed software, consumers are generally less sophisticated, the products tend to be low-cost, and the marketing techniques of developers are similar. ⁴⁵ Such consumers are less likely to have seen the product demonstrated before purchase. Purchasers may buy solely on the basis of recommendations, ⁴⁶ packaging, advertising, etc., and never have the opportunity to be confused by the appearance of a program's audiovisual displays.

Nonetheless, where consumers generally are exposed to the look and feel of a product prior to purchase ⁴⁷ the potential likelihood of confusion could exist.

The courts, not the Congress, have insisted that protected elements be limited to "nonfunctional" features. ⁴⁸ The reason for this limitation is an overriding public policy of preventing market monopolization. ⁴⁹ "[A] product feature is functional if it is essential to the use or purpose of the article or if it affects the cost or quality of the article." ⁵⁰ Judicial inquiry is addressed to "whether the whole collection of elements taken together are functional." ⁵¹ Although courts refuse to protect functional features, "functional elements that are separately unprotectable can be protected together as part of a trade dress." ⁵²

Is computer software look and feel functional? As a product feature, it certainly plays a "functional" or "utilitarian" role. For almost all software programs, a user interface is essential to the use or purpose of software -- by definition, it is the means by which the user interacts with the software to achieve the desired result. ⁵³

Perhaps specific features need to be defined more precisely, e.g., are "pull-down" menus ⁵⁴ essential to the use or purpose of software? Probably not. Products have displayed menus in many different ways, and some display no menu at all. ⁵⁵ They do not affect the "cost" of the product, in the sense of reducing production costs. On the other hand, pull-down menus, as well as look and feel generally, may be said to affect the quality of the product in terms of ease of use, efficacy, or salability.

This "plain language" application of the rule shows that the line between functionality and nonfunctionality "is not brightly drawn in every case." ⁵⁶ In fact, the distinction is unworkable unless one recognizes, as Judge Rich did in In re Morton-Norwich Products, Inc., ⁵⁷ that the designation is a legal conclusion, rather than the prerequisite to one. It is necessary to distinguish between de facto functionality and de jure functionality. ⁵⁸

De facto functionality is functionality in the lay sense, i.e., directed toward a use or purpose. As noted above, a feature that is de facto functional may or may not receive protection. Similarly, a feature that at one point is nonfunctional may become functional later. ⁵⁹ De jure functionality refers to the legal conclusion: what product features may be copied (de jure functional) and what features may be protected (de jure nonfunctional).

Given that de jure functionality is a conclusion, the question remains as to what considerations precede the conclusion. A single, clear criterion emerges from the case law: whether protection against imitation will unduly hinder the competitor in competition. ⁶⁰ Courts seek to balance and reconcile the public's interest in making use of a design, the public's interest in "producer identification," ⁶¹ and the originator's interest in being the sole vendor. ⁶²

This criterion is consonant with the rationale for the doctrine of functionality: to prevent market monopolization. ⁶³ If trade dress protection is available only when such protection still allows for effective market competition, then no monopolies will result. Consequently, where competitors may fairly compete without imitating trade dress, protection should be extended to such trade dress without regard to metaphysical distinctions between de facto functional and de facto nonfunctional features.

Applying this test to the look and feel of computer software, we find that situations exist where competition might, and might not, be unduly hindered by trade dress protection. For example, in markets with standard user interfaces, consumers are likely to insist that products adhere to these standards. ⁶⁴ Products with similar capabilities following the same interface guidelines are expected to have a similar appearance and organization. Broad trade dress protection would prevent effective competition among similar products and result in excessive market power for a vendor. Opportunities for protection in such markets must necessarily be very narrow.

Where no standard user interface exists, the argument that user interface features are de jure nonfunctional is much stronger. However, the existence of alternative user interfaces is not conclusive. If an interface is particularly efficient, or if for other reasons the only means of effective competition is through imitation, ⁶⁵ then no protection should be granted.

It appears that Section 43(a) provides a viable cause of action in certain situations, particularly where the appropriation takes place in a market with no standard user interface, where consumers are exposed to the look and feel of software before it is purchased, and where the grant of protection would not unduly burden competition in the relevant market.

However, the potential importance of trade dress protection for computer software look and feel is limited by two factors. First, copyright protection is likely to remain more attractive as a means of protection to software developers than trade dress. Copyright protection is available at little or no cost, while trade dress protection is not available until sufficient time and marketing efforts have established the necessary secondary meaning. While trade dress protection lasts indefinitely, the minimum duration of copyright protection is already much longer than the expected useful life of most software products. ⁶⁶ Further, the trade dress rubric of "confusion as to source of the product" results in a narrower scope of protection than copyright. ⁶⁷

Second, a trade dress theory of protection is tenable only in a software market without a standard user interface; historically, the IBM PC market has been such a market. The present trend toward standard user interfaces, ⁶⁸ however, will necessarily be a trend away from the applicability of Section 43(a) to look and feel.

III. THE COPYRIGHT ACT OF 1976

Copyrightable subject matter is described by the Copyright Act as:

Two categories of copyrightable works applicable to computer programs include "literary works" and "audiovisual works." ⁷⁰ Literary works include the written programming instructions of software as stored on the computer. ⁷¹ Computer software video displays are considered audiovisual works. ⁷²

Audiovisual works are defined as a "series of related images which are intrinsically intended to be shown by the use of machines ...." ⁷³ As such, all aspects of computer software displays would ordinarily come within the scope of copyright protection; this includes audiovisual look and feel. ⁷⁴

There is an affirmative statutory limit on otherwise copyrightable subject matter, however, and it is discussed in the following section.

Section 102(b) of the Copyright Act provides:

This limitation is known as the "idea/expression" dichotomy, ⁷⁶ and has proven itself to be extremely difficult to administer. ⁷⁷ Judge Learned Hand's famous "levels of abstraction" test ⁷⁸ shows how a line might be drawn, but does not offer guidance on where to draw it. ⁷⁹ "Obviously, no principle can be stated as to when an imitator has gone beyond copying the `idea,' and has borrowed its `expression.' Decisions must therefore inevitably be ad hoc." ⁸⁰

The difficulty in distinguishing idea from expression should come as no surprise. How is it possible for an idea to exist, except in some expression? How is it possible to excise an idea from its expression without recasting it into another expression, for example by stating the idea more abstractly, or contemplating it in a different medium? ⁸¹

The "unprincipled" nature of these determinations forces courts to grapple with a metaphysical issue. ⁸² They are set adrift without guidance as to how to separate by law what seems inseparable in fact. Ad hoc determinations where no clear criteria exist leave attorneys unable to map for their clients reliable bounds of protection. Further, the uncertainty of protection diminishes the incentive effect for authors and hinders the promotion of useful arts and sciences.

This need not be the case.

This comment proposes a new focus for analyzing the limits of copyrightable subject matter found in Section 102(b). ⁸³ The proposal is this: courts should distinguish "de jure idea" from "de facto idea." Where the grant of copyright protection to a feature would unduly hinder competition within the market for a writing, the feature should be held a de jure idea and dedicated to the public domain. Protection would unduly hinder competition when society's interest in competitive markets outweighs society's interest in rewarding the author.

This proposal reflects a test of balancing, not metaphysics, that is required by the very rationale of distinguishing idea from expression. It is consonant with the purpose of copyright law: to create the most efficient balance between society's interest in competitive markets, and incentives ⁸⁴ for the production and dissemination of information. ⁸⁵ Further, it is fully consistent with the Copyright Clause of the Constitution. As Justice O'Connor states:

This general form of analyzing the scope of copyrightable subject matter has several advantages. First, it focuses on considerations that courts are much more experienced at resolving, making it administrable. ⁸⁷ Second, the focus in no way departs from the wording or the purpose of the Copyright Act; rather this proposal adds the necessary gloss to Section 102(b). Third, this analysis will promote the progress of useful arts and sciences by balancing the need to encourage innovation with the need to foster markets for such work. Finally, this analysis will result in greater predictability of protection. ⁸⁸

Applied to computer software look and feel, the proper analysis requires placing the subject matter in the context of its market. Where a market has adopted a standard user interface, the use of such standards in software design becomes essential to effective competition. In such a market, courts should recognize a virtual per se rule: the general look and feel of computer software audiovisual displays is a de jure idea and unprotectable. ⁸⁹

More specifically, consider two uses of "icons" in the Macintosh interface. "Disk" icons ⁹⁰ display a small graphic image selected by the developer that appears on screen and identifies the disk when it is inserted into the computer. The graphic images can be fanciful. Another form of icon, a "tool" icon, ⁹¹ exists when a program is running, and the icon represents a certain command or function that is available. For instance, a "paint" program for drawing pictures might have a column of icons that represent tools to draw boxes, circles, ovals, freehand strokes, lines, and so on. The purpose of these icons is to represent the capability without words, using only small pictures.

Protection for original disk icons would probably not hinder competition at all. Protection for original tool icons might present a different problem. There may be very few ways to represent a software command or function, or there may be many ways to represent it but only a few that effectively convey to a user the command or function represented by the icon. If protection for certain tool icons would allow a less-than-acceptable number of developers to monopolize the only effective means of expressing the capability of a certain tool, the tool icons should be a de jure idea and unprotectable. ⁹²

Where a market has no standard user interface, however, competitors can generally compete effectively without copying a competing product's look and feel. The wide competitive playing field justifies a wider scope of protection for authors in such markets. On the other hand, circumstances may exist in these markets where protection would nonetheless unduly hinder competition. The extraordinary success of a product might result in formidable barriers to entry. For example, a vast installed base of users may be unwilling to purchase similar products with different user interfaces because of retraining costs. At some point, society's interest in competition in that market will no longer tolerate the use of copyright protection as a means of perpetuating market domination.

In conclusion, copyright law should shield original works of authorship; that shield should not, however, be misapplied as a barrier to market entry.

IV. THE PATENT ACT OF 1952

The law of patents in the United States is governed by the Patent Act of 1952. ⁹³ Two different forms of patents are relevant to this issue and will be discussed separately: utility patents and design patents.

Thomas Jefferson ⁹⁴ drafted the first description of proper subject matter for utility patents, and the substantive statutory requirements have seen little change since the Patent Act of 1793. ⁹⁵ Section 101 of the Patent Act of 1952 defines patentable subject matter as "any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof ...." ⁹⁶ Committee reports accompanying the Patent Act indicate that Congress intended statutory subject matter to include "anything under the sun that is made by man." ⁹⁷

This broad scope of patentability is narrowed, however. Congress has relied on the Patent Office and the judiciary to resolve questions of subject matter patentability on a case-by-case basis, providing only the barest statutory guidance. ⁹⁸ Consequently, subject matter patentability is shown by coming within the plain language of the statute and avoiding areas excepted by the judiciary. ⁹⁹

Patent applications for computer programs are generally made in the form of "process" claims. ¹⁰⁰ The following sections will analyze such claims and the relevant judicial doctrines of exclusion.

The Patent Act defines process to mean "process, art or method, and includes a new use of a known process, machine, manufacture, composition of matter, or material." ¹⁰¹ A dictionary definition of the term includes "a particular method of doing something, generally involving a number of steps or operations." ¹⁰²

A plain reading of the statute suggests that the de facto functional, interactive aspects of a program's user interface constitute a process -- for example, the process of making a greeting card in Broderbund Software, Inc. v. Unison World. ¹⁰³ Computer "implemented processes are encompassed within 35 U.S.C. [§] 101 under the same principles as other machine implemented processes, subject to judicially determined exceptions ...." ¹⁰⁴ The question, then, is whether the judiciary will except such a process from the statutory definition. ¹⁰⁵

Three doctrines potentially relevant to computer software look and feel exclude mathematical algorithms, printed matter, and mental steps from patentable subject matter. ¹⁰⁶

This doctrine has been relevant in the few cases on patents involving software to reach the Supreme Court. The Court's last decision on the issue, Diamond v. Diehr, ¹⁰⁷ was the first Supreme Court case in which a patent including a computer program was upheld as within Section 101 patentable subject matter. In Diehr, the Court applied the test of In re Freeman. ¹⁰⁸

The first step of the Freeman test is to determine whether the claim directly or indirectly states an algorithm. ¹⁰⁹ If so, the second step is to analyze the claim to ascertain whether, in its entirety, it wholly preempts the use of that algorithm (in which case it is nonstatutory subject matter). ¹¹⁰ In the case of look and feel, we never reach the second step of the Freeman test. A user interface does not directly or indirectly state a (mathematical) algorithm. Its implementation may involve the use of mathematical formulae in programming, but it would probably not depend on any one algorithm as a means for implementing the screen displays. ¹¹¹ Such programming issues are transparent to the user and unnecessary in describing or implementing the process.

Diehr, as well as the Patent Office's Manual of Patent Examining Procedure, ¹¹² provides guidelines on the patentability of computer programs when they involve an algorithm. They are silent, however, on how to analyze claims that do not involve algorithms. The conclusion is that this doctrine, which involves the only Supreme Court pronouncements on the patentability of software-related inventions, will not exclude computer software look and feel from patentable subject matter.

In the case of computer programs, both the written source code and the audiovisual displays are arguably printed matter. Historically, patents were denied when the sole distinction over the prior art involved printed matter. ¹¹³ The rationale was that printed matter did not relate to the physical structure of the invention and therefore was not within the scope of the patent statute. ¹¹⁴

Where printed matter relates to the physical structure of the invention, however, courts have upheld the patentability of such inventions. For example, an accounting system utilizing columns covered with movable strips of paper was treated as a "structure," rather than printed matter, and held to be patentable. ¹¹⁵ Similarly, an educational mathematical device in the form of a band containing digits printed at certain intervals could not be excluded from patentable subject matter simply because the differences between the invention and prior art were to be found in the content of printed matter. ¹¹⁶

Computer software source code and audiovisual displays play active, essential, de facto functional roles in the operation of the computer and any process implemented thereby. In particular, dynamic audiovisual displays in a user interface play a structural role analogous to the movable strips of paper in In re Hansen. ¹¹⁷ Consequently, as a general rule, the look and feel of computer software should not be excluded from patentable subject matter on the grounds that the invention takes the form of printed matter. ¹¹⁸

In the 1940's the "mental steps" doctrine gained express recognition in the Patent Office, Ninth Circuit Court of Appeals, and the Court of Customs and Patent Appeals, and excluded processes composed of "purely mental steps" from patentable subject matter. ¹¹⁹ Some cases suggested that a patentable process must operate to transform and reduce matter to a different state or thing. ¹²⁰ "It is self-evident that thought is not patentable." ¹²¹

When the issue of patents involving software arose in the 1960's, the Patent Office refused to allow such applications, relying in part on the mental steps doctrine. ¹²² Software-implemented processes might fall into the doctrine's scope of exclusion since a process taking place wholly within a software program would not necessarily operate to transform and reduce matter to a different state or thing. ¹²³ One might analogize the steps executed in a computer program to a "thought process" that could be performed mentally by a human.

Look and feel-related processes, on the other hand, arguably transform audiovisual displays and reduce them to a different state; for example, the manipulation of phosphors by cathode rays to create video displays on a computer monitor could provide the necessary nexus to physical matter to make such processes patentable. ¹²⁴ The process reflected in computer software look and feel cannot fairly be likened to a situation where a human could readily achieve the same result with pen and paper. A human might mentally follow the programming steps in software source code (with great difficulty), but the only practical value of following such steps in the case of computer software look and feel is to implement a user interface on a computer.

While convincing arguments may be made that look and feel- related patents would not violate the traditional mental steps doctrine, such arguments may be unnecessary. The continued existence of the doctrine has been called into question, ¹²⁶ and its viability is discussed below.

From 1969 to 1972, the Court of Customs and Patent Appeals dismantled the mental steps doctrine while reviewing software-related applications rejected by the Patent Office. ¹²⁷ As stated in In re Musgrave: ¹²⁸

The C.C.P.A. noted that the exclusion of "purely mental steps" from patentable subject matter might lead to a correct result if construed to encompass only those steps which are incapable of being performed by a machine; the fact that machine-implemented steps could also be performed by a human mentally did not make them "purely mental steps." ¹³⁰

Surprisingly, the Supreme Court's subsequent decision in Gottschalk v. Bensonn ¹³¹ made a fleeting reference to the unpatentability of "[p]henomena of nature, ... mental processes, and abstract intellectual concepts ...." ¹³¹ Professor Chisum suggests that this "disturbingly terse" reference to the mental steps doctrine may have been inadvertent and not intended to discard developments in the C.C.P.A. ¹³² The reference is made without a discussion of, or even citation to, the lower court cases on the mental steps doctrine. "It is quite irregular for any court to accomplish so much by a short, ambiguous, dogmatic statement." ¹³³

The Supreme Court's decision in Parker v. Flook quoted the Benson language, ¹³⁴ but identified it as the "established rule that a law of nature cannot be the subject of a patent." ¹³⁵ Earlier in the Flook opinion, the Court referred to the argument that a patentable process must operate to change materials to a different state or thing; ¹³⁶ however, the Court appears to disclaim this view without actually rejecting it. "As in Benson we assume that a valid process patent may issue even if it does not meet one of these qualifications of our earlier precedents." ¹³⁷

Two years later in Diamond v. Diehr, ¹³⁸ the Supreme Court stated: "Excluded from such patent protection are laws of nature, natural phenomena, and abstract ideas." ¹³⁹ The Court cited Flook and Benson, and yet omitted reference to "mental processes." The Diehr Court then stated that Flook and Benson stood for no more than the long-established rule that "principles" were not patentable processes.

This suggests that the use of the term "mental processes" in Benson, and as quoted by Flook, was never intended to refer to the mental steps doctrine. ¹⁴¹ It would appear that, for all intents and purposes, the mental steps doctrine has not been squarely before the Supreme Court since the doctrine was repudiated by the C.C.P.A. in the late 1960's and early 1970's.

Lower court decisions, both before and after Diehr, have upheld the patentability of computer programs that do not appear to change materials to a different state or thing. ¹⁴² Thus, the current state of the law appears to be that as stated in Musgrave, i.e., a process is not made unpatentable simply because some or all of the machine-implemented steps could also be performed by a human mentally. ¹⁴³

Consequently, the mental steps doctrine should not survive to exclude look and feel-related processes from statutory subject matter. Alternatively, if the doctrine does survive, the Court should hold that look and feel-related processes have a sufficient nexus to the transformation of matter and the technological arts to avoid the exclusive effect of the mental steps doctrine.

In conclusion, it appears that computer-implemented processes involving audiovisual look and feel are within the statutory scope of patentable subject matter and are not categorically excluded by the judicial doctrines relating to mathematical algorithms, printed matter, or mental steps. These inventions, if they meet the further requirements of the Patent Act such as novelty, ¹⁴⁴ nonobviousness, ¹⁴⁵ and disclosure, ¹⁴⁶ are the proper subjects of valid patents.

The first design patent law was enacted in 1842 and defined subject matter as "any new and original design for a manufacture, whether of metal or other material or materials" or any of a list of items, such as for the printing of fabrics, design for a statue, any ornament to be placed on an article of manufacture, etc. ¹⁴⁷ In 1902 it was amended to define subject matter as "any new, original, and ornamental design for an article of manufacture" and eliminating the enumerated items. ¹⁴⁸ The courts have added a further requirement to the statutory language: designs that are primarily functional or dictated by functional considerations are not proper subject matter for design patents. ¹⁴⁹

The Patent Office granted the first design patents for computer displays on May 10, 1988 to Xerox Corporation. ¹⁵⁰ United States Design Patent No. 296,339, dated June 21, 1988, states as its claim the "ornamental design for an icon for freehand drawing softkey display or the like, as shown and described." This demonstrated that the Patent and Trademark Office believed design patentability tests, including subject matter, had been met. ¹⁵¹ One commentator proclaimed that "Xerox has discovered a new form of protection for the `look and feel' of software." ¹⁵²

This novel development warrants further analysis; the scope and requirements of design patent protection are discussed further in the following sections.

The term "article of manufacture" has been construed broadly. In the case of a design patent claim for a water fountain with rotating sprays, the court held "a manufacture is anything made `by the hands of man' from raw materials, whether literally by hand or by machinery or by art. Certainly the fountains are so made." ¹⁵³ Computer monitors are made this way as well, and consequently their ornamentation (including audiovisual displays appearing on the monitor) is likely to be included within a broad construction of the term "article of manufacture."

"Ornamental implies that the design must be the product of aesthetic skill and artistic conception." ¹⁵⁴ However, a design that is "primarily functional" or "dictated by functional considerations" is not proper subject matter for design patents. ¹⁵⁵ There are two reasons for this rule of functionality. First, where function dictates configuration, there is no ornamental creativity. ¹⁵⁶ "It was certainly not the intent of the law to grant monopoly to purely conventional design which is in itself little more than a necessary response to the purpose of the article designed." ¹⁵⁷ Further, design patent protection for functional features would in effect grant a utility patent, while circumventing Section 101 subject matter requirements. ¹⁵⁸

Computer software look and feel is often the product of aesthetic skill and artistic conception; ¹⁵⁹ in most cases "aesthetic quality is critical." ¹⁶⁰ The question is whether look and feel is "primarily functional" or "dictated by functional considerations." Here again we encounter what is a test of metaphysics. ¹⁶¹ Fortunately, the rationale for the distinction here appears to be the same as that in the case of trade dress and copyright law: the scope of legal protection must be balanced against society's interest in competitive markets. ¹⁶² Further, design patent law may not be used as a means of attaining what amounts to a utility patent. ¹⁶³

This comment proposes that courts applying design patent law adopt a concept of de jure functionality. ¹⁶⁴ The ornamental design of an article of manufacture is de jure functional if protection would unduly hinder competition, without regard to the de facto functional or de facto nonfunctional nature of the design. Additionally, the subject matter of utility patents is de jure functional, and not the proper subject matter of design patents. However, if protection for an ornamental design is not an attempted "end run" around utility patent protection, and further would not result in the undue hindrance of competition, then such protection should be upheld as valid.

Whether the scope of protection in Xerox's design patents extends to the look and feel of the audiovisual displays remains uncertain. In fact, it was the Xerox Star interface that was the inspiration for the Macintosh interface. ¹⁶⁵ The recently patented screens by Xerox may bear a familial relationship to the Macintosh interface and virtually all graphics-based user interfaces available. If that familial similarity suggests that design patents could be used by Xerox to secure judgments or injunctions against virtually all major software developers (and thereby bestow market monopoly power on Xerox), then the patents should be adjudicated invalid. On the other hand, if granting protection to Xerox would protect their interest without unduly burdening competition, then the ornamental features of the patented displays should be de jure nonfunctional, and validly protected. ¹⁶⁶

Further, if Xerox's audiovisual displays embody a look and feel that is within the scope of utility patent subject matter, ¹⁶⁷ and protection for that aspect of the user interface is not disclaimed in the design patent they received, then the patent should be adjudicated invalid with respect to those aspects.

CONCLUSION

The myriad ways in which computer programs manifest themselves make them unamenable to per se rules of protection or exclusion from federal sources of intellectual property protection. Consequently, courts are forced to distinguish between various manifestations in a program, often applying rules that on their face require a metaphysical separation of what otherwise seems inseparable. This approach frequently leaves courts, counsel, and clients without reasoned principles to guide their actions.

A better approach is to abandon metaphysics and analyze each case in light of the rationale for the rule. If, in the case of trade dress, the rationale of limiting protection to nonfunctional features is to protect our interest in producer identification without unduly hindering competition, then we may safely protect trade dress as long as competition has not been unduly hindered.

The rationale for distinguishing idea from copyrightable expression is to promote writings by rewarding authors, without unduly hindering competition for such works. Consequently, we may safely reward the author by protecting an aspect of her work as long as competition in the market for such works is not unduly hindered.

Finally, the rationale applies to design patents, with further deference to the scope of utility patent protection, and the same result should follow. We may safely protect the ornamental design of an article of manufacture as long as protection would not unduly hinder competition and would not result in protection for the subject matter of utility patents without meeting the requisite tests.

This approach is more administrable for the courts, focusing on more familiar tenets of competitive analysis rather than the metaphysics of new technologies. This approach is wholly consistent with the purpose of the statutory and judicial rules; it cuts to the very rationale of the rules and provides the necessary gloss to statutory language. Further, the results are consistent with the constitutional grant of authority under which the federal statutes are enacted.