The federal intellectual property regime encourages innovation and creativity in a critical subject matter of the information age, computer software.[1] The computer user interface, in particular, represents an innovative and creative subject matter which should similarly be granted strong protection under federal intellectual property law.[2] The "user interface" is generally "the means by which the software and user interact.[3] For example, a student walks into a computer lab and observes an illuminated computer screen. She double-clicks on the "START" icon and then observes a strip of words across the top of the screen, labeled folders, and a recycle bin. The student immediately recognizes this as the Microsoft Windows 95 user interface. With the use of the mouse she manipulates the menu command hierarchy to execute the "Help" function, which provides informational text explaining the functions of the menu commands. She then double-clicks on the folder marked with the identifiable Netscape logo. She uses the Netscape software to download video and audio of a recent sporting event from ESPN's Web site.

The sound, graphics, pull-down menus, help text, and commands, as well as the mouse and keyboard, all constitute elements of a computer user interface. Some of these elements, such as the mouse, keyboard, and commands, are purely functional. Other elements, for example, the entertaining sounds and graphics and the informational help text, are purely informational or even aesthetic. Additionally, some of the elements, including logos or the overall look and feel of a user interface, can provide expression that consumers identify with a particular supplier of goods or services. This potential for functional, aesthetic, and consumer-identifiable expression demonstrates the hybrid nature of user interfaces.

This unique hybrid of functionally expressive subject matter can provide a single aspect (the user interface) of a single work (the computer software), which can qualify for all three forms of federal intellectual property protection: patent, copyright, and trademark.[4] Because of the unprecedented potential overlap of the three domains of federal intellectual property protection in a particular aspect of a work, judicial attempts to protect user interfaces reveal a tension in federal intellectual property law.

Recent developments, however, suggest the emergence of a coherent model for protecting the intellectual property of the user interface. In In re Alappat,[5] the United States Court of Appeals for the Federal Circuit suggested that the new functionality resulting from programming a general purpose computer creates a new machine and thus constitutes patentable subject matter.[6] In Lotus v. Borland,[7] the First Circuit Court of Appeals suggested that functional elements of a user interface constitute non-copyrightable subject matter.[8] These decisions suggest a paradigm shift in the federal intellectual property scheme for user interfaces. While the scope of patent protection for functional aspects of user interfaces is expanding, the scope of copyright protection for functional aspects of user interfaces is contracting.

This Comment argues that the observed paradigm shift is consistent with a "utilitarian" limitation, which excludes elements of user interfaces constituting statutory subject matter of patent law from copyright, trademark, and trade dress protection.[9] This exclusionary doctrine recognizes that functional (i.e., utilitarian) aspects of the user interface are the exclusive domain of patent law.[10] In other words, a user interface's functional method of operating a computer should not be protected by copyright, trademark, or trade dress, because it now appears, after Alappat, that such functionality constitutes patentable subject matter and "should be protected, if at all, by patent law."[11] Thus, the utilitarian limitation emanates from patent law and requires that the stricter requirements of patent law be met for protection of functional aspects of user interfaces.[12]

Part II of this Comment argues that the scope of patentable subject matter has recently expanded to include a user interface's method of operating a computer and potentially encompasses functionally expressive elements of a user interface. Part III demonstrates that a failure to apply the utilitarian limitation extends expansive copyright protection to user interface embodied functionality that now constitutes patentable subject matter. Part IV illustrates that the utilitarian limitation must be applied as well to exclude this newly patentable subject matter from expansive user interface trademark and trade dress protection.

Part V contends that a coherent approach to user interface intellectual property protection requires the application of the utilitarian limitation. This Part suggests that the utilitarian limitation strikes the proper balance between the need for strong intellectual property protection for user interfaces and the need for access to functional aspects of user interfaces that are essential building blocks for innovation in the information age. This Comment concludes that the utilitarian limitation requires that the scope of copyright, trademark, and trade dress protection for user interfaces shrink to appropriately reflect the recent expansion of the scope of patentable subject matter for user interfaces.

II.   PATENT PROTECTION FOR USER INTERFACES

Recent developments in patent law have significantly expanded the scope of patent protection for the functionality of computer software. This trend suggests that patent law will be the exclusive domain of federal intellectual property protection for the functionality of user interfaces. A brief introduction to patent law is necessary to understand the impact recent developments will have on the scope of patent protection for user interfaces.

A.   Patent Law

The United States Constitution gave Congress the power to promote "Science" and the "useful Arts" by granting "Inventors" rights in their inventions Afor limited Times."[13] Under this authorization, Congress passed the Patent Act, permitting the federal government to issue a utility patent[14] (also called a functional patent) to an inventor.[15] A patent, which provides the strongest form of federal intellectual property protection, grants the patentee the right

to exclude others from making, using, selling, or offering to sell the claimed invention for a period of twenty years, measured from the date of filing.[16]

According to the Supreme Court, the statutory subject matter of patent law includes "anything under the sun that is made by man."[17] The patent statute, however, imposes important restrictions upon the scope of statutory subject matter. An inventor can obtain a utility patent for only the following statutory classes of new and useful inventions:[18] processes[19] (such as a computer-implemented method), machines (such as a programmed computer), articles of manufacture (such as a computer program[20] stored on a computer-readable medium), and compositions of matter (such as chemical compounds).[21]

Not all statutory subject matter, however, is patentable. To obtain patent protection, the inventor must proceed through an examination process before the Patent and Trademark Office (PTO). The examination process is used to determine whether the claimed invention satisfies the patent statute's high standard of "invention."[22] Sections 102 and 103 of the Patent Act require that statutory subject matter be "novel" and "non-obvious," respectively.[23] For example, although functional aspects of user interfaces may constitute statutory subject matter, not all would satisfy the requirements of novelty and non-obviousness.[24]

Despite the high threshold requirements for invention under patent law, several trends demonstrate the advantages of patent protection for user interfaces. Since 1982 the Federal Circuit, which has exclusive appellate jurisdiction over patent rulings,[25] has been developing uniform and favorable law for patent validity.[26] Also, the PTO is decreasing patent application processing time and improving the prior art collection for software inventions.[27] These favorable trends, combined with the fact that patent protection provides the strongest form of federal intellectual property protection, suggest that companies should aggressively pursue patent protection for innovative user interfaces.[28]

B.   Software Patent Case Law

No cases directly address the validity of patent protection for user interfaces.[29] Courts, however, have addressed the validity of software patent protection for three categories of software inventions: software implementing mathematical algorithms,[30] operating system programs,[31] and application programs.[32]

Software implementing mathematical algorithms has presented the most difficult case for software patent protection because mathematical algorithms constitute non-patentable abstract ideas or natural laws. The Supreme Court's software patent decisions have primarily wrestled with the issue of whether computer-related processes that involve mathematical algorithms recite statutory subject matter.

In 1972, the Supreme Court addressed the patentability of claims that involved a computer program implementing a mathematical algorithm in Gottschalk v. Benson.[33] Benson involved a method of converting binary-coded decimal numerals directly into binary numerals for use with a general purpose computer. The Court held that the patent recited nonstatutory subject matter because Athe claims were not limited to any particular art or technology, to any particular apparatus or machinery, or to any particular end use. They purported to cover any use of the claimed method in a general-purpose digital computer of any type."[34]

In 1978, in Parker v. Flook,[35] the Court similarly held that claims which involved a program implementing an abstract mathematical formula were not statutory subject matter. The Court specifically held that conventional or post-solution activity is not sufficient to transform an unpatentable principle into a patentable process.[36] While neither Benson nor Flook suggested that software inventions could not recite statutory subject matter,[37] these two decisions left the patentability of software inventions uncertain.

In Diamond v. Diehr,[38] the Supreme Court, in its third and most recent software patent decision, clarified that a computer-related process can constitute statutory subject matter. The Supreme Court's decision in Diehr resolved much of the confusion caused by Benson and Flook. In Diehr, the Court reexamined the patentability of a program implementing a mathematical algorithm. The Court held that the claimed process for curing rubber constituted statutory subject matter, despite the claim's recitation of a step involving a computer that was programmed with a mathematical formula.[39]

In reaching this conclusion the Court distinguished Benson and Flook. The Court reasoned that in those cases the patentee merely sought to patent abstract mathematical formulas, whereas in Diehr, the patentee did "not seek to patent a mathematical formula [but rather] a process of curing synthetic rubber."[40] In addition to utilizing a programmed computer to calculate Arrhenius' equation, Diehr's claimed process "involves the transformation of . . . raw, uncured synthetic rubber, into a different state or thing."[41]

The Diehr Court concluded that a "claim drawn to subject matter otherwise statutory does not become non-statutory simply because it uses a mathematical formula, computer program, or digital computer."[42] Therefore, the Diehr Court suggested that computer-related processes, when properly disclosed and claimed, can constitute statutory subject matter under 35 U.S.C. § 101.[43]

Post-Diehr cases have held that claims drawn to operating system programs may recite statutory subject matter. For example, In re Pardo,[44] held that a compiler program for converting source code programs into object code programs constituted statutory subject matter. The court reasoned that the program provides "a method for controlling the internal operations of a computer to govern the manner in which programs are executed."[45]

Post-Diehr cases have also held that claims involving application programs were statutory subject matter.[46] For example, in Paine, Webber, Jackson & Curtis, Inc. v. Merrill Lynch, Inc.,[47] the United States District Court for the District of Delaware held that claims drafted to a "system" for cash management constituted statutory subject matter. The court stated that the method would be nonstatutory if carried out by hand, but held that the apparatus claims required machine implementation and, as such, did not preempt the general method.[48] The court reasoned that "the focus should be on the operation of the program on the computer."[49] The court held that the claims constituted subject matter because they recite a new method of operating a computer to effectuate a business activity.[50]

In holding that operating system and application programs constitute statutory subject matter, courts have focused on the functionality of the software as a process or method. Although there are no cases that appear to address directly the validity of user interface patents, software patent case law suggests that functional aspects of operating system and application user interfaces could similarly provide a statutory method of "operat[ing] the program on the computer,"[51] or a statutory "method for controlling" a computer.[52]

C.   In re Alappat: Software as Patentable Machine

In re Alappat[53] represents the most important post-Diehr case to address the patentability of computer-related subject matter. The patent in Alappat disclosed a method of varying the intensity of the pixels of an oscilloscope display in order to create a smoother wave depiction. The Federal Circuit en banc held that the claimed apparatus[54] considered as a whole constituted statutory subject matter.[55] Six of the eight Federal Circuit judges who expressed an opinion on the merits voted in favor of holding the claims statutory.[56]

The Federal Circuit reasoned that a claim is not nonstatutory simply because it reads on a general purpose digital computer as a means to perform various steps under program control. On the contrary, the Federal Circuit suggested that "such programming creates a new machine, because a general purpose computer in effect becomes a special purpose computer once it is programmed to perform particular functions pursuant to instructions from program software."[57] In particular, the Federal Circuit concluded that although Alappat's claimed apparatus recited disclosed electronic circuitry elements that perform mathematical calculations, the claimed invention as a whole was directed to a combination of elements that constituted a machine for creating an anti-aliased waveform from discrete waveform data input.[58]

Therefore, Alappat suggests that functional aspects of software claimed in connection with a computer can recite a statutory machine. The recent Federal Circuit pronouncement in Alappat confirms a significant expansion of patentable subject matter for computer software.[59] Alappat refocuses the statutory subject matter inquiry upon whether the claimed subject matter falls within one of the four categories of statutory subject matter under the plain language of 35 U.S.C. § 101, and reins in the application of the judicially developed mathematical algorithm exception to patent protection for computer-related inventions.[60]

1.   Examination Guidelines

Primarily in response to Alappat, the PTO recently released new Examination Guidelines for Computer-Related Inventions (Examination Guidelines).[61] The Examination Guidelines set forth the official PTO policy with respect to the patentability of software inventions.[62] In particular, these guidelines are meant to direct PTO examiners during the examination of software inventions. Most importantly, the Examination Guidelines suggest a significant expansion of the patentability of software user interfaces.

The Examination Guidelines adopt a two-step protocol first articulated by, or at least emanating from, the Alappat majority.[63] The first step requires a comparison of the claimed invention with a list of nonstatutory subject matter taken from court holdings.[64] If the claimed invention falls squarely within any of these categories then the Guidelines mandate the conclusion that the claims recite nonstatutory subject matter. Otherwise, the Guidelines instruct that the examination proceed to a comparison of the claimed subject matter to categories of statutory subject matter derived from Supreme Court and Federal Circuit holdings.[65]

The second step requires a determination of whether the claimed subject matter properly falls within a statutory category: process, machine, article of manufacture, or composition of matter.[66] For example, the Guidelines direct examiners to rely on the following presumption, which clearly emanates from Alappat: a computer or other programmable apparatus whose actions are directed by a computer program or other form of "software" is a statutory machine.[67] Accordingly, a user interface may be claimed as a statutory machine if the claim recites a computer whose actions are directed by software that implements the user interface.

Software can also be claimed as a method or process. The Examination Guidelines teach that if a claimed process is limited to a practical application of the abstract idea or mathematical algorithm in the "technological arts," then the claim recites a statutory process.[68] Accordingly, a user interface may be claimed as a statutory process if the claim language sufficiently limits the process to a method of operating a computer. This, of course, is a practical application within the technological arts.

The new PTO policy does not, however, extend statutory subject matter so far as to include claims for computer programs per se, data structures per se, or compilations of nonfunctional information.[69] These items are not statutory subject matter because "they are neither physical `things nor statutory processes,' as they are not acts being performed [i.e., they are not functional aspects of the software]."[70]

Accordingly, the Examination Guidelines require a demarcation between functional and nonfunctional elements embodied in computer software.[71] The attempt to draw such a line is consistent with a policy that aims to ensure that protection of user interface functionality is the exclusive domain of patent law and that nonfunctional information is protected only by the other domains of the federal intellectual property regime.

In conclusion, the Examination Guidelines suggest that a user interface claimed as a programmed computer and/or a method of operating a computer may constitute a statutory machine. The recent trend of Federal Circuit and PTO liberalization in software patent policy suggests that the scope of allowable and valid claims drawn to various functional aspects of user interfaces can be expected to continue to expand.[72]

2.   Survey of Issued User Interface Patents

Hundreds of patents have actually been issued[73] for user interface functionality. User interface patents typically claim particular user interface functionality as a method of operating a programmed computer (i.e., a system), or user interface software stored on a computer-readable medium. For example, Patent Number 5,230,063[74] ("Hoeber et al.") discloses an object-oriented graphical user interface. Hoeber et al. includes both computer-implemented method claims and machine claims (i.e., a programmed computer).[75]

Generally, graphical user interfaces (GUIs),[76] such as those of word processing systems, provide common types of user interface methods which have recently been the subject of issued patents.[77] For example, a "method for viewing icon contents on a video display" has been patented.[78] The patent claims a method that allows a user to add or remove virtual sheets of information to and from the labeled virtual folder as well as open the folder in order to read its contents (the virtual sheets of information).[79]

Similarly, the PTO issued a patent for a type of help screen for a GUI.[80] The patent claims a system and method for providing "hints" for screen objects of interest. This user interface system and method includes a window which is continually updated with "hints" as the screen cursor moves from one screen object to another, which is generally common functionality among GUIs of currently available commercial software.

Menu-driven user interfaces of data entry systems have also been the subject matter of recently issued patents. For example, IBM obtained a patent for a method of operating a data entry program through the use of a menu-driven GUI.[81] Many other variations upon menu hierarchical user interfaces have also been patented.[82]

An interesting commonality among the GUI patents recently issued is that many of these patents explicitly refer to their respective user interfaces as providing a particular functional "look and feel."[83] For example, IBM patented a multimedia user interface system.[84] The patent discloses a system for controlling a plurality of application programs and devices via "consistent interfaces having a common look and feel."[85] Similarly, many user interface patents also refer to a particular functional "selection and arrangement" of elements in the user interface's embodied method of operation.[86]

The PTO has also issued patents for touch screens, which represent a more innovative type of user interface.[87] For example, a touch screen system that acts as an electronic mouse simulator for a computer has been patented.[88] A patented method and apparatus for accessing touch screen desktop objects via fingerprint recognition illustrates an even more advanced touch screen user interface.[89]

Handwriting recognition systems allow a user to operate a computer via the user's handwriting (either in script or print). Variations upon this type of intuitive user interface have been patented.[90]

Perhaps the most innovative type of commercially available user interface to date, however, is the voice recognition system.[91] IBM, for example, sells voice command systems that enable the user to say the commands to the computer instead of manually interacting with the machine through various types of "command functions."[92] IBM and others have obtained patents for such innovative voice-recognition user interfaces and other variations on voice-related user interfaces.[93]

Functional aspects of user interfaces, described as methods and systems of operating a computer and programmed computers, clearly have provided statutory subject matter for numerous issued patents at least as far as the PTO is concerned. The recent Federal Circuit decision in Alappat suggests that user interface functionality can recite statutory subject matter as a new machine. Moreover, the Examination Guidelines suggest that user interfaces may be claimed as statutory methods if sufficiently limited by the claim language to a practical application within the technological arts. Further, the PTO also views software stored on a computer-readable medium as a statutory article of manufacture. Most importantly, Alappat, combined with the PTO's continued liberalization of its policy toward software patents, will most likely expand the availability and scope of user interface patent protection.

As the Federal Circuit and PTO continue to expand the scope of user interface patent protection, it becomes critical that courts apply the utilitarian limitation to properly exclude such newly patentable subject matter from copyright, trademark, and trade dress protection. However, the next Part of this Comment demonstrates that courts have too rarely applied the utilitarian limitation and, in effect, have granted copyright protection to aspects of user interfaces which, at least now after Alappat and the Examination Guidelines, constitute allowable and valid statutory subject matter under patent law.

III.   COPYRIGHT PROTECTION FOR USER INTERFACES

Intuitive user interfaces frequently contain text, graphics, and compilations of information. These aspects of a user interface are easily copied by competitors. Copying by competitors can remove much of the competitive advantage gained by the costly development of intuitive and creative user interfaces.[94] Because copyright prohibits the unauthorized copying of original expression and until recently the validity of user interface patent protection was uncertain, it is not surprising that for the past two decades, developers of user interfaces have attempted to protect their intuitive and creative user interfaces primarily through the vehicle of copyright law.[95] A brief review of copyright law and the initial decision to extend copyright protection to the original expression of computer software is necessary to understand the current approach to copyright protection for user interfaces.

A.   Copyright Law

Copyright seeks to promote literary and artistic creativity by protecting "writings" of "authors: for limited times.[96] The Copyright Act was passed under the same clause of the United States Constitution as the Patent Act.[97] Copyrightable works include "literary works," which are "works, other than audiovisual works, expressed in words, numbers, or other verbal or numerical symbols or indicia" that are fixed in a "tangible medium of expression."[98]

Copyrightable literary works must meet the threshold requirement of "original[ity]."[99] Originality generally requires that the work cannot be copied from another source. The Supreme Court held that originality is a constitutional standard of copyright, requiring that a work not be so "obvious" or "garden-variety" that it lacks a sufficient "modicum of creativity."[100]

Computer programs are protected under copyright law as original works of authorship. The Copyright Office began issuing registrations for claims of copyright in computer programs in 1964. After Congress passed the Copyright Act of 1976, computer programs were explicitly protected, as they fell squarely within the copyrightable category of "literary works." Copyright protection, which grants less protection than patent law, can be used to prevent unauthorized reproduction of the copyrightable expression of a computer program for the life of its author.[101]

Despite the apparent copyrightability of computer programs under the Copyright Act of 1976, Congress sought input regarding the issue of copyright protection for computer programs and created the National Commission on New Technological Uses of Copyrighted Works (CONTU) to study the issue. In its final report, CONTU observed that "[t]he cost of developing computer programs is far greater than the cost of their duplication."[102] To counteract this imbalance CONTU determined there was a need to provide copyright protection for the expression embodied in computer programs.

In 1980, Congress adopted the recommended amendments of the Final Report of the National Commission on New Technological Uses of Copyrighted Works (1978) ("CONTU Final Report") almost verbatim.[103] Congress thereby confirmed that computer programs are protected under the Copyright Act of 1976 as a type of "literary work."

Although computer programs themselves clearly constitute copyrightable subject matter, the scope of copyright protection for user interfaces is not clear from the legislative history. CONTU provided only "that [the] line (between copyrightable elements and excluded elements under § 102(b)) should be drawn on a case-by-case basis by the institution designed to make fine distinctions—the federal judiciary."[104] Accordingly, the courts have struggled with the difficult task of applying copyright protection to the hybrid subject matter of user interfaces.

B.   User Interface Copyright Case Law

The courts have not been comfortable applying copyright law to computer software in general.[105] "Applying copyright law to computer programs is like assembling a jigsaw puzzle whose pieces do not quite fit."[106] In particular, courts have struggled in defining the proper scope of copyright protection for user interfaces.

Courts have recognized two primary limitations upon the scope of copyrightable subject matter for user interfaces. First, one may use any "idea" expressed in the user interface.[107] This limitation represents the idea/expression dichotomy.[108] Under the idea/expression dichotomy, copyright protection extends only to the particular expression of ideas, not to the abstract ideas themselves.[109] In applying copyright to user interfaces, courts have primarily focused upon the idea/expression dichotomy and have developed a variety of tests to apply that limitation.[110]

Courts have been less consistent in recognizing and applying the second limitation—the utilitarian limitation for user interface copyright. The utilitarian limitation assures that another person can use any "method of operation," "procedure," "process," or "system" embodied in the user interface.[111] The following survey of user interface copyright case law demonstrates that courts have too rarely applied the utilitarian limitation.

Copyright protection has been sought for three primary aspects of user interfaces: (1) individual elements (such as innovative icons, graphic images, or help text); (2) compilations of information (such as the selection and arrangement of commands in a menu command hierarchy); and (3) the overall "look and feel" of the user interface.[112] This section examines the success in obtaining copyright protection for each of these aspects of user interfaces.

1.   Individual Elements

Copyright has been sought for individual elements of user interfaces such as the icons of GUIs. Iconic images of GUIs present difficulties for copyright because of their functional or virtual nature. In Apple Computer, Inc. v. Microsoft Corp.,[113] for example, Apple was unsuccessful in its attempt to protect iconic images such as the virtual files, virtual pages, and virtual trash cans of its GUI, because these individual elements were held to be either too functional or lacking sufficient originality for copyright protection.[114]

The district court in Apple analogized functional elements of the user interface to the buttons, knobs, and switches of a television, VCR, clock, or oven.[115] The court concluded that the functionality of an individual user interface element may deprive it of protection under copyright. Because of the functionality of Apple's individual elements, the court denied copyright protection for particular methods of presenting different views of objects or dialog boxes when selecting menu items.[116] Thus, the Apple decision properly applied a utilitarian limitation to exclude functionally expressive individual elements from user interface copyright.[117]

On the other hand, nonfunctional individual elements of the user interface, such as detailed nonfunctional graphic images or detailed help text would not present much difficulty for copyright protection.[118] These aspects of the user interface are most closely analogous to nonfunctional literary, audiovisual, and graphic works as defined in Title 17. Most importantly, nonfunctional expressive individual elements of a user interface do not constitute patentable subject matter and, therefore, would not be excluded under the utilitarian limitation. Unfortunately, the numerous cases addressing the compilation argument for user interface copyright have been inconsistent in applying the utilitarian limitation.

2.   Selection and Arrangement

Synercom Technology, Inc., v. University Computing Co.,[119] which appears to represent the first user interface copyright case, addressed the selection and arrangement argument for user interface copyright.[120] Synercom alleged that defendants infringed the copyright of the input formats of their program's preprocessing interface. The court analogized the sequence and ordering of data of Synercom Technology's input formats to the "familiar `figure-H' pattern of an automobile stick [shift]."[121] The court reasoned that the "figure-H" pattern presumably could be protected through copyright, but that copyright protects the particular expression of the pattern, and does not prohibit another manufacturer from marketing an automobile using the same pattern.[122] The Synercom court then held that the selection and arrangement of the input formats did not constitute copyrightable expression.[123]

Although the court did not explicitly apply the utilitarian limitation but rather applied the idea/expression dichotomy, the Synercom holding is consistent with the utilitarian limitation. Even if the input formats conveyed information to the user through their selection and arrangement, these formats primarily provided users with a functional method for inputting data.[124] The utilitarian limitation requires that such methods of operation be excluded from user interface copyright protection.[125] Subsequent decisions have generally not followed Synercom, however, as courts have held that Synercom-like aspects of user interfaces constitute copyrightable subject matter.[126]

Fifteen years after Synercom, for example, the Fifth Circuit addressed nearly the same issue but reached the opposite result. In Engineering Dynamics, Inc., v. Structural Software, Inc.,[127] Engineering Dynamics (EDI) alleged copyright infringement of the selection and arrangement of its input formats, some of which, ironically, EDI had itself copied from Synercom's input formats. The Fifth Circuit reasoned that the expressive purpose of the input formats was not outweighed by their utilitarian function of acting like switches in the electrical circuits of the program.[128] The Fifth Circuit concluded that the EDI input formats constituted copyrightable expression because, although the degree of interaction might not be as high as that of the Lotus 1-2-3 menu command hierarchy,[129] EDI proved original expressive elements in the selection and arrangement of its input formats.[130]

The Fifth Circuit's analysis is unpersuasive because the court views the utilitarian limitation as a balancing test. This is a fundamental misunderstanding of the utilitarian limitation. Copyright simply does not extend to methods of operation.[131] Correctly applied, the utilitarian limitation would dictate that the EDI court, like the Synercom court, hold that the input formats constituted a non-copyrightable method of operation.[132]

The scope of user interface copyright reached its maximum breadth in Lotus Development Corp. v. Paperback Software International.[133] In Paperback, Judge Keeton addressed the scope of copyright protection for the selection and arrangement of the menu command hierarchy of the popular Lotus 1-2-3 spreadsheet application.[134] Judge Keeton held that the command menu hierarchy of Lotus 1-2-3 constituted copyrightable subject matter.[135] However, if the court had properly applied the utilitarian limitation, it would have concluded that the menu command hierarchy of Lotus 1-2-3 constituted a non-copyrightable method of operation.[136]

These cases demonstrate that plaintiffs have met with fair success when arguing for selection and arrangement copyright of user interfaces. However, many of these decisions reveal that courts have failed to properly apply the utilitarian limitation and have effectively granted copyright protection to patentable subject matter, i.e., statutory methods of operating a computer. The "look and feel" cases demonstrate that courts have been more consistent in applying the utilitarian limitation when asked to grant copyright protection to the overall "look and feel" of a user interface.

3.   "Look and Feel"

The "look and feel"[137] argument for user interface copyright attempts to gain the most expansive scope for user interface copyright. This argument basically asks courts to extend copyright to the whole user interface, as opposed to protecting individual icons, a selection and arrangement of elements of a user interface.

The "look and feel" argument for user interface copyright essentially seeks to monopolize a particular user interface's functional behavior and any particular embodied methods of operating a computer.[138] Accordingly, the argument for "look and feel" copyright protection of the user interface has appropriately not succeeded to date;[139] this is primarily due to application of the utilitarian limitation.

In Apple Computer, Inc. v. Microsoft Corp.,[140] the Ninth Circuit addressed this "look and feel" argument. Apple alleged that Hewlett Packard's (HP) NewWave program and versions of Microsoft's Windows program infringed Apple's copyright[141] in the GUI for the Apple Lisa and Macintosh computers.[142] Both of Apple's GUIs were developed to provide an intuitive user-friendly environment for users to operate Apple computers. Many elements of the Apple environment are based on a desktop metaphor. For example, Apple's GUIs provide iconic representations of familiar objects from the office environment, including a filing cabinet, file folders, and a trash can for discarding (deleting) documents.

After unsuccessfully arguing for copyright protection for the individual icons of its GUI at the district court level,[143] Apple argued, on appeal to the Ninth Circuit, that the particular combination of graphic icons and overlapping windows of defendants' user interfaces were too "Mac-like" in their total concept and feel. Apple contended that Microsoft and HP simply "mimicked the composition, organization, arrangement and dynamics of the Macintosh interface, as shown by striking similarities in the animation of overlapping windows and the design, layout and animation of icons."[144]

The Ninth Circuit affirmed the district court's "dissection" of the unprotectable elements of Apple's GUIs and refused to grant copyright protection to the overall "total concept and feel" of the elegant Macintosh user interface.[145] In effect, the Ninth Circuit applied the utilitarian limitation: "Apple cannot get patent-like protection for the idea of a graphical user interface, or the idea of a desktop metaphor which concededly came from Xerox."[146]

The Ninth Circuit remanded the case so that the district court could revisit both Microsoft's and HP's requests for attorney's fees.[147] The ultimate result of this litigation provides an ironic conclusion: Apple was required to defend against having to pay attorney's fees for its litigation efforts to utilize copyright law to protect what was arguably the most intuitive graphical user interface for home computers on the market to date.[148]

The user interface copyright case law reveals that several courts have extended copyright to methods of operation embodied in user interfaces, thereby granting patent-like protection through the vehicle of copyright.[149] Although a few cases such as Apple have excluded certain user interface functionality from copyright, the First Circuit's recent decision in Lotus v. Borland illustrates the dramatic effect of the utilitarian limitation upon user interface copyright, and, in particular, the selection and arrangement argument for user interface copyright.

C.   Lotus v. Borland: Methods of Operating a Computer as Non-Copyrightable Subject Matter

Lotus Development Corp. v. Borland International, Inc.[150] presents the most significant user interface copyright decision to date. It also represents the first software copyright case ever heard by the United States Supreme Court. Most importantly, the Borland decision demonstrates the proper effect of the utilitarian limitation on copyright protection for user interfaces.

The case centered around the well-known Lotus 1-2-3 software.[151] The Lotus 1-2-3 menu command hierarchy consists of 469 commands arranged into more than fifty menus and sub-menus. Lotus 1-2-3 utilizes a two-line menu. The top line of the two-line menu consists of the 1-2-3 command, which a user may select to execute a particular function of the 1-2-3 program. A user executes commands, such as "Copy" and "Print," by highlighting the command in the menu hierarchy or by typing the first letter of the menu command.

The bottom line of the two-line menu contains the 1-2-3 "long prompts." The long prompts represent nonfunctional information. They provide a concise help text explaining what the highlighted menu command above it will do if entered by the user, or informing the user of the options available at the next level of nested menus. A user can execute a 1-2-3 command but cannot execute a long prompt. For example, the long prompt for the "Quit" command provides the following explanation to the user, "End 1-2-3 session (Have you saved your work?)."

Lotus 1-2-3 also provides macro capabilities.[152] A user cannot write a Lotus 1-2-3 macro unless the user has access to the commands and knows the order of those commands as they appear in the 1-2-3 menu command hierarchy. Nor can a user modify, enhance, or debug such a macro without access to the 1-2-3 menu command hierarchy. Moreover, macros requiring input from the user to complete their execution, called interactive macros, cannot be executed without the user's having visual access to the 1-2-3 menu command hierarchy.[153]

By its own admission, Borland included "a virtually identical copy of the entire 1-2-3 menu tree" in its Quattro and Quattro Pro spreadsheet programs.[154] Borland provided a copy of the 1-2-3 menu command hierarchy, called the "Lotus Emulation Interface," as an alternate user interface for Quattro. The Lotus Emulation Interface had a different-looking screen so that its overall appearance was not identical to that of the Lotus 1-2-3 user interface.[155] The emulator also included additional Borland commands not present in the 1-2-3 menu command hierarchy. The Lotus Emulation Interface enabled Borland to achieve user interface compatibility with Lotus 1-2-3.[156] Users could elect to operate the Quattro programs by using Borland's independently created native mode menu command hierarchy[157] or the Lotus Emulation Interface.

Lotus sued Borland, contending that Borland had copied the 1-2-3 menu command hierarchy, except for the long prompts.[158] On July 31, 1992, the district court, using an analysis similar to the Paperback court's, concluded that the Lotus menu command hierarchy constituted copyrightable expression and held that Borland had infringed Lotus's 1-2-3 copyright as a matter of law.[159] The district court also found that Borland's Key Reader[160] infringed the copyright of the Lotus 1-2-3 program, as it contained "a virtually identical copy of the Lotus menu tree structure, but represented in a different form and with first letters of menu command names in place of the full menu command names."[161] Therefore, the district court also extended copyright to the 1-2-3 macro language.

1.   The Utilitarian Limitation Applied to User Interface Embodied Functionality

While Lotus contemplated seeking over $100 million in damages, Borland appealed to the First Circuit Court of Appeals. Just one week before the scheduled damages trial in district court, the First Circuit unanimously reversed the district court and held that the Lotus 1-2-3 menu command hierarchy constituted a non-copyrightable method of operation under 17 U.S.C § 102(b).[162]

The First Circuit defined a "method of operation" as "the means by which a person operates something, whether it be a car, a food processor, or a computer."[163] The court analogized computers to VCRs.[164] The First Circuit suggested that, just as one could not operate a button-less VCR, one could not operate Lotus 1-2-3 without utilizing the menu command hierarchy.[165] The First Circuit reasoned that any expression in the arrangement of the parts of a VCR is not capable of existing separately from the VCR itself, and, therefore, would not be protectable as original selection and arrangement under copyright law. The court specifically held that the Lotus 1-2-3 menu command hierarchy is a non-copyrightable "method of operation" because it provides the means by which users control and operate the Lotus 1-2-3 spreadsheet program.[166]

According to the First Circuit, "[i]f specific words are essential to operating something, then they are part of a `method of operation' and, as such, are unprotectable. This is so whether they must be highlighted, typed in, or even spoken, as computer programs no doubt will soon be controlled by spoken words."[167] The existence of alternative methods of operation, which was the basis for the district court's finding of copyrightable expression, therefore, is not relevant under the First Circuit's analysis.[168] No other circuit has issued such a far-reaching holding respecting limitations on the scope of user interface copyright.

The First Circuit holding is correct because it is consistent with the utilitarian limitation. However, the First Circuit's opinion provides very little general guidance for the difficult task of separating functional methods of operation from copyrightable expressions contained in user interfaces.[169] The utilitarian limitation can properly guide courts in this task and confirms that the result in Borland is correct.

The utilitarian limitation excludes from copyright any expression which one must copy in order to use a particular method of operation. For example, to use a spreadsheet program, one must use the menu command[170] hierarchy to trigger operations controlled by the spreadsheet program itself. Similarly, Borland had to copy the selection and arrangement of commands in the 1-2-3 menu command hierarchy to implement the 1-2-3 method of operating a spreadsheet. As a result, the selection and arrangement of the 1-2-3 menu command hierarchy constitute a functional method of operation that must be excluded from copyright under the utilitarian limitation.

The utilitarian limitation also would deny copyright protection to 1-2-3's macro language. Prior to Judge Keeton's decision, there was no protection for macro languages under copyright and for good reason. Such protection would be equivalent to extending copyright to a computer programming language;[171] this raises the problem that any program coded in that copyrighted language would represent an infringing derivative work.[172] Extending copyright protection to computer and macro languages would seriously hinder innovation in computer science.[173] The First Circuit's analysis is again consistent with the utilitarian limitation: macro and computer languages are clearly functional methods of operating a computer that the utilitarian limitation would exclude from copyright.

On the other hand, the underlying code that implements the 1-2-3 menu command hierarchy would be copyrightable. A computer programmer could implement the same menu command hierarchy without copying Lotus's underlying code.[174] Like the First Circuit's approach, the utilitarian limitation would not exclude from copyright the elements of the 1-2-3 menu command hierarchy referred to as the long prompts[175] or the help text.[176] These aspects of the 1-2-3 user interface are not necessary to run the 1-2-3 program.[177] One could use the Lotus 1-2-3 method of operation even if there were no long prompts or help text, because such expressive aspects of the 1-2-3 user interface have no bearing on the actual functional use of the program.

In sum, the result in Borland is consistent with the policy behind the utilitarian limitation that protection for the building blocks of user interfaces resides within the exclusive domain of patent law. The First Circuit properly recognized that building upon innovation in computer science requires relying on the method of operation already employed.[178]

2.   The First Circuit Holding as Consistent with Patent Law

The First Circuit's decision in Borland is clearly inconsistent with the Tenth Circuit's holding in Autoskill, Inc. v. National Education Support System.[179] In Autoskill, the Tenth Circuit rejected defendant's argument that plaintiff's keying "procedure," which required users to select responses to the program's queries "by pressing 1, 2, or 3 keys," was a "process" or "method of operation" and held that the procedure constituted a copyrightable element of the user interface.[180] The First Circuit recognized that its decision was in conflict with the Tenth Circuit's Autoskill decision and criticized the Tenth Circuit's failure to hold that the keying procedure was anything but a non-copyrightable method of operation.[181]

The Supreme Court was offered the opportunity to resolve this split in the circuits. Lotus[182] appealed to the Supreme Court and the Court granted certiorari.[183] When the Supreme Court granted certiorari, many predicted that the case would have a significant impact on copyright and patent law.[184]

The case presented the Court with a unique opportunity to draw the line between copyright and patent. Borland essentially asked the court to provide a demarcation between user interface copyright and user interface patent, based upon the utilitarian limitation.[185] The Court's per curiam decision failed to provide an answer[186] and left the circuits in a quandary.[187]

Without Supreme Court guidance, one can only hope that courts will recognize that the First Circuit's decision provides the proper framework of analysis for user interface copyright: it is consistent with the utilitarian limitation and properly accounts for the recent expansion of software patentability under Alappat and the Examination Guidelines.[188] The Borland court viewed the 1-2-3 menu command hierarchy as analogous to the buttons of a VCR, or as a type of virtual machine.[189] Thus, in Borland the First Circuit confirmed that the functional elements of a menu command hierarchy in a user interface are not protected under copyright because such copyright protection would grant patent-like protection to a method of operating a computer. Therefore, the First Circuit approach is consistent with the Examination Guidelines and Alappat, which open the door to patent protection for the functional elements of user interfaces as particular methods for operating a computer.[190]

Indeed, the menu command hierarchy can easily be described in patent terminology as a method of operation. For example, in Lotus's European patent application for "a method of providing a plurality of controls" or simply a "method of operati[on]," Lotus described a user interface as a "control hierarchy": "divid[ing] controls, assign[ing] each control a label, display[ing] the labels according to a design hierarchy, and operat[ing] on the controls according to a functional hierarchy."[191] The PTO has similarly issued utility patents on menu command hierarchies.[192] For example, a spreadsheet software developer obtained patent protection for a system using a command hierarchy similar to Lotus's.[193]

Although patents often issue for a whole class of functionality (for example, all pull-down menu systems), that a very general method of operation may state statutory subject matter for patent does not mean, however, that a more specific method of operation cannot also recite statutory subject matter. As the First Circuit insightfully observed, "`methods of operation' are not limited to mere abstractions."[194] For example, a user interface's particular method of operation can constitute statutory subject matter as an efficient method of operating a computer.[195] In hindsight, Lotus would have been served equally well by a very narrow patent (i.e., narrow claims) because Lotus was merely seeking to prevent Borland from using 1-2-3's exact method of operating a spreadsheet program.[196]

In conclusion, Lotus v. Borland properly contracts the scope of user interface copyright to account for the recent expansion of user interface patent. However, because of the uncertain scope of user interface copyright and the recent expansion of trademark law, several commentators have persuasively argued for an expansive use of trademark law to protect similar aspects of user interfaces.[197]

IV. TRADEMARK AND TRADE DRESS PROTECTION FOR USER INTERFACES

Trademarks represent a third form of federal intellectual property. The recent expansion of this domain of federal intellectual property should have interesting consequences for protecting the intellectual property of user interfaces.[198] A brief overview of trademark law is necessary before considering its application in the user interface context.

A.   Trademark and Trade Dress Law

Trademark subject matter includes almost any mark, such as a design, symbol, word, or combination thereof used by a merchant to identify his or her goods and distinguish them from others.[199] Trademark rights in a mark last forever as long as the mark is not abandoned or does not become a generic term.[200]

Although trademark rights potentially last in perpetuity, federal trademark law provides far less protection to the owner than patent law. The goal of trademark law is to protect against consumer confusion as to the source or origin of goods in the marketplace.[201] It is therefore trademark infringement for someone else to use a confusingly similar mark on the same or closely related goods.[202]

Trade dress law, which is a subset of trademark law, protects a product's overall appearance to the purchasing public. "Trade dress" subject matter includes the total image of a product, possibly including features such as size, shape, color or color combinations, texture, and graphics; and it may encompass the design of the product itself.[203] Trade dress rights, like trademark rights, can last in perpetuity.[204]

As with trademarks, the policy of trade dress is based upon the consumer protection model. Trade dress law, however, provides a slightly broader scope of protection than trademark because it is not necessary to show confusion as to source or origin of goods, but merely confusion as to their association or sponsorship.[205] Trade dress law aims to protect the consumer from use of trade dress that creates a likelihood of confusion or deception about the affiliation, approval, association, or sponsorship of the goods of another.[206]

The hybrid nature of user interfaces can constitute trademark and trade dress subject matter insofar as user interfaces may embody marks or overall images that consumers can identify or associate with commercial goods. Individual elements of user interfaces can provide trademark subject matter. For example, the multicolored apple icon is an individual element of the Macintosh GUI, which identifies the software developer and therefore constitutes trademark subject matter.

As a result of the recent expansion of trade dress protection, several commentators have argued that trade dress law can be used to protect the "look and feel" of a user interface.[207] Interestingly, this argument for expansive trade dress protection of the user interface is analogous to the broad user interface "look and feel" copyright argument which actually imported trademark-like "confusing similarity" standards.[208] The trade dress version of the "look and feel" argument asserts that the identifiable "look and feel" of a user interface can constitute trade dress subject matter. For example, the Macintosh GUI has a distinct and identifiable "look and feel."

This argument for expansive trademark protection for user interfaces, however, overlooks the utilitarian limitation.[209] This Comment asserts that the utilitarian limitation necessarily poses the same exclusionary limitation upon the scope of trademark and trade dress subject matter for the user interface as it does in the user interface copyright context.[210] For example, the Apple district court found many of the individual icons of the Macintosh environment to be too functional for copyright protection. The Ninth Circuit in Apple similarly refused to extend copyright protection to the "look and feel" of the Macintosh GUI.[211] If the utilitarian limitation were properly applied to provide a coherent approach to intellectual property protection for user interfaces, then the result would not have been any different if Apple had argued for trademark or trade dress protection for these functionally expressive elements. The utilitarian limitation should exclude the same aspects of user interfaces from trademark and trade dress that have been properly excluded from user interface copyright, i.e., functional aspects of user interfaces that now constitute patentable subject matter.

B.   User Interface Trademark and Trade Dress Case Law

There are relatively few cases that address the issue of trademark and trade dress protection for user interfaces. The cases addressing user interface trademark and trade dress protection can be divided into two principal categories: (1) the individual elements of the user interface; and (2) the "look and feel" of the user interface.

1.   Individual Elements

In Sega Enterprises Ltd. v. Accolade, Inc.,[212] the Ninth Circuit addressed the issue of trademark and trade dress infringement for an individual element of the user interface of a video game system's screen display. The Sega Genesis console utilizes a trademark security system (TMSS) to restrict the access to its game console to its own game cartridges.[213] When a game cartridge is inserted into the Genesis console, TMSS prompts the game console's microprocessor to search for the initialization code stored on the cartridge. The code consists of 20 bytes plus the letters "S-E-G-A." Upon finding the code TMSS prompts a visual display for approximately three seconds. The screen display reads "PRODUCED BY OR UNDER LICENSE FROM SEGA ENTERPRISES LTD" (the "Sega Message").[214] TMSS is analogous to a functional lock and key. Sega had intended that only licensees would have the necessary key to unlock the Genesis console.[215]

Defendants, who lacked a license, reverse-engineered TMSS to unlock the Genesis console and thereby design compatible game cartridges.[216] Defendants' game cartridges would, however, also prompt the Sega Message on the screen display. Sega sued Accolade under § 32(1)(a) of the Lanham Act (trademark infringement) and §; 43(a) (trade dress infringement).[217]

The Ninth Circuit observed that trademark law does not protect essentially utilitarian elements because "such protection would constitute a grant of a perpetual monopoly over features that could not be patented."[218] The court noted that the only issue in the case was whether competitors had feasible alternatives for cartridge compatibility without having the Sega Message display on the screen or whether the TMSS initialization code was functional.[219] The Ninth Circuit held that Sega had not yet met its burden in proving feasible alternatives and was not entitled to preliminary relief.[220]

This case suggests that while trademark protection is available for individual elements of the user interface, a "trademark owner may not enjoy a monopoly over the functional use of the mark."[221] Thus, the utilitarian limitation as implicitly applied in Sega prevents companies from utilizing user interface trademark protection to avoid the stricter requirements of patent law, thereby gaining a perpetual monopoly over a functional user interface element.

2.   "Look and Feel"

In Engineering Dynamics, Inc. v. Structural Software, Inc.,[222] the district court addressed claims for trade dress protection of the "look and feel" of a user interface. Plaintiffs developed a highly technical structural engineering program after considerable time and effort. Plaintiffs alleged that some engineers in the field would recognize the "look and feel" of the program's user interface—its input and output formats. Defendants developed a competing program with a substantially similar user interface. The primary users of both products were engineers in the offshore engineering community.

The district court held that given the relatively sophisticated nature of the offshore engineering community, the users of these programs were not likely to confuse the two software products.[223] However, the district court did not reach the merits of whether the "look and feel" of the program was protectable under trade dress law.

On appeal, the Fifth Circuit held that the same input formats of the user interface were copyrightable.[224] The Fifth Circuit did not find that the functionality of such formats (as methods of operating a computer) excluded copyrightability. But the Fifth Circuit suggested in dicta that the functionally expressive aspects of such input formats and output reports (as a method of operating a computer) may exclude them from protection under trade dress law.[225] Therefore, the Fifth Circuit indicated that elements of a user interface can be too functional for trademark and trade dress law protection but yet not too functional for copyright protection.

The Fifth Circuit's dicta implicitly recognize the utilitarian limitation of trademark and trade dress protection insofar as they suggest restricting the scope of "look and feel" protection for functional aspects of a user interface. Thus, the Fifth Circuit's dicta appropriately suggest that the court would exclude trade dress protection for the functional "look and feel" of a user interface. This proposition is consistent with the Ninth Circuit's refusal to grant patent-like protection to the "look and feel" of the elegant Macintosh user interface through the vehicle of copyright.[226]

Unfortunately, the Fifth Circuit's trade dress dicta are inconsistent with its copyright holding. The Fifth Circuit's internally inconsistent opinion demonstrates that it lacked a coherent approach for applying federal intellectual property protection to the user interface. If the Fifth Circuit had properly applied the utilitarian limitation, it should have reached a coherent result and denied copyright as well as trade dress protection to the functionality of EDI's user interface.

V. THE UTILITARIAN LIMITATION

Protecting the hybrid nature of the user interface has presented courts with a challenging task. In Engineering Dynamics the Fifth Circuit demonstrated that the federal courts still lack a coherent approach to this problem. The question remains as to what type of guidance can be gleaned from the statutes and case law, and in particular, In re Alappat and Lotus v. Borland, in providing courts with a coherent approach for applying patent, copyright, and trademark law to the hybrid subject matter of the user interface. This Comment argues that the utilitarian limitation provides the correct and most coherent approach.[227]

A.   User Interface Embodied Functionality as the Exclusive Domain of Patent Law

The Supreme Court has emphasized that a statutory provision should be interpreted in light of "`the provisions of the whole law, and to its object and policy.'"[228] The Supreme Court has also suggested that statutes should be construed in harmony with the Constitution and other applicable statutes.[229] Accordingly, the entire federal intellectual property regime must be examined, not just its individual domains. This suggests that copyright and trademark law should be read in accordance with the requirements and policy aims of the patent system, which translates into statutory form the Patent Clause of the United States Constitution.

Applying this approach, trademark and trade dress protection clearly should not extend to "functional" aspects of the user interface that now constitute statutory subject matter under patent law. According to the Supreme Court in Qualitex Co. v. Jacobsen Products Co.,[230] "`[i]n general terms, a product feature is functional,' and cannot serve as a trademark, `if it is essential to the use . . . of the article.'"[231] The utilitarian limitation for trademark and trade dress subject matter[232] embodies the related principle that trademark or trade dress law may not be used to secure patent-like protection for useful product features.[233]

Moreover, in Elmer v. ICC Fabricating, Inc.,[234] the Federal Circuit recently stated that "[i]t is the province of patent law, not trademark law, to encourage invention by granting inventors a monopoly over new product designs or functions for a limited time. . . . "[235] Thus, trademark and trade dress protection for patentable elements of a user interface would disrupt the balance envisioned by the patent system. The policy underlying the exclusion of useful methods and systems from copyright justifies the limitation on trademark and trade dress protection for user interfaces.[236]

The "historic kinship" between patent and copyright law was born in the Constitution itself.[237] Congress created this kinship by enacting separate and distinct statutes.[238] This division forms the basis of the utilitarian limitation for user interface copyright. Granting copyright protection to the patentable subject matter of user interfaces undermines this intentional division and disrupts the careful balance envisioned by the patent system.

The Supreme Court recognized the exclusivity of the patent domain as a limitation on the scope of copyright protection for functionality in a long line of cases. In Baker v. Selden,[239] the Court established the utilitarian limitation as a doctrine stating or holding that useful methods and systems should be protected, if at all, only if they meet the standards for patentability.[240]

Selden, who developed a bookkeeping system, failed to secure a patent on his system.[241] Selden, however, copyrighted his book, entitled Selden's Condensed Ledger, or Book-keeping Simplified.[242] The Selden system achieved the same results as double-entry bookkeeping, but effected the operation by a peculiar arrangement of columns and headings in an account-book.[243]

The Court made a distinction between copyright and patent for utilitarian subject matter:

The novelty of the art or thing described or explained has nothing to do with the validity of the copyright. To give the author of the book an exclusive property in the art described therein, when no examination of its novelty has ever been officially made, would be a surprise and fraud upon the public. That is the province of letters-patent, not of copyright. The claim to an invention or discovery of an art or manufacture must be subjected to the examination of the Patent Office before an exclusive right therein can be obtained; and it can only be secured by a patent from the government.[244]

The Selden Court suggested that an author cannot gain patent-like rights over patentable subject matter through the vehicle of copyright. The Court specifically rejected Selden's attempt to enforce an exclusive right to use a particular system through a copyright infringement action, because that right can only be secured by a patent.[245]

Selden also excludes copyright protection for functional expression of a "system" or "method of operation" "where the art it teaches cannot be used without employing the methods and diagrams used to illustrate the book, or such as are similar to them, such methods and diagrams are considered as necessary incidents to the public."[246] In Selden, the Court first used the term "method of operation" to describe that which "is the province of letters-patent (functional patent), not of copyright."[247] Thus, patent and copyright should protect exclusive aspects of a work,[248] for example, a user interface.

The Supreme Court, in Bonito Boats, Inc. v. Thunder Craft Boats, Inc.,[249] recently reiterated that "the efficient operation of the federal patent system depends upon substantially free trade in publicly known, unpatented design and utilitarian conceptions."[250] According to the Bonito Boats Court, providing "patent-like protection" for "the functional aspects of a product which had been placed in public commerce absent the protection of a valid patent" upsets the balance set out by "the federal patent system."[251] Similarly, a software developer, through the assertion of copyright, trademark, or trade dress, should not be allowed to secure all the benefits of patent protection for user interface functionality without meeting the strict statutory requirements of patent law including "novelty," "non-obviousness," and "best mode" disclosure.[252]

In Lotus v. Borland, the First Circuit relied on Baker v. Selden in a controversial decision to exclude a user interface's functional "method of operat[ing]" a computer from copyright.[253] But as one Supreme Court Justice noted in the Lotus v. Borland oral arguments:

if the only way to practice the art in Baker v. Selden is to copy the concise words at the head of the accounting columns, the headings are unprotected. Your answer may be that use of those same words is not absolutely necessary, but how much do we read into "absolutely necessary."[254]

The Justice asked the critical question: how far does Baker v. Selden reach—how do courts separate copyright, trademark, and trade dress subject matter from "method[s] of operation" and "system[s]" which are the exclusive domain of patent law?

Assuming that one must interpret and apply patent law to answer the above question, it is not surprising that the Supreme Court chose not to attempt an answer.[255] The Federal Circuit clearly is the federal court with the most expertise in patent law. The Supreme Court admittedly lacks patent law expertise. The Court has stated, for example that, "the issue . . . involves the practicalities of the Federal Circuit's specialized patent jurisdiction, rather than matters of statutory or constitutional interpretation with which we are familiar."[256] Perhaps even more telling is that in the most controversial software patent decision since Diehr, the Justice Department elected not to appeal the Federal Circuit's decision in Alappat, presumably because it did not trust the Supreme Court to give as thoughtful and expert an opinion as the Federal Circuit had. Thus, the Supreme Court has not given a clear answer to the question; and it is possible that the Court may never do so because answering the question essentially requires the Court to determine the patentability of the user interface.

B.   The Utilitarian Limitation Defined and Applied

In the absence of Supreme Court guidance, the Federal Circuit pro vides the best insights and guidance into the proper application of the utilitarian limitation.[257] Applying the utilitarian limitation essentially requires courts to issue shadow patent rulings. In other words, to determine whether a particular aspect of a user interface should be excluded from copyright, trademark, or trade dress protection, a court must first determine if that aspect of the user interface constitutes statutory subject matter under patent law. Thus, the Federal Circuit's expertise in patent law suggests that it is the court best prepared to give guidance for the proper application of the utilitarian limitation.

The Federal Circuit defined and applied the utilitarian limitation in Atari Games Corp. v. Nintendo of America, Inc.[258] The case involved video game systems. Nintendo utilized a program that acted as a lock and key, restricting the access of game cartridges to its game console. Defendants made unauthorized copies of the program and reverse-engineered the lock and key system.[259]

The Federal Circuit reasoned that processes or methods of operation must be protected under patent law.[260] It found that the arrangement of instructions in the 10NES computer program "does not merge with the process so long as alternate expressions are available. In this case, Nintendo has produced expert testimony showing a multitude of different ways to generate a data stream which unlocks the NES console."[261] Thus, the Federal Circuit determined that the utilitarian limitation would not remove copyright protection from this computer program even though the purpose of the program was to operate or access the NES console.

The Federal Circuit's decision in Atari basically defined the utilitarian limitation. The court simply stated that "[i]n conformance with the standards of patent law, title 35 provides protection for the process or method performed by a computer in accordance with a program."[262] Accordingly, only patent law should grant the exclusive right to use an abstract or particular method of operation embodied in a user interface.[263]

The utilitarian limitation does not require a per se bar of copyright, trademark, and trade dress protection for the aspects of a user interface that embody a method of operation. If one can use a particular method of operation without copying the exact expression (i.e., alternative expressions of that particular method exist or, alternatively, there is no merger under the principles of Baker v. Selden), then such functional expression may be copyrightable.[264] "If the patentable process is embodied inextricably in the line-by-line instructions of the computer program, however, then the process merges with the expression and precludes copyright protection."[265]

However, according to the Federal Circuit, the existence of alternative methods (rather than alternative expressions for a particular method) for performing the same function is irrelevant under the utilitarian limitation.[266] This approach is consistent with the patent law's approach to methods of operation. Patents often recite prior methods of performing the same function in distinguishing the claimed invention from the prior art.[267] Thus, just as the existence of alternative methods does not bar a user interface patent for a particular method, the existence of alternative methods for achieving the same functionality cannot be the basis for justifying an extension of copyright, trademark, or trade dress to that user interface's particular method of operating a computer. As a result, even if alternative methods exist, a functionally expressive aspect of a user interface providing a particular method of operating a computer must be excluded from user interface copyright, trademark, and trade dress protection if one must copy that particular expression in order to use that embodied method of operation.[268]

Accordingly, when applying the utilitarian limitation, a court must first determine whether the subject matter at issue constitutes statutory subject matter under patent law as defined by Alappat and by the PTO in the Examination Guidelines.[269] For example, a software company clearly cannot protect functional aspects of menu-driven command systems like the selection and arrangement of commands in the 1-2-3 menu command hierarchy.[270] These aspects of a user interface constitute statutory subject matter under patent law that the utilitarian limitation would exclude from copyright, trademark, and trade dress.[271]

On the other hand, nonfunctionally expressive aspects of a user interface such as help text and logos generally do not embody patentable subject matter. The utilitarian limitation would not bar these aspects from protection under copyright, trademark, or trade dress.[272]

Applying the utilitarian limitation requires courts like the First Circuit in Lotus v. Borland to make difficult decisions. Even though user interface functionality did not constitute patentable subject matter in 1983 when Lotus 1-2-3 was first created, the First Circuit correctly withheld copyright protection to the 1-2-3 menu command hierarchy for user interface functionality that now constitutes patentable subject matter. The First Circuit's controversial ruling demonstrates the proper approach to copyright for functionally expressive aspects of the user interface because it properly accounts for the recent Federal Circuit and PTO expansion of the scope of patentability for user interfaces. Courts will confront similarly difficult cases as the Federal Circuit and PTO continue to expand the scope of user interface patents and as software companies continue to aggressively provide protection for their innovative and creative user interfaces.

VI.   CONCLUSION

In conclusion, the utilitarian limitation properly accounts for the recent Federal Circuit and PTO expansion of user interface patent protection because it requires courts essentially to issue shadow patent rulings. As a result, the utilitarian limitation dictates a contraction of the scope of user interface copyright, trademark, and trade dress protection to ensure that functional aspects of user interfaces that are patentable are the exclusive domain of patent law.

User interface copyright, trademark, and trade dress protection will also shrink because of changes in the marketplace and in computer technology itself. Computer software, for example, attempts to emulate the real world to create the most intuitive computing environment—a virtual world.[273] As user interface technology approaches "virtual reality," it will cease to contain original (copyrightable) expression. Because it will mimic the real world so closely, it will lack sufficient originality for copyright.[274] The trend toward compatible user interface environments will also decrease the availability of trademark and trade dress law protection as user interfaces will no longer be uniquely identifiable to the consumer.

These trends in user interface technology and the marketplace, combined with the PTO's Examination Guidelines and the Federal Circuit's recent expansion of patentable subject matter, suggest that innovative user interfaces will primarily be protected under patent law. During this period of transition, courts will continue to address vexing questions about federal intellectual property law as applied to user interfaces. The utilitarian limitation can properly guide courts during this transition period and assure that they apply a coherent approach to protecting the functional aspects of user interfaces under federal intellectual property law.

"Lord Ellenborough expressed the inherent tension in the need simultaneously to protect copyrighted material and to allow others to build upon it when he wrote, `while I shall think myself bound to secure every man in the enjoyment of his copy-right, one must not put manacles upon science.'"[275] The utilitarian limitation recognizes that copyright as well as trademark and trade dress protection for functional aspects of the user interface can stifle computer software innovation. It therefore excludes such elements from copyright, trademark, and trade dress protection and requires that such aspects can be protected, if at all, only by patent law. The utilitarian limitation ensures that developers have access to the specifications needed to make compatible products. This in turn spurs innovation and competition in user interface development, which is a critical subject matter in the information age. Software companies will continue to adjust to this rapidly changing marketplace and shifting legal paradigm to protect their investments in research and development through expanded patent protection for truly innovative user interfaces.

MICHAEL J. SCHALLOP[*]

© 1998 Kiosks.Org.
All Rights Reserved.