ICTLINTELL – 2007
Computer program protection under copyright law
(19th September 2007)
Lee Bygrave
1. Introduction -
Economic importance of software protection.
-
“Constitutional” importance of software protection.
-
Various legal means of protecting software (e.g., contract; design law, trade secrets law / law of breach of confidence, patent law) – i.e., copyright = just one avenue of protection.
2. What = computer program?
Not directly defined in international legal agreements.
Computer Program / Software Directive provides that term “computer program” shall embrace “preparatory design material” (Art. 1(1)) and shall include programs “in any form, including those which are incorporated into hardware” (recital 7).
Query: What is advantage of this failure to define? Cf. EC Commission’s rationale in OJ C 91, 12.4.1989, p. 5. What is disadvantage?
WIPO Model Provisions on the Protection of Computer Software 1978 define “computer program” as “a set of instructions capable, when incorporated in a machine-readable medium, of causing a machine having information-processing capabilities to indicate, perform or achieve a particular result” (Art. 1).
US Copyright Act 1976, s. 101: “A ‘computer program’ is a set of statements or instructions to be used directly or indirectly in a computer in order to bring about a certain result”.
Australia’s Copyright Act 1968, s. 10(1) used to define “computer program” as:
“an expression, in any language, code or notation, of a set of instructions (whether with or without related information) intended, either directly or after either or both of the following:
(a) conversion to another language, code or notation;
(b) reproduction in a different material form;
to cause a device having digital information processing capabilities to perform a particular function”.
That definition was amended, however, by the Digital Agenda Act 2000 to follow the definition in US legislation.
Note distinction between source code and object code:
-
Object code = computer program expressed in machine-readable form (i.e., can be interpreted directly by computers (machines) without a translation process).
-
Source code = computer language expressed in human-readable form and which allows a human programmer to work with instructions (rather than binary or hexadecimal numbers).
Source code is most valuable and hence most in dispute. Hence, decompilation and, more generally, reverse engineering are contentious processes.
Note distinction between code and algorithm:
-
Algorithm = a finite set of well defined rules for the solution of a problem in a finite number of steps (ISO).
Considerable litigation on issue as to what = computer program (as literary work).
See, e.g., series of Australian cases on this issue during 1980s. Most (in)famous = Computer Edge Pty. Ltd. v. Apple Computer Inc. (1986) 161 CLR 171; 6 IPR 1 (the Apple case) – holding, i.a., that programs written in source code (but not programs in object code) = literary works (for purposes of Copyright Act as it then stood); programs in object code held not to be literary works b/c they consisted of electrical impulses which could not be perceived by the senses and were not intended to convey any message to a human being and did not represent words, letters, figures or symbols). Decision of historical interest only as Copyright Act subsequently amended to cover object code.
Cp. situation in USA and UK where courts were, from early 1980s more accepting of copyrightability of object code, though legislative amendments still made to respective copyright legislation to clarify protection of object code. In USA, these amendments were partly in response to the problem of “invisible copies” in traditional US copyright law (viz. White-Smith Music Publishing Company v. Apollo Company 109 US 1 (1907)).
Cp. Germany: works must be product of personal intellectual creations (“persönliche geistige Schöpfungen”); decision of Federal Supreme Court (Bundesgerichtshof) in 1985 held that computer software not copyrightable unless it went beyond skills of average computer programmer – i.e., represented “Überdurchschnittlichkeit”. Doubt also whether object code was copyrightable. Some court cases deciding that screen displays not copyrightable. See Hoeren, in H.D.J. Jongen & A.P. Meijboom (eds.), Copyright Software Protection in the EC (Deventer/Boston: Kluwer, 1993), p. 73ff.
A related issue attracting litigation = protectability of “look and feel” of programs
Two overlapping aspects here:
-
graphic user interface – “look” [presentation on computer screen of icons generated by computer program and facilitating access to and use of programmed material];
-
the “sequence, structure and organisation” of a program – “feel”.
Leading case law on point is from US Court of Appeals:
-
Whelan Associates v. Jaslow Dental Laboratory, 797 F.2d 1222 (3rd Cir. 1986) – holding that structure, sequence and organisation (SSO) of program = protectable as part of literary work;
-
Lynx Ventures v. Miller, 45 Fed. App. 468 (2nd Cir. 2002) – holding that substantial reproduction of database (which was literary work) was infringement even if no substantial similarity in terms of “total concept and feel”.
-
Computer Associates International Inc. v. Altai, 23 IPR 385 (2nd Cir. 1992) – rejecting Whelan rationale on SSO and adopting “abstraction-filtration-comparison” test.
See further Sterling §§ 13.25 – 13.26 (though not entirely satisfactory description of state of play).
Note importance of idea/expression distinction (and associated “merger doctrine”) in this case law.
-
Cf. Lotus Development Corp. v. Borland International litigation – US Sup. Ct. failed to conclusively resolve issue whether menu structure in Lotus 1-2-3 could be protected; affirmed (by 4-4 vote) 1st Circ. Ct. of Appeal decision (which was also equally divided!) that menu structure = method of operation that could not be copyrighted.
-
Note role played by patent in US law; cf. Europe where computer programs in themselves are not patentable (European Patent Convention s. 52), though they can be patented if part and parcel of invention that goes beyond such programs.
Query: what authority does US case law have for European jurisdictions?
No express reference to computer programs in Berne Convention, but category of “literary and artistic works” as defined in Art. 2(1) is broad enough to encompass such programs.
In 1970s consideration given to developing sui generis protection for computer software: WIPO Model Provisions on the Protection of Computer Software 1978 = manifestation of this way of thinking. But failed to gain widespread support. See further Sterling §§ 6.19, 91.04.
General approach since 1980s has been to extend protection to computer programs under copyright law, deeming them as “literary works”.
TRIPS Agreement: computer programs (in source and/or object code) to be protected as literary works (Art. 10(1)).
WCT: similar – protection (as literary works) to be given to programs “whatever may be the mode or form of their expression” (Art. 4).
Query: would sui generis approach as proposed by WIPO in 1970s provide better regulatory solution?
Transitional arrangements: TRIPS Agreement stipulates that least developed countries do not have to extend protection until 2006 (Art. 66(1)).
4. EC Computer Program Directive
For overview, see Sterling 2003, § 26E (required reading).
Background: Lack of harmony – Germany as major “problem” jurisdiction due to stringent originality requirement. Competition with USA and Japan. Part of “Europe 1992” wave of reform to realise internal market. Controversy during drafting of Directive, espec. on reverse engineering.
Object of protection: “computer programs” and their “preparatory material” (both terms undefined). What is extent of protection of preparatory material? Cf. Meijboom in H.D.J. Jongen & A.P. Meijboom (eds.), Copyright Software Protection in the EC (Deventer/Boston: Kluwer, 1993), p. 8.
Basic exclusions: “ideas” and “principles” including those underlying program’s “user interface”. Note also reference to “logic” and “algorithms” in recital 14.
Copyright ownership in employment context: Provisions in Article 2(3) important due to previously conflicting approaches (cp. previous situation in Belgium, Germany and France with situation in UK and Netherlands).
Rightholder rights: exclusive control over reproduction (permanent or temporary), translation, arrangement or other alteration of program, and distribution to public (up to point of first sale) of original or altered version (Art. 4). Rental (but not public lending: recital 16) of program not permitted without owner’s consent (Art. 4(c)).
User rights (or rather freedoms):
User = “lawful acquirer” (not defined; cf. category “persons having right to use” in Art. 5(2) and (3) – also not defined) of program is able to reproduce program, and to translate, adapt, arrange or otherwise alter it (but not distribute it to public), when these actions are “necessary” for the program’s “intended purpose” (Art. 5(1)). One such purpose = “error correction”.
What = “necessary”?
What = “intended use”?
What = “error correction”?
Cf. slightly narrower approach in US law: US Copyright Act grants “owner” of copy of software program the inherent right to make additional copies or adaptations for two limited purposes: (i) where copy or adaptation = “essential step” in utilization of program on computer, or (ii) when it is for archive purposes. The few courts that have interpreted “essential step” right have limited it to right to copy program into non-permanent memory to run program. Also, since right is given only to “owner” of copy, it is easily negated by program owner granting license, rather than selling “ownership” of copy.
The ability to carry out the actions in Art. 5(1) can be overridden by contract (but see recital 18 which on one reading seems to say that some of the actions designated in Art. 5(1)).
Back-up copies may be made if “necessary” for persons having “right to use” program, to use program – contractual restrictions on this ability are unenforceable (Art. 5(2)).
Another ability = observe, study or test functioning of program in order to determine its underlying “ideas and principles”. Only applies if exercised while loading, displaying, running, transmitting or storing the program, and only if person is entitled to perform the function in question.
“Decompilation” (full reverse engineering?): Directive allows reproduction and translation of the form of program code, without consent of owner, but only for purpose of achieving interoperability of program with some other program, and only if this reverse engineering is “indispensable” for this purpose. Note too recital 22 (indicating that circumstances in which decompilation permitted are “limited”). If information needed to achieve interoperability is “readily available” (Art. 6(1)(b)), then decompilation not permitted. What = readily available? See further Meijboom, op cit., p. 16). Note too Art. 6(3): the permitted reverse engineering may not be interpreted so as to “unreasonably prejudice” the program owner’s “legitimate interests” or conflict with program’s “normal exploitation” (cf. Berne Convention’s 3-step test in Art. 9(2)).
The approach to decompilation is roughly in line with US law: see espec. Sega Enterprises v. Accolade 977 F. 2d 1510 (9th Cir. 1992) – described in Sterling § 10.29(h).
Term of protection (Art. 8): is now life of author + 70 years (viz. Directive 93/98/EC).
Share with your friends: |