Kremer, Ben --- "Copyright and Computer Programs: Data Access v Powerflex Before the High Court" [1998] SydLawRw 12; (1998) 20 (2) Sydney Law Review 296
[*] BSc (Hons), LLB (Hons) (Syd). MSc candidate at the University of Sydney and DPhil candidate at Magdalen College, Oxford)
[1] On appeal from Powerflex Services Pty Ltd v Data Access Corporation [1996] FCA 460; (1997) 37 IPR 436.
[2] For instance, “the expression of thought in print or writing”: University of London Press Ltd v University Tutorial Press Ltd [1916] 2 Ch 601 at 608–9.
[3] Autodesk Inc v Dyason [1992] HCA 2; (1992) 173 CLR 330.
[4] Autodesk Inc v Dyason (No 2) [1993] HCA 6; (1993) 176 CLR 300 at 303 per Mason CJ.
[5] The compatibility of this low-level source code with that produced by the Dataflex program itself is not mentioned in the judgment; it would be a great commercial advantage for Powerflex if it were compatible. This fact would not affect the legal issues in this case.
[6] For example, Autodesk Australia Pty Ltd v Cheung [1990] FCA 121; (1990) 94 ALR 472; possibly termed piracy jure pentium.
[7] They are so called because the user may not use these words themselves; ie, cannot ascribe a custom function to them. The language already has ascribed one and the word is thus “reserved” to the language and unavailable to the user.
[8] “High-level” is computer industry jargon for “more like English”, low-level means “less like English”, so that a high-level source code (like BASIC) is easier to comprehend, even to a professional programmer, than a low-level one such as assembly language.
[9] The reasons for utilising a two-step high-level source to low-level source to object code conversion in place of the usual source code to object code conversion are not relevant here; they involve considerations of practicality and convenience to the vendor of the Dataflex program and its users.
[10] It is unclear whether Powerflex comprises just one program or whether it too has a language and compiler program but this fact is immaterial to the legal issues.
[11] Above n4.
[12] Id at 310–11, 329 per Brennan and Gaudron JJ respectively.
[13] Id at 322.
[14] Id at 329.
[15] This example is over-simplified; the highly uneconomical use of registers is to keep the illustration conceptually simple.
[16] It is worth noting that, theoretically, it is impossible to distinguish between data and instructions as any data may potentially be executed; equally, it is theoretically impossible to prove that any program will work. Practically, the first issue at least is moot.
[17] Above n3 at 343.
[18] Id at 346.
[19] Above n4 at 310–11.
[20] Computer Edge Pty Ltd v Apple Computer Inc [1986] HCA 19; (1986) 161 CLR 171.
[21] Above n3 at 343.
[22] That is, the process might be originally: take a number; send it to the peripheral; wait a short time; retrieve a number from the peripheral; calculate the correct response; check the number received against the correct response; does it match: if yes, go to point A and continue; if no, go to point B and continue. Point A would be the position to resume the program’s calculations, display etc; point B would initiate a message something like “you are not using the correct peripheral; goodbye” and quits the program. “Cracking” the protection might involve changing the last check to: does it match: if yes, go to point A and continue; if no, go to point A and continue. This may sometimes be as simple as changing one hexadecimal character in memory.
[23] Above n1 at 451–2.
[24] The cause is the instructions for which the label stands. Even where the word is “used” by the operator to achieve a result, the cause of the computer performing the result is the issuance of a command by the user, not the word itself.
[25] Above n3 at 343.
[26] Id at 335 per Mason, Brennan and Deane JJ.
[27] In the sense that, in Autodesk, Widget.C was a program in and of itself despite the fact that it made up the AutoCAD program when combined with many other programs: see above n1 at 447. See also at 455.
[28] Above n3 at 335–6.
[29] Above n4.
[30] Above n3 at 345.
[31] Id at 346.
[32] Above n4 at 306.
[33] Id at 304.
[34] Above n4 at 310–11 and 329.
[35] That is, as both programs would contain the words in their source and object code as literal strings, one copied from the other.
[36] Above n3 at 346.
[37] Above n4 at 311–12.
[38] Id at 330.
[39] Above n4 at 305–6.
[40] Above n3 at 343.
[41] Id at 344–5.
[42] Baker v Selden [1879] USSC 9; (1880) 101 US 99.
[43] Above n3 at 345.
[44] See the judgment of the Cour d’Appel of Paris on similar facts to Autodesk in Sarl Artware v Groupe D’Utilisation Francophone D’Informatique [1993] ECC 279 noted [1993] FSR 703.
[45] Skone James, E P, Mummery, J F, Rayner James, J and Garnett, K M, Copinger and Skone James on Copyright (13th edn, 1991) at par 8–26.
[46] Francis Day & Hunter Ltd v Twentieth Century Fox Corporation Ltd [1940] AC 112 at 122–3. But cf Ladbroke (Football) Ltd v William Hill (Football) Ltd [1964] 1 WLR 273 at 286.
[47] Note that quality is not used exclusively in place of quantity: see Lord Cottenham LC in Bramwell v Halcomb (1836) 3 My & Cr 738 at 738 [1836] EngR 890; [40 ER 1110 at 1110] but cf Lord Pearce in Ladbroke (Football) Ltd v William Hill (Football) Ltd [1986] 1 AC 577 at 613, 631–632, 647 “quality rather than its quantity”.
[48] Catnic Components Ltd v Hill & Smith Ltd [1982] RPC 183 at 223 (CA).
[49] Hawkes & Son (London) Ltd v Paramount Film Service Ltd [1934] 1 Ch 593 at 609.
[50] Ladbroke (Football) Ltd v William Hill (Football) Ltd [1964] 1 WLR 273 at 283.
[51] Above n23 at 454.
[52] See memorandum to the bill amending the Copyright Act discussed above n51.
[53] Data Access Corporation v Powerflex Services Pty Ltd (1996) 33 IPR 194 at 202.
[54] Id at 203.
[55] Above n50 at 293 per Lord Pearce and Mason CJ in Autodesk v Dyason (No 2) above n4 at 305.
[56] Above n53 at 202.
[57] Cf John Richardson Computers v Flanders [1993] FSR 497.
[58] British Leyland Motor Corporation v Armstrong Patents [1986] 1 AC 577 at 613, 631–2, 647.