NZLII Home | Databases | WorldLII | Search | Feedback

Court of Appeal of New Zealand
You are here:  NZLII >> Databases >> Court of Appeal of New Zealand >> 2014 >> [2014] NZCA 389

Database Search | Name Search | Recent Decisions | Noteup | LawCite | Download | Help

Karum Group LLC v Fisher & Paykel Financial Services Ltd [2014] NZCA 389; [2014] 3 NZLR 421 (13 August 2014)

Last Updated: 30 January 2018

For a Court ready (fee required) version please follow this link
     
IN THE COURT OF APPEAL OF NEW ZEALAND
BETWEEN
 Appellant
AND
 Respondent
Hearing:
9, 10, 11, 12 June 2014
Court:
Harrison, Stevens and Miller JJ
Counsel:
J G Miles QC, Z G Kennedy and M Toulmin for Appellant A R Galbraith QC and J F Anderson for Respondent
Judgment:


JUDGMENT OF THE COURT

  1. The appellant’s appeals and the respondent’s cross-appeals against [2012] NZHC 240 and [2012] NZHC 794 and CIV-2006-404-6646, 11 October 2011 are dismissed.
  2. The appellant’s appeal against [2012] NZHC 3314 is dismissed.
  1. The appellant is ordered to pay costs for a complex appeal on a Band B basis together with usual disbursements. We certify for two counsel.

____________________________________________________________________

REASONS OF THE COURT

(Given by Harrison and Miller JJ)

Contents

Introduction

[1] This claim for enforcement of intellectual property rights falls within the orthodox parameters of misrepresentation, misleading and deceptive conduct, infringement of copyright and breach of confidence. Its complexity lies in the novelty of the copyright claim, its technical detail and the structure of the relief sought.
[2] Fisher & Paykel Financial Services Ltd (FPF) provides financial products and services including credit cards and hire purchase loans. In 2004 the company acquired Retail Financial Services Ltd (RFS), a wholly owned subsidiary of The Farmers Trading Company Ltd (Farmers), a substantial retailer. RFS operated the company’s credit business by providing a revolving credit product and hire purchase accounts to its customers. FPF’s intention was to transfer all RFS’ financial business to its software programme, known as Lending, which was then being developed.
[3] At the time of acquisition FPF assumed that RFS owned its own credit management software system, known as CMS. It was unaware that RFS had for ten years used the system under licence from its developer, an American company, Karum Group LLC, previously known as Credit Management Services Inc. FPF was unaware also that the licence provided for termination if Farmers’ retail and credit divisions ceased to remain in common ownership.
[4] When he learned of the acquisition, Peter Johnson, Karum’s founder and chief executive officer, immediately claimed that FPF was operating the CMS software without a licence and infringing Karum’s rights during the transition process to Lending. He was not placated by FPF’s assurances to the contrary. Attempts to negotiate a settlement of the resulting dispute were unsuccessful. In late 2004 Karum issued a proceeding against FPF in the High Court seeking permanent and interim injunctive relief.
[5] Negotiations between the parties resumed immediately and within a month terms of settlement were agreed. In essence, in exchange for Karum’s licence to use its software through the transition process, FPF agreed to pay transfer and licensing fees and warranted that its own software system was not and would not be based on Karum’s system.
[6] Just over a year later, when the licence expired, Mr Johnson’s concerns resurfaced. Karum gave notice of cancellation of the settlement agreement on account of prior misrepresentations allegedly made by FPF. Karum also issued another proceeding in the High Court seeking a declaration of valid cancellation of the settlement agreement together with damages. FPF defended the claim and an eight week trial eventually ensued.
[7] Rodney Hansen J gave judgment in FPF’s favour.[1] He dismissed all of Karum’s claims of misrepresentation or misleading and deceptive conduct (collectively the misrepresentation claims). He held further that if, contrary to his primary finding, FPF did make misrepresentations, they did not induce Karum to enter into the settlement agreement. He also dismissed Karum’s separate claims for infringement of copyright and breach of confidence (collectively the intellectual property claims). It was unnecessary for the Judge to consider relief.
[8] Karum appeals against all of Rodney Hansen J’s findings. We shall address its submissions on its principal causes of action in the same order as the Judge.
[9] Karum emphasises that it is exercising a general right of appeal. Mr Miles QC correctly asserts that the company is entitled to judgment in accordance with this Court’s opinion even though the impugned decision amounts to an assessment of fact and degree and involves a value judgment.[2] Nevertheless, Karum must still show that the judgment is wrong.[3] In that respect we record that Rodney Hansen J enjoyed the particular benefit of seeing and hearing the witnesses and an evolving appreciation of all the evidence, including documentary material produced and referred to by witnesses, over a lengthy trial where all possible issues were exhaustively explored.[4]
[10] Three separate appeals were also filed against related earlier judgments and a ruling delivered by Rodney Hansen J.[5] The first two were Karum’s appeals against the Judge’s decisions that it was not the owner of the software and that further evidence was inadmissible. FPF crossappealed against the Judge’s decisions that Karum was the exclusive licensee of the CMS system. For reasons given later in this judgment,[6] these appeals and cross-appeals are dismissed. Mr Miles withdrew a separate appeal by Karum against a costs judgment for post-trial applications.[7]

Background

[11] Rodney Hansen J set out in considerable detail the background to the settlement agreement reached on 7 January 2005 following a meeting between representatives of the parties in San Francisco.[8] It is unnecessary for us to attempt to replicate the Judge’s narrative, except to emphasise three features.
[12] One feature is that the parties and their lawyers were engaged in extensive dialogue from June 2004 to early January 2005. Representatives met on at least two occasions. There were also telephone conferences and a great deal of correspondence. As noted, it came to nothing, and Karum issued its proceeding in the High Court in mid December 2004.
[13] Another feature is Farmers’ purpose in entering into a licensing agreement with Karum in 1994. It was to develop a new credit management system to support Farmers’ longstanding revolving credit policy, enabling customers and later third party merchants to purchase goods on credit using the Farmers card. Karum provided some maintenance and support services until 1998. But after that Farmers and RFS modified and adapted the CMS system themselves. As Rodney Hansen J pointed out, when FPF acquired RFS it had been using the CMS system for 10 years, for most of that time independently of Karum and its predecessor companies.[9]
[14] A third and related feature is that for a time after it acquired RFS, FPF maintained parallel business systems for each entity. RFS staff maintained the CMS platform while processes were developed for transferring the RFS ledger across to the FPF platform Lending. As a general practice, staff from the two entities remained geographically separate while the transfer occurred.
[15] By early 2005, as the Judge noted, an internal debate was under way among FPF’s information technology staff. Should the transition proceed on the basis that RFS’ business rules and protocols were to be absorbed within the existing FPF rules and policies? Or should RFS’ discrete rules and policies be maintained in the new composite system? The latter view prevailed, leading to FPF replicating aspects of CMS software in the Lending platform as the project progressed during 2005. Karum’s argument in this Court that FPF breached its intellectual property rights focussed heavily on FPF’s incorporation within the Lending platform of five features of the CMS software: (1) the payment calendar, (2) the software delinquency calendar, (3) the aged debt system, (4) some special codes, and (5) intercept codes.
[16] Four people attended the settlement meeting in January 2005. On Karum’s side were Mr Johnson and Robert Scott, an experienced lawyer who had acted as the company’s legal counsel in the United States for many years. On FPF’s side were two senior Fisher & Paykel in-house lawyers: Lindsay Gillanders, an FPF director and the Fisher & Paykel Group’s legal director, and Rebecca Holbrook, Fisher & Paykel Appliances Ltd’s general counsel. Both had first become involved in negotiations about a month earlier.
[17] Mr Gillanders and Ms Holbrook brought to the meeting a document known in the High Court as the “clean room memorandum”. In this Court counsel agreed that it should be more appropriately named the Nobbs memorandum after its author, Pamela Nobbs. Karum’s allegations of misrepresentation and misleading or deceptive conduct are based upon it, and we shall return to the document and its effect.
[18] Ms Nobbs had only joined FPF on 10 December 2004 as its general manager for information services. She prepared her memorandum over the next three days at the request of either Mr Gillanders or Ms Holbrook. Its purpose was to report on the operation of the RFS credit management systems. In a telephone conference on 22 December 2004, Mr Gillanders advised Mr Scott of Ms Nobbs’ engagement to carry out a review and of his intention to give Mr Johnson a copy of her report – to enable him to understand FPF’s position and reassure him.

Settlement

[19] Rodney Hansen J summarised the progress of the settlement meeting as follows:

[56] Negotiations concluded with a meeting on 6 January 2005 in San Francisco. Mr Gillanders and Ms Holbrook met with Mr Johnson and Mr Scott in Mr Scott's office. Again, the discussion was agreed to be “without prejudice” and “off the record”. Mr Johnson and Ms Holbrook each took notes. Both sets of notes are agreed to be accurate and there are no material differences as to what was said at the meeting. After discussing the start date for the licence (eventually agreed to be 1 April 2004) and the definition of “account” for the purpose of fixing the number of active customer accounts that would be processed, the licence fee was addressed. Mr Gillanders said FPF was prepared to pay an up-front fee of US$400,000 for two years but not the proposed US$150,000 contract initiation fee. According to Mr Gillanders, as soon as he told Mr Johnson that FPF were prepared to pay US$400,000 as a licence fee, his whole demeanour changed. He became much more relaxed “and it was as if we had in one movement addressed his greatest concern”.

[57] Discussion then turned to the migration of data to the FPF system. The clean room memorandum was handed over at this point. Mr Gillanders said that Mr Johnson seemed “not the least bit concerned” about the statements made in the memorandum because he was satisfied that any issues that may arise would be dealt with by the Exhibit A review process. ...

[58] There was discussion of modifications required to the licence. It was agreed that a clause should be added in which FPF warranted that its software would not be based upon or derived from the CMS system. Mr Johnson said that was to cover the possibility that FPF could inadvertently incorporate a small amount of CMS proprietary information into its new platform during the process of data migration. He compared this to a composer remembering a few notes or a melody over dinner and incorporating them into a new song the following morning.

[20] The settlement agreement signed by the parties on 7 January 2005 comprised three interrelated instruments: a heads of agreement (the HOA), a licensing agreement (SLA 2005) and what was known as Exhibit A. We shall return more fully to those documents. But for now we note that Exhibit A provided a discrete dispute resolution mechanism, with rights of review, objection and arbitration. SLA 2005 provided a different arbitral mechanism.

Post-settlement events

[21] On 30 March 2006 FPF advised Karum that it did not need to renew SLA 2005 because it had transferred all data from CMS software to the Lending platform. In July 2006 Karum’s nominee, Paul Hart, carried out the Exhibit A review in New Zealand. He reported that FPF had incorporated components of the CMS software into its Lending program by a variety of methods. He described its scale as extensive and pervasive.
  1. FPF invoked the objection procedure provided by Exhibit A. Karum countered by issuing arbitration proceedings in San Francisco under SLA 2005. The arbitrator upheld FPF’s objection to jurisdiction. He decided that the parties had agreed to resolve their disputes under the Exhibit A procedure, which is subject to New Zealand law, and not under SLA 2005, which is governed by Californian law.[10]
[23] In the normal course, Karum would have been expected to continue with the Exhibit A dispute resolution procedure. But the company’s apparent purpose in seeking the High Court’s validation of its cancellation of the settlement agreement is to avoid that process. A favourable judgment would allow Karum to rely upon its accrued rights under SLA 2005 – in particular, the right to invoke arbitration in California. Karum accepts that it will be unable to pursue its intellectual property claims in this litigation unless the settlement agreement is validly cancelled. If the agreement remains on foot, Karum’s intellectual property claims would fall within Exhibit A’s jurisdiction.
[24] Thus the primary question is whether Rodney Hansen J was correct to dismiss Karum’s application for a declaration that it validly cancelled the settlement agreement: if so, Karum’s appeal must fail; if not, its appeal against the Judge’s dismissal of its intellectual property claims will require determination. In the event, we will decide both substantive elements of its appeal, to cover the contingency that we have erred in determining either of them.

Misrepresentation claims

[25] Karum’s alternative claims for misrepresentation and misleading and deceptive conduct were based upon identical factual allegations. In argument on appeal counsel did not differentiate between the two causes of action. We shall proceed on the premise that the result of Karum’s claim for misrepresentation will be decisive: if that claim fails, then the company cannot succeed under the Fair Trading Act 1986.
[26] Karum divided its misrepresentation claims into two categories. In the first category were so-called non-infringement representations, to the effect that:
[27] Rodney Hansen J considered this first category when determining Karum’s claims for breach of confidence and copyright; and we shall do likewise.
[28] In the second category were three so-called clean room representations, based on passages from the Nobbs memorandum:
[29] To the extent that these statements are representations as to FPF’s future intentions, we recognise that to be actionable misrepresentations must be as to facts.[11] However, we also recognise such representations can amount to misrepresentations if FPF’s intentions are not reasonably held or if the present facts on which they are based are untrue at the time the representations are made.[12]
[30] Rodney Hansen J observed:[13]

Karum developed this part of its case in [mis]representation and for breach of the Fair Trading Act on the basis that the clean room representations led Karum to believe that it had no reasonable basis to doubt FPF's noninfringement representations. Karum’s position is that, based on the clean room representations, it assessed the risk of its intellectual property being compromised during the data conversion and migration process as low. It considered that a technical review of the Lending software at the conclusion of the licence agreement in terms of Exhibit A would be adequate to address any infringement which occurred, notwithstanding the clean room protections.

[31] The Judge dismissed each allegation of misrepresentation. We shall review his findings at a later stage in this judgment.[14]

Inducement

[32] Our examination of Karum’s appeal against the Judge’s findings on the Nobbs memorandum misrepresentations does not start with the threshold question of whether representations were made or whether they were false. It more appropriately begins with the question of whether any representations, assuming they were made, induced Karum to enter into the settlement agreement and, if so, whether it is entitled to a declaration of valid cancellation.

(a) Contractual Remedies Act 1979

[33] Karum’s right to cancel the settlement agreement is codified by the Contractual Remedies Act as follows:

7 Cancellation of contract

...

(3) Subject to this Act, but without prejudice to subsection (2), a party to a contract may cancel it if—

(a) he has been induced to enter into it by a misrepresentation, whether innocent or fraudulent, made by or on behalf of another party to that contract; or

(b) a term in the contract is broken by another party to that contract; or

(c) it is clear that a term in the contract will be broken by another party to that contract.

(Emphasis added.)

[34] Karum was entitled by s 7(4) of the Act to cancel the settlement agreement if, and only if, it could establish that (1) it had been induced to enter into the contract by FPF’s misrepresentations, and (2) the parties had agreed – expressly or impliedly – that the truth of the representations was essential to Karum or that the effect would be (a) to substantially reduce the benefit of the contract to it or increase its burden under the contract; or (b) to make the benefit or burden substantially different from that which was represented. Karum must prove both elements if it is to succeed, and we shall address each in the same order.

(b) Settlement agreement

[35] The question of whether Karum was induced by FPF’s representations to enter into the settlement agreement is ultimately a question of fact. Our inquiry in that respect must be guided by the contractual framework. That is because the three relevant instruments expressly identified FPF’s warranties and representations, their status and Karum’s rights in the event of FPF’s breach. We shall discuss the relevant terms of each document.

(i) The HOA

[36] The HOA is a short document which relevantly provided:
  1. CMC Inc. (“CMS”), under the terms below and of the attached form of License Agreement, will grant to Fisher & Paykel Financial Services Limited Finance Ltd (“Client”) a nonexclusive right and license to use CMS’s proprietary software entitled Credit Management System described in the License Agreement (the “Software”), for the purpose described in the License Agreement.
  2. At the conclusion of the term of the License Agreement, the parties will follow the procedures described in Exhibit A hereto to determine whether any proprietary trade secret information of CMS has been used by Client in connection with creating the software which will succeed the CMS Software.

...

  1. The parties will enter into comprehensive mutual releases in mutually agreed form and substance in respect of the claims covered by the litigation currently pending in New Zealand, including the filing in Court of all documentation required to cease the current proceedings. CMS acknowledges and agrees that the payments referred to in the License Agreement include (a) a full and final settlement of all claims howsoever arising out of the use of the Software by any member of the Client Group prior to 1 April, 2004; and (b) a full and final settlement of all claims made in relation to any alleged infringement, or breach of any agreement in relation to, CMS proprietary trade secret information by Farmers Trading Company Limited or its related companies. Each party shall be responsible for its own legal costs associated with the litigation and settlement thereof.

...

(ii) SLA 2005

[37] SLA 2005 was apparently a standard form Karum licence but with important modifications drafted by Mr Scott to cover this particular arrangement. SLA 2005 granted FPF a non-exclusive and non-transferrable right and licence to use the source and object code of the CMS software programme, and further provided:

1.4 Notwithstanding the foregoing, CMS acknowledges that the Software is installed on Equipment owned by The Farmers Trading Company Limited (“FTC”) and located at FTC premises, and that FTC provides infrastructure and operations support to the Client Group in connection with the Client Group’s use of the Software.

...

2.1 The Software may be copied by Client for use solely at the Location and with the Equipment provided that Client may make the number of copies that are needed for proper operation and backup. The original and all copies of the Software or any portion thereof made by Client shall be and remain the property of CMS. Client shall display the copyright and proprietary notices of CMS or any other person on all copies of the Software or any portion thereof in the same form as such notices appeared in the original Software.

2.2 CMS shall own all right, title and interest in and to all intellectual property rights, including patent, trademark, copyright and trade secrets rights, related to the Software and, subject to Section 2.4, any and all services provided by CMS in connection with Client’s use of the Software. Client agrees to execute and deliver such confirmations, assignments and other documents as CMS may request to confirm such ownership.

...

2.4 CMS acknowledges and agrees that:

(i) Nothing in Section 2.3 shall apply to the extent that the Software, the systems, ideas, methods of operation, and information contained therein has become generally available to the public other than as a result of Client’s unauthorised disclosure, or which has been disclosed to Client by CMS on a non-confidential basis.

...

(iii) All original software or modifications developed by CMS under consulting at the request and expense of Farmers Trading Company Limited (“FTC”), if any, FTC, Client and/or a member of the Client Group, in each case which does not contain proprietary information of CMS and which is not derived from or related to the Software or other proprietary information originally developed by CMS shall be the sole property of Client.

...

2.10 CMS acknowledges that the Client Group has developed its own software and system for the processing of credit card and hire purchase transactions and is in the process of further developing that software and system to enable the migration of all transactions for which the Client Group requires use of the Software to the Client’s own system. Client represents and warrants that the Client Group’s own software and system is not and will not be based upon or derived from the Software; provided that the exclusive remedy for any breach of this representation is provided for in the Heads of Agreement between the parties of even date herewith.

[38] SLA 2005 commenced retrospectively on 1 April 2004 and was to continue for as long as FPF used the CMS software. A schedule incorporated FPF’s agreement to pay a once only fee (the transfer fee) of US $300,000 on or before 17 January 2004.[15] The licence fee of US $100,000 was also then payable to cover the period to 31 March 2006. FPF expected by that time to complete the transition of RFS’ information to its Lending platform.
[39] Clause 10.1 provided:

This Agreement, and the Schedules hereto, contains the entire agreement of the parties with respect to the subject matter hereof and supersedes all prior oral or written discussions, representations, understandings and agreements.

[40] By cl 10.11, the agreement was subject to the laws of California and any claim arising from it was to be settled by binding arbitration in San Francisco pursuant to American arbitration rules.

(iii) Exhibit A

[41] Exhibit A provided Karum with a right of preliminary review of FPF’s software products and product specifications relating to revolving credit card and hire purchase transactions of the type previously processed by the CMS hardware. The right was to be exercised within 90 days of termination of SLA 2005. Its purpose was to determine whether FPF had incorporated within the Lending system any of Karum’s “proprietary trade secret information”. If Karum’s nominee concluded that CMS trade secrets had been incorporated, then Exhibit A provided a carefully structured objection and arbitration procedure.
[42] The arbitrator’s decision was agreed to be “final and binding upon both parties and neither shall require confirmation nor be subject to review by the Courts”. If the arbitrator found in Karum’s favour, he or she had the sole power to direct the remedial actions FPF must take to cease using Karum’s intellectual property. Specifically, the parties also agreed that the sole and exclusive remedy for this breach was FPF’s obligation to pay usage fees under SLA 2005 until the proprietary trade secret information was removed.

(iv) Summary

[43] As Mr Miles accepted, these three instruments formed one interlinking and composite settlement agreement. SLA 2005 was the most extensive, reciting formally the parties’ acknowledgement that:
[44] Mr Scott drafted cl 2.10 while the parties were negotiating in San Francisco. Its purpose was to record FPF’s primary representation and warranty that its system was not and would not be based upon or derived from Karum’s software. However, the parties agreed upon an “exclusive remedy” if FPF breached its undertaking – the carefully structured dispute resolution mechanism provided by Exhibit A. In combination with cl 10.1, this provision was intended to address the contingency that FPF might have made the first category of misrepresentations, the non-infringement misrepresentations.
[45] Clause 10.1 assumes central importance in determining the status of the second category, the Nobbs memorandum misrepresentations. It is the parties’ acknowledgement that “this agreement” (whether restricted to just SLA 2005 as Mr Miles contended or applying to all three instruments) was the whole agreement “with respect to the subject matter hereof” – FPF’s right and licence to use Karum’s CMS software. Critically, the parties acknowledged that the instrument superseded “all prior ... written ... representations”. On Karum’s own case, the Nobbs memorandum was a prior written representation.
[46] In combination, cls 2.10 and 10.1 operated as the parties’ agreement that the only representations made by FPF which had a material effect, and for which a carefully tailored remedy was available in the event of a breach, were those relating to its use of its own software and its undertaking not to misappropriate Karum’s intellectual property. The only available inference is that the parties agreed all other written representations made by either of them, such as the Nobbs memorandum, were of no legal consequence. The corollary is that those representations did not operate as an inducement to contract and Karum did not rely on them.
[47] By this means the parties intended to record, as they expressly said, their entire agreement, avoiding the risk of collateral litigation of the very type which has now eventuated. The commercial rationale is obvious. The falsity of any representations of the type made in the Nobbs memorandum – such as whether FPF software developers had access to Karum’s system during the migration process – would not be of itself relevant unless it resulted in a breach of Karum’s intellectual property rights. And that result would become apparent when its nominee undertook the Exhibit A review.
[48] Mr Miles sought to negate the obvious potency of these provisions by reference to Mr Johnson’s evidence. Mr Johnson asserted that Mr Scott drafted cl 2.10 on his instructions because by then, having seen the Nobbs memorandum, Mr Johnson was satisfied that the possibility of FPF infringing Karum’s intellectual property rights was remote. By reference to a musical analogy, Mr Johnson described cl 2.10 as being drafted to cover “stray notes” only.
[49] The short answer is that our inquiry requires an objective construction of the contract. Mr Johnson’s subjective evidence of his intention is irrelevant.[16] Clause 2.10 speaks for itself to the contrary.

(v) Other evidence

[50] We add also that, even if we had not found that the relevant provisions of SLA 2005 were decisive against Karum’s claim of inducement, we are satisfied that other evidence supports Rodney Hansen J’s conclusion that Karum was not in fact induced by the Nobbs memorandum to enter into the settlement agreement. In summary the Judge found that:
[51] In support of Karum’s appeal Mr Miles argued that the Nobbs memorandum representations were material and must have operated as an inducing factor given that:
[52] We accept Mr Miles’ submission that FPF intended the contents of the Nobbs memorandum to influence Mr Johnson’s decision to settle the claim. In legal terms, objectively speaking, it was provided in the expectation that it would be operative. In this respect Mr Miles’ reliance on what he described as the discomfort displayed by both Mr Gillanders and Ms Holbrook under cross-examination about the extent to which they intended Karum to rely on the Nobbs memorandum is misplaced. On the other side of the ledger, however, he is correct that evidence given by those two witnesses about Mr Johnson’s conduct at the meeting, upon which the Judge placed some reliance, is also irrelevant. We repeat that our inquiry is of an objective nature. With respect, the subjective views or impressions of the participants, which the Judge gave some weight, are of no consequence.[24]
[53] However, our conclusion that FPF intended that Karum would rely on the Nobbs memorandum is not the end of the inquiry. The decisive question is whether the memorandum did in fact operate as a materially contributing factor in Karum’s decision. Rodney Hansen J’s primary findings were that from the time he learned of FPF’s acquisition of RFS in mid 2004 Mr Johnson believed that FPF was infringing and would continue to infringe Karum’s intellectual property rights; and that he never wavered from that belief. Any assurances given by FPF to the contrary were immaterial and disbelieved.
[54] Having reviewed for ourselves the relevant evidence, we are satisfied that the Judge’s factual finding is well founded. Mr Johnson’s correspondence, and that emanating from Mr Scott, is consistent with his conclusion that by early December 2004, well before preparation and provision of the Nobbs memorandum, Karum was set upon the course towards settlement which it later implemented. This judgment would be extended unnecessarily if we were to recite extracts from that correspondence. We simply record that Mr Johnson’s letters to FPF satisfy us that by 22 December 2004 Karum had resolved to settle the dispute in principle subject only to agreeing on the amount of the licence fee, the starting date of a licensing agreement and the terms of the first stage of the Exhibit A process.
[55] Moreover, as Mr Miles accepted, the Nobbs memorandum did not convey anything new to Mr Johnson; it simply assembled pre-existing information into one document. Each of its representations had been made by different FPF personnel to Mr Johnson at various times since June 2004. His state of disbelief remained constant throughout. Indeed, before us, Mr Miles acknowledged Mr Johnson’s scepticism: there is nothing to suggest that it changed upon receipt of the Nobbs memorandum during the course of the settlement meeting. We are satisfied that the memorandum had no effect upon Karum’s decision to settle at all. The evidence is overwhelming in its reliance on other factors.

(vi) Statutory discretion

[56] Clause 10.1 engages s 4 of the Contractual Remedies Act which states as follows:

4 Statements during negotiations for a contract

(1) If a contract, or any other document, contains a provision purporting to preclude a Court from inquiring into or determining the question—

(a) whether a statement, promise, or undertaking was made or given, either in words or by conduct, in connection with or in the course of negotiations leading to the making of the contract; or

(b) whether, if it was so made or given, it constituted a representation or a term of the contract; or

(c) whether, if it was a representation, it was relied on—

the Court shall not, in any proceedings in relation to the contract, be precluded by that provision from inquiring into and determining any such question unless the Court considers that it is fair and reasonable that the provision should be conclusive between the parties, having regard to all the circumstances of the case, including the subject-matter and value of the transaction, the respective bargaining strengths of the parties, and the question whether any party was represented or advised by a solicitor at the time of the negotiations or at any other relevant time.

...

[57] The question then is whether it is fair and reasonable that cl 10.1 should be conclusive in precluding the Court from inquiring into whether Karum relied on the Nobbs memorandum representations. Before us both counsel accepted Rodney Hansen J’s adoption of this Court’s description of the purpose of an entire agreement clause[25] in PAE (New Zealand) Limited v Brosnahan as follows:[26]

[15] An entire agreement clause, however, is not absolute or conclusive. Section 4(1) recognises a wide judicial discretion to determine whether it is “fair and reasonable that the provision should be conclusive”. While the issue is to be determined “having regard to all the circumstances of the case”, the specified criteria focus the inquiry on an assessment of the relative positions of the parties and their access to independent legal advice. Its apparent purpose is to protect one party’s relative vulnerability from another party’s power to impose an exemption from liability which is contrary to the factual reality or an existing legal obligation and is thus unreasonable and unfair. Section 4(1) is a mechanism for striking balances, both individually between parties and conceptually between freedom of contract and unfair or unreasonable commercial conduct. (See also Dawson and McLauchlan The Contractual Remedies Act 1979 (1981) at 36-40.)

[58] Rodney Hansen J found:

[127] The party relying on the entire agreement clause has the initial evidential burden to prove that it is fair and reasonable to accept the terms of the exclusion clause as conclusive.[[27]] In my view FPF has discharged that burden. This was an arm’s length commercial negotiation between two well resourced parties. Karum had the benefit of Mr Scott's involvement throughout the negotiation. He is experienced counsel who had been Mr Johnson’s legal adviser for many years. Karum clearly had the superior bargaining position. There is little doubt that RFS had no right to continue using the CMS software following its separation from Farmers retailing activities. Karum had strong grounds for injunctive relief, which would have had disastrous consequences for FPF. I am satisfied that cl 10.1 should take effect according to its tenor. Karum could not have relied on prior representations to cancel the licence.

[59] Mr Miles did not attempt to challenge Rodney Hansen J’s reasoning, which we endorse. Instead, he challenged the Judge’s conclusion on a different ground. He submitted that the application of cl 10.1 is limited to SLA 2005 alone, with the result that it does not extend to the HOA and Exhibit A. Mr Miles’ argument was based upon the words “this agreement” where they appear in cl 10.1 which, he said, could only refer to SLA 2005. By contrast, cl 2.10 draws a distinction with “the Heads of Agreement” (which appears to include Exhibit A as well). The parties’ obvious intention was, he said, to ensure a temporal separation between the currency of SLA 2005 and the Exhibit A process; the former is subject to the laws of California and the latter to New Zealand law.
[60] We reject this submission. Its subtlety belies the commercial context and contradicts Mr Miles’ acceptance of the interlinkage between the three instruments as constituting one composite settlement agreement. It is plain that the entire agreement between the parties was not just SLA 2005 but all three contracts. In this context the phrase “this agreement” must be construed consistently with the contractual framework as a whole. Clause 2.10 of SLA 2005, which is the governing provision in this respect, expressly refers to the exclusive remedy available to Karum if FPF breached its obligations. It is artificial to seek to isolate the instruments in the way attempted by Mr Miles.
[61] Mr Miles submitted also that Karum relied entirely on FPF to accurately inform it of the use it was making of the CMS software and the protections in place to prevent infringement; and accordingly it would be unjust to permit FPF to take advantage of its own wrong unless it could clearly establish the parties’ agreement to exclude all representations from their bargain.
[62] However, the parties did not agree to exclude all representations from their bargain. They expressly agreed to include the representations which they regarded as material, principally FPF’s assurances about its past and intended use of the CMS programme. Karum drafted its own provision to protect itself against any adverse consequences of reliance. And the parties unequivocally agreed to exclude all other representations from their bargain.
[63] We add that neither counsel argued that the parties’ agreement that SLA 2005 was to be construed and enforced in accordance with the laws of California operated to exclude our consideration of its terms in the context of deciding whether Karum was induced by FPF’s representations to enter into it and the two related instruments.

Whether representations were essential or substantially reduced benefit

[64] Even if it proved inducement, Karum would have to satisfy us on the second element of s 7(4) of the Contractual Remedies Act: either that the truth of the Nobbs memorandum representations was essential to it or that they substantially reduced the benefit of the settlement agreement to it. Karum pleaded that the parties agreed expressly or by implication that the truth of those representations was essential to it. But the four particulars given in support are simply variations of its expressed concerns about FPF’s past and prospective breaches of its intellectual property rights. This unilateral emphasis does not advance Karum’s case where s 7(4) expressly requires mutuality.
[65] Instead, in argument Mr Miles emphasised Karum’s alternative pleading that the alleged misrepresentations made the benefit or burden of the settlement agreement substantially different from that which was represented or contracted for. In particular, Karum asserted that under the agreement it forewent its right to injunctive relief which it enjoyed in the existing High Court litigation and settled for the lesser remedy of a right to damages. Karum also asserted that it accepted a concessionary licence fee.
[66] Rodney Hansen J did not make a finding on this second element of substantial difference, presumably because he was satisfied that Karum’s claim failed at the inducement hurdle. Mr Galbraith QC submitted that Karum’s decision to forego its right to injunctive relief was valueless, and did not constitute such a qualifying change in benefit or detriment as to justify cancellation. He pointed out that Karum is in the business of licensing its software, it agreed the level of the licence fee and agreed under Exhibit A to accept the fee in return for allowing FPF’s use of the system. Furthermore, there was no evidence that the licence fee was concessionary.
[67] We agree with Mr Galbraith. In this respect we note Rodney Hansen J’s finding that the licence and transfer fees were windfalls for Karum. That is because the company would not have been entitled to any further payment if Farmers had not sold RFS or RFS had ceased using the system on sale.[28] We agree with him that in this sense Karum obtained a substantial benefit.
[68] Karum’s asserted detriment – the loss to a right to injunctive relief – is problematic. While at the time the company may have considered its rights to an interim injunction were strong, giving it the upper hand in any negotiations, an objective analysis suggests otherwise. FPF may well have been able to resist an application with an undertaking as to damages and appropriate undertakings about its use of the CMS system.
[69] In any event, we give little weight to Karum’s assertion that it suffered a detriment in this respect. A part of the company’s bargain for which it received substantial payments was retention of a right to what were effectively damages in the form of a usage fee if FPF was found to be in breach under the Exhibit A process. Karum could have stipulated if it had wished for the additional remedy of a prohibitory direction akin to an injunction.

Nobbs memorandum

[70] In the event that we are wrong in concluding that Karum has failed to establish both statutory prerequisites for an order for cancellation, we will consider its three specific allegations of misrepresentation made in the Nobbs memorandum.
[71] Rodney Hansen J addressed these allegations comprehensively in his judgment.[29] At trial Karum had embarked upon detailed analyses of what certain senior employees were said to have known or done. All this was based upon what was apparently a minute review of documents discovered by FPF in the course of this litigation. After a careful survey of the particular allegations,[30] Rodney Hansen J concluded that:

[113] I consider the clean room memorandum to have been an honest and fair attempt to summarise the situation at the time it was written. It emerged largely unscathed from a searching line-by-line examination undertaken with the advantage of hindsight and a great deal of information which would not have been known to its authors. It largely succeeded in conveying the essential features of a complex and dynamic process that was really only just getting underway.

[72] We do not propose to address Karum’s allegations in the same degree of detail as Rodney Hansen J. That is because his findings reflect a commendably careful analysis of all the relevant evidence. Karum must satisfy us that the Judge was wrong in a material respect. For reasons which we shall explain shortly, it has failed to do so.
[73] In argument before us Mr Kennedy helpfully combined Karum’s specific allegations of the Nobbs memorandum misrepresentations into a collective proposition that FPF was independently developing its own revolving credit platform which had not and could not incorporate any of the CMS software. In formulating the case in this way, Mr Kennedy may have inadvertently confirmed what we have already found. The effect of the Nobbs memorandum representations was no more than ancillary to or part of FPF’s representations or warranties expressly recited in cl 2.10 of SLA 2005. Mr Kennedy’s proposition serves to illustrate that the essence of the Nobbs memorandum representations was subsumed within the contractual structure, which, as we have discussed, provided an exclusive remedy in the event of breach.
[74] Nevertheless, we shall address each of the particulars of the misrepresentation which Mr Kennedy identified as: (1) IT support staff for RFS and FPF were technically, geographically and culturally separate; (2) systems and document access to CMS was restricted to previous RFS staff; (3) only two RFS team members – Ms Horvat and Ms Mason – had accepted roles relating to Lending and of such a limited nature that there was little or no risk of infringing Karum’s intellectual property rights; and (4) FPF was undertaking a data migration only rather than a systems integration.
[75] As to the first of these four particulars, Mr Kennedy identified a number of documents to the effect that, while the two teams may have been separated when the representations were made, the RFS team was soon to be relocated to FPF. These documents confirm that relocation was under consideration before 7 January 2005. But no firm decision had been made by that date. The Nobbs memorandum did not misrepresent the situation when stating that:

“The RFS and FPF systems are separately supported with dedicated technical specialists. Support staff are located at separate sites, use different methodologies ... and have specific technical skills relating to the very different environments they support. ...”

[76] We agree with Rodney Hansen J that this statement was correct when it was made and was not rendered false by subsequent developments,[31] and with his conclusion that:

[95] I do not accept that what was said was intended to describe the state of affairs that would continue for the duration of the licence. In describing the separation of RFS and FPF systems, the memorandum is carefully expressed to describe only present arrangements. There is nothing which might have conveyed that the situation was static. And the way in which discussions developed gave no hint to FPF that Mr Johnson was interpreting the document in any other way. Mr Gillanders said that in discussions with Mr Johnson the question of whether or not separation would continue simply was not an issue. He did not seek elaboration or clarification of what was in the memorandum. I am satisfied that this part of the memorandum was true at the time it was made and was neither intended nor had the effect of misleading Karum. In my view, FPF management understood Karum’s concerns and were committed to ensuring that the two systems were operated independently until integration had been completed.

[77] As to the second particular, Mr Kennedy again identified documents which showed that a number of former RFS staff (then employed by FPF) may have had access to the CMS software. He relied also upon what he portrayed as concessions made in cross-examination. However, having reviewed this evidence we are satisfied that the Judge was correct in finding that:

[97] The statement in the clean room memorandum was in fact that none of the FPF system developers have access to CMS development, test or production systems or documentation. Except in relation to [Ms] Mason, that statement is accurate. The system developers of FPF did not have access to CMS proprietary information at the time the memorandum was written and I am satisfied that the intention of senior management was that that should continue to be the case. FPF developers relied on FTC/RFS personnel to provide the information necessary for them to assess the additional functionality required in Lending to accommodate RFS business requirements. Karum says that in the course of this process (and after the settlement was concluded in January 2005), FPF developers and business analysts were provided with CMS proprietary information. However, that is a different issue which I will consider in relation to the specific claims of breach of confidence and copyright.

(Footnote omitted.)

[78] As to the third particular, Mr Kennedy devoted considerable attention to an argument that Ms Mason’s role in FPF was more technically involved than was represented. She was described in the memorandum as being in a non-technical management role and having skills which were primarily in the business administration area. Mr Kennedy pointed to documents establishing that Ms Mason had been head of RFS’ IT development team, had a role in selecting the CMS software for use by Farmers and was appointed development manager information systems at FPF after the shift. He also noted that she was designated in FPF documents as a “subject matter expert”.
[79] Taken in isolation, some of these documents might show Ms Mason’s role was more technical than was represented. However, we agree with the Judge that the essential description of her role was accurate.[32] She was a non-technical employee whose skills lay elsewhere.
[80] As for Ms Horvat, Mr Kennedy’s submissions relied on two isolated pieces of evidence to assert that her role was misrepresented. Both the representation relating to her and this evidence were inconsequential, tending to reflect Karum’s misplaced emphasis on peripheral factors in an attempt to bolster its claim. The Judge’s observation on this allegation is apposite:

[76] The relevant section of the clean room memorandum was plainly never intended and could not have been understood to have been more than a brief summary of the essential nature of what those two employees were doing. Subject to the reference to Mrs Mason not having access to CMS documentation, I consider this part of the memorandum to have been accurate.

[81] As for the fourth particular, we have noted previously that there was as at 7 January 2005 internal debate within FPF about the nature and extent to which the CMS system should be integrated into the Lending platform. We agree, however, with the Judge that the Nobbs memorandum statement of intention – that after business integration was completed (projected for June 2005) FPF would decommission all CMS modules currently used by RFS and customer data would be migrated to the Lending programme – correctly stated FPF’s intention. Mr Kennedy has not identified any evidence which might show that Ms Nobbs did not then have reasonable grounds to make this statement. The existence of an unresolved debate does not advance Karum’s cause. This allegation also fails.

Summary

[82] The Nobbs memorandum was just over two pages long; we are satisfied that while FPF intended that Karum should rely on its consent it purported to be nothing more than a summary of Ms Nobbs’ findings about systems usage, security, development and support methodologies within FPF relating specifically to the RFS system. Like Rodney Hansen J, we are not satisfied that the document materially misrepresented FPF’s position; instead, like the Judge, we are satisfied that it represented a snapshot in time and nothing more. Nor, for the sake of completion, are we satisfied that its statements were misleading or deceptive.
[83] In summary, Karum has failed to prove both that (1) the representations made in the Nobbs memorandum, even if false, were material and operated as an inducement to enter into the settlement agreement; and (2) either (a) the truth of the representations was essential to the company or (b) they substantially reduced the benefit of the settlement agreement to it. In our judgment Karum’s assertion that it relied on the Nobbs memorandum to assess the risk of its intellectual property being compromised as low is an after the fact rationalisation, designed to circumvent its own carefully structured contractual framework for settling its rights of recourse against FPF.
[84] It follows that our dismissal of Karum’s appeal against Rodney Hansen J’s dismissal of its application for a declaration that it validly cancelled the settlement agreement is sufficient to dispose of its entire appeal.[33] However, in case our conclusion on this part of the appeal is wrong, we shall now consider Karum’s separate appeal against the Judge’s dismissal of its intellectual property claims for infringement of copyright and breach of confidence.

Copyright

[85] A plaintiff in a copyright claim must prove that the defendant infringed copyright subsisting in a work that the plaintiff owns.[34]
[86] The plaintiff must begin by accurately identifying the work and establishing that it falls within one of the categories in s 14(1) of the Copyright Act 1994, which include, relevantly, a literary work. A literary work must have been recorded, in writing or otherwise.[35] It is the expression of an idea, rather than the idea itself, that copyright protects.[36]
[87] The plaintiff must show that the work is original,[37] meaning that it does not reproduce another work[38] and its author expended more than minimal skill and labour in its production.[39] Novelty is not the test, but the greater the degree of originality the more extensive the protection supplied by copyright.[40]
[88] Having established that the work enjoys copyright protection, the plaintiff must next show that the defendant infringed that copyright by making substantial use of the skill and labour which went into the work.[41] To decide this issue the court makes a qualitative assessment, looking for what has been described as the “essence” of the work.[42] The assessment may be influenced by the work’s degree of originality; the less original it is, the less likely that something substantial was copied. Infringement in the case of a literary work may take the form of adoption of the work’s structure or compilation of content rather than its language or form of expression.[43]
[89] The parties agree that copyright subsists in CMS at a basic code or “literal” level, that FPF did not copy anything at that level, and that FPF did copy certain features or details found at a somewhat higher level. Karum characterises the things copied as “design elements” and claims that copyright protection extends to them. FPF does not accept that Karum owns copyright in CMS so as to have a cause of action, or that Karum has identified anything capable of protection, or that copyright protection extends to the things copied, which it characterises as ideas, and further, as insubstantial and non-original.

(a) Has Karum a cause of action for copyright infringement?

[90] As we have noted,[44] Rodney Hansen J found in earlier decisions[45] that Karum does not own copyright in CMS but is the exclusive licensee, which status conferred upon it rights and remedies concurrent with those of the owner[46] such that it could pursue an infringement action by leave[47] without joining the owner.
[91] The issue arose in an unsatisfactory manner, as the Judge confirmed in a subsequent costs judgment.[48] Karum failed to anticipate, when counterclaiming for breach of copyright, that its right to sue might be in issue. When the point was taken at trial, Karum was allowed to plead that it owned CMS or held an exclusive licence, and to adduce some further evidence at a very late stage. The evidence was not entirely satisfactory. The copyright was originally held by a US firm called Zale Corp Inc. That firm later formed a partnership, through a subsidiary called ZRCM2 Inc, with Peter R Johnson & Associates Inc. The partnership was later dissolved on the bankruptcy of ZRCM2, but not before it granted (in 1988) Peter R Johnson & Associates rights to CMS.
[92] In his No 2 judgment, which was issued following a partheard application for leave to bring proceedings, Rodney Hansen J found that ownership of the software remained with Zale although Peter R Johnson & Associates had been granted the exclusive right to exploit it.[49] He also found that there were several licensees, including Peter R Johnson & Associates and Zale subsidiaries.[50]
[93] The No 3 judgment was issued after the leave hearing concluded. It focussed on whether Karum, which had acquired the exclusive right to exploit CMS that was granted to Peter R Johnson & Associates, was an exclusive licensee.[51] The Judge found that a licence might be exclusive although the owner retained the right to use the software;[52] and that the right to commercially exploit the software amounted to an exclusive licence, there being no overlap with rights retained by Zale.[53] The fact that Zale still owned the copyright was not a bar to proceedings, since it might be joined to the proceeding. In the circumstances, however, the Judge did not think it necessary to join Zale.[54] Leave to sue without Zale was accordingly granted.[55]
[94] We observe that the Judge recognised FPF might be disadvantaged because of the unsatisfactory way in which the issue had arisen and been handled.[56] He was prepared to make the grant of leave conditional upon further discovery and the recalling of witnesses at Karum’s expense,[57] but FPF elected not to take advantage of that opportunity.
[95] Karum appealed, challenging the Judge’s decision that it does not own CMS. It contended that on the true interpretation of the relevant agreements it owns the copyright in CMS, having acquired it by assignment, and that FPF is estopped under SLA 2005 from denying Karum’s copyright.
[96] FPF cross-appealed, challenging the Judge’s decision that Karum is an exclusive licensee. Its short point was that an exclusive licensee for purposes of the Act is a person holding an exclusive written licence of the relevant copyright and Karum’s licence does not qualify; it licenses not the copyright but the product (CMS), conferring only the right to use and commercialise the software.
[97] On the view we take of the case it is not necessary to address these issues, and we decline to do so. The result does not depend upon them. They do raise questions about divisibility of copyright and what the Act means by exclusive licensee, but as the summary above indicates there are good reasons not to decide those questions here. The Judge’s approach managed an issue which might otherwise have derailed the hearing and allowed him to decide the substantive issues dividing the parties. The evidence on the issues was and is unsatisfactory; Karum says that there are other documents which demonstrate that it does own the copyright in CMS. FPF was offered the opportunity to have further discovery and witnesses recalled at Karum’s expense but it did not take up the invitation, electing instead to proceed on the basis that Karum was an exclusive licensee.

(b) Copyright in computer programs

[98] One of Karum’s witnesses, Dr A J Nichols, supplied a general definition of a computer program:

[C]omputers are machines that execute instructions in sequence. Each type of computer has a specific set of instructions that it can execute. These instructions are quite simple and generally fall into arithmetic operations, logical operations, tests, jumps to an instruction other than the next one, data exchanges with memory, or data exchanges with external devices. Each instruction is specified with a unique numeric code. To get a computer to do something useful, it is necessary to create an algorithm and then to determine the specific instructions that will cause the computer to perform that algorithm. This is called a program.

[99] Counsel also referred us to authorities which contain descriptions or definitions. The authorities appear to deal with conventional imperative programs.[58] We understand COBOL, the language in which CMS is written, to be of an imperative type. Although some of the authorities are now dated, it was not suggested that they are inapplicable here. Notably, counsel relied on Computer Associates International Inc v Altai Inc.[59] We observe that caution is needed when dealing with authorities from the United States, which diverge from Commonwealth authorities in some respects.[60] But we are dealing here with the nature of a computer program, and in particular the distinction between literal and structural elements. In these respects English and United States authorities are both germane when dealing with copyright claims.[61]
[100] Consistent with the authorities and the evidence, the following sequence summarises how a program such as CMS is created:[62]
[101] The literal elements of a computer program are normally taken to comprise the source and object code. Copyright protection undoubtedly extends to source and object code.[64] It is through code that the “selection, ordering, combination and arrangement of instructions” within a program find particular expression.[65]
[102] The question is whether copyright protection extends beyond literal elements to the program’s structure.[66] The term “structure” refers to the programmer’s choice of subroutines and relationships and prescribed interactions between them.[67]
[103] It is not in dispute that most of the programmer’s skill and labour lies in the structure, through which the programmer seeks to achieve speed, efficiency and simplicity within external constraints which may be imposed by hardware or regulatory requirements or business rules. As the High Court of Australia put it in Data Access Corp v Powerflex Services Pty Ltd:[68]

It is the ability to express in a computer language an algorithmic or logical relationship between an identifiable function ... to be performed and the physical capability of the computer, which is the true skill of the programmer. This remains true even if the programmer is working via the medium of a high level language ...

[104] In Altai, the Court accepted that structural elements, although non-literal, are in principle capable of copyright protection. It adopted the following syllogism: copyright in a literary work extends to its non-literal components; computer programs are literary works; therefore copyright in computer programs extends to their non-literal elements.[69] A similar view has been taken in the United Kingdom; in SAS Institute Inc v World Programming Ltd Arnold J accepted in the High Court that copyright may extend to a program’s “structure, sequence and organisation”;[70] and the Court of Appeal held that protection is conceivable from the point at which the selection and compilation of a program’s elements indicate the author’s creativity and skill.[71]
[105] But this is merely to admit the possibility of copyright protection for nonliteral elements of a computer program. Counsel drew our attention to only one case in which protection has actually been extended so far. That case, Whelan Associates Inc v Jaslow Dental Laboratory Inc, appears to be the source of the terms “structure”, “sequence” and “organisation”, which the Third Circuit Court of Appeals used interchangeably to mean the programmer’s choice of a particular arrangement of modules and subroutines.[72] In issue was a dental laboratory management program which the defendant was said to have appropriated. The Court reasoned that the idea of the program was its purpose or function – in that case, the efficient management of a dental laboratory – and everything else was expression. It did not accept that the organisation and structure of the program were inseparable from the underlying idea. It also observed that the work of turning the business processes of a dental laboratory into the modules and subroutines of a computer program required a good deal of skill; more, indeed, than the actual coding.
[106] In Altai, which appears to be the leading US decision, the Court noted that Whelan had met with a mixed reception in the courts and much criticism in the academic community.[73] The essence of the criticism is that the Court in Whelan assumed that only one idea underlies any given computer program and all else is expression, but that assumption is founded on a misunderstanding; in fact, every subroutine is its own program, so may be said to have its own idea.[74]
[107] Courts in other jurisdictions have refused to extend copyright protection to structural elements for a series of reasons. First, copyright protects the skill, judgment and labour employed when devising the form of expression of a literary work.[75] Because a program is a literary work, protection is denied to ideas having nothing to do with its literary, dramatic, musical or artistic nature, and to ideas which are commonplace or unoriginal.[76]
[108] Second, the structure of a traditional literary work – such as the plot of a novel – may be protected, but a computer program is unlike other literary works. It has no plot, merely a pre-defined set of operations intended to achieve a desired result or functionality. That functionality may be replicated precisely by a completely different program and cannot be protected by copyright.[77]
[109] Third, it is not always easy to distinguish the idea or a program from its expression; at higher levels of abstraction there is no distinction,[78] and at lower levels the task is complicated because programs are essentially utilitarian or functional in nature.[79]
[110] Fourth, elements dictated by function or efficiency, or by external factors must be excluded.[80] Those elements may include, on the evidence in this case, mechanical specifications of the host computer, compatibility requirements, design standards, industry demands and standard programming practices.[81] “Procedures, methods of operation or mathematical concepts”,[82] and elements drawn from the public domain, are also excluded.
[111] Fifth, a program may implement “business rules”, meaning the policies and procedures of the client business which is using the program, but business rules are not protected; they are not original, from the programmer’s perspective, and they fall into the category of functions or ideas.[83]
[112] Sixth, what remains must be substantial, in a copyright sense. A single subroutine may or may not be substantial, depending on the skill and labour involved in it.[84] But general ideas having little to do with the programmer’s skill are not substantial.[85] For this reason protection is denied to items which it is strictly necessary to reproduce, such as names of procedures or fixed text elements in log files.[86]
[113] Finally, for all of these reasons a plaintiff may not find it easy to prove precisely what, if any, non-literal elements of the program merit protection.[87]
[114] That aligns with the European Union Software Directive, which provides that the expression of a computer program is protected but ideas and principles underlying any element of a program are not.[88] Mr Kennedy urged us to read English and European cases with care for this reason and because the European standard requires not mere originality but creativity. But as Jacob LJ said in Nova Productions Ltd v Mazooma Games Ltd, the directive merely makes clear that the universally recognised dichotomy between idea and expression applies to software.[89] We return to the English authorities later, when discussing infringement.[90]
[115] We note here that Karum relied heavily on SAS Institute v The World Programming Ltd, in which Arnold J accepted, as noted above, that copyright may extend to the design or structure of a program. We observe, however, that the plaintiff failed on the facts to identify anything structural that was protected and had been copied. It claimed that the defendant infringed copyright in SAS’ software by creating a program, called the World Programming System (WPS), which enabled users to run other programs written in the SAS language. The defendant developed WPS without access to the SAS software source code, but it did use software manuals that documented functionality. Arnold J held that the defendant was entitled to use the manuals to ascertain functionality, and the Court of Appeal agreed.
[116] The rules under which the international authorities were decided vary, but the authorities are notable for their consistency of approach. We consider that the same approach is appropriate in New Zealand. The authorities lead us to conclude that while copyright may extend to non-literal elements of a computer program, the scope for protection of such elements is constrained; a plaintiff may not find it easy to distinguish expression from idea, or to establish originality or substantiality, or to prove that the specific elements concerned are not mere functionality or business rules or matters of external necessity.

(c) The issues in this case

[117] We have already outlined the general issue; whether copyright subsists in the works allegedly copied, and whether it was infringed. We record at this point that certain things are not in issue. First, a copyright work must originate from its author, but there is not now any dispute about authorship; not all of the many authors of every relevant version of CMS were identified by name, but Rodney Hansen J held that it was enough that the original authors had been identified as CMS employees holding specified job descriptions that called for skill and labour, and that conclusion is not contested on appeal. Second, FPF does not contest the Judge’s finding that CMS had been published so as to qualify for protection in New Zealand.[91] Third, insofar as FPF concedes that it copied certain features or details from CMS, objective similarity and derivation are obviously not in issue.

(d) For what does Karum now claim protection?

[118] Karum claimed at trial that FPF had infringed intellectual property rights in seven CMS components. Five are the subject of the appeal: the payment calendar, delinquency calendar, aged debt module, special codes and intercept codes.[92] FPF is said to have infringed copyright in the first three, and to have misused confidential information about all five.

(e) The copyright claim

[119] We will deal with the copyright claim first. Karum pleaded that FPF copied the “logic and functionality” in the CMS payment calendar, delinquency calendar and aged debt components. Aged debt is a subset of the delinquency calendar, used to determine the value shown in the calendar but not visible to the calendar’s user.
[120] At trial the term “processing logic” was used, and much effort was expended on the process by which elements in CMS found their way into Lending; Karum characterised this as evidence of infringement, and FPF as evidence of customer performance data and credit policies, some of which predated CMS or had been added by FPF’s pre-CMS program (called F Credit), being migrated into Lending.
[121] On appeal, Karum alleges not that FPF copied logic and functionality but that the components were “design elements” or “design components” or simply “design” expressing their authors’ creativity, to be distinguished from the business rules that they implemented. To a substantial extent its case still focuses on the process by which FPF migrated information and policies from one system to another; it claims that FPF dissected and disassembled the design of these elements in CMS before replicating it in Lending, having recognised the sophistication of the design and decided to save the time and expense of replicating business functionality in an entirely new system. Of course that process does not in itself supply the answer to the question whether anything taken was protected by copyright. Karum accepts that under its licence FPF could study the functionality of CMS with a view to replicating it in a new program.
[122] It is fair to say, as Mr Galbraith emphatically did, that the claim has exhibited a protean quality as Karum sought to find some defensible way of securing protection for non-literal elements of the CMS system. Karum concedes that it has moved ground somewhat; it maintains still that “logic” is protectable but admits that that term is a synonym for “business rules”. It now characterises “design elements” as a better description of the non-literal elements copied by FPF.
[123] Little is gained, in our opinion, by adding another to the abstract nouns that already afflict the subject of copyright in computer programs. Generally speaking, what Karum seeks to protect falls into the same category as that described in the authorities we have discussed and by the trial Judge in this case;[93] it is the programmer’s choice of structure.
[124] Of course it does not follow that the claim must fail. Structural elements are in principle capable of protection, and Karum has identified specific elements or details which it says are protected. It seeks to protect, for example, specific values which users employ as codes to instruct CMS to perform some function or another. We must examine the detail, looking for anything substantial that has been copied.
[125] As noted, the copyright claim is now confined to three components; the payment calendar, delinquency calendar and aged debts module. We will deal with each in turn, discussing the evidence and the Judge’s findings. We will then answer the question whether FPF infringed copyright in CMS.

(f) Payment calendar

[126] The Judge described the payment calendar as follows:

[202] The payment calendar in the CMS system records the payment history of an individual customer. It tracks a customer’s payment history for the preceding 24 months by reference to the proportion of a scheduled payment actually made by the customer. Payment classifications include full payment of the amount due, partial (more than half but not all of what was due), less than half of the scheduled payment and no payment at all. These payment conditions are calculated by the CMS software from the raw customer data in the database and by the application of the applicable product data tag to which each payment is coded. The codes are a letter “0” for no payment, “L” for a payment of less than half, “P” for a payment of more than half and “F” for the full scheduled payment. By this means the payment history of a customer can be shown over a 24month period. Business users apply the codes to interpret what is displayed on a screen.

[127] The values shown on the screen and their meaning are recorded in the table below. It is Karum’s case that FPF copied the code values and the methodology of determining when a given code would be applied, and the sequence in which values were created. It claims the fields were designed to present a customer services representative with the information needed to deal with a customer, and no more. It refers to the CMS “Automated Customer Service System External Specification”:

The Payment Summary Calendar indicates how the account was paid month by month, in reverse chronological order beginning with the most recent month. Twenty-four months of account history are listed, including inactive months. The codes are interpreted as follows:

0 No payment

  1. (Less) Payment received that is less than 1/2 of one scheduled payment.
  1. (Partial) Payment received that is less than one scheduled payment but greater than 1/2 a scheduled payment.
  2. (Full) One full scheduled payment received.
  3. (More) Payment received that is greater than one scheduled payment but less than past due amount.
  4. (Delinquent) Payment received that is greater than or equal to the past due amount but less than the total due.
  5. (Total) Payment received that is greater than or equal to the total due.
[128] The Judge found that FPF used identical values to those employed in the CMS payment calendar when developing RFS. However, he accepted that this behaviour followed from FPF’s decisions that existing business rules should continue and users should see the same “contract performance indicators” after migration. He made the following findings, which apply to all of the elements of CMS that are in issue, about the process that FPF followed:

[206] Mr Philip McDiarmid, a business analyst contracted to FPF from December 2003, was tasked by Paul McCarthy, the Team Leader, to undertake five streams of work in February 2005, including the performance indicators project of which the development of the payment and delinquency calendars (and credit rating fields) were a part. Mr McDiarmid explained that the migration of the payment and delinquency calendars and credit rating fields required an investigation of the business rules of RFS. The objective was to ensure that the same business rules applied following integration. Among the documents he examined was a summary of how contractual delinquency, payment summary, credit rating and credit limits were determined in RFS. On the basis of his enquiries, he prepared and, on 3 October 2005, circulated a first draft of a Business Requirements Definition (BRD) report. It was headed “Integration Project: Establishment of RFS Contract Performance Measures in the Lending System”. The document concluded with a series of questions including the following:

  1. What aspects of this, if any, could be regarded as CMS IP? For example the CD Cal and Payment Cal, and the deriving of credit ratings and limits from these” Sheila advises these measure preceded CMS, and so are not JP.
  2. Business rules for calculating CD Cal – will need code inspection to determine.
  3. There are two delinquency calendar fields referred to during credit limit calculation, i.e. CD Cal i.e. and Memo CD Cal. Will need code inspection to determine if both are needed, and why.

What aspects of this, if any, could be regarded as CMS IP? For example, the CD Cal and Payment Cal and the deriving of credit ratings and limits from these? Sheila advises these measures preceded CMS, and so are not IP.

[207] Mr McDiarmid said he posed the first question because he and others involved in the integration programme had been advised during 2005 that concerns had been raised by Karum in respect to intellectual property in the FTC-CMS software. They were told to ensure that in the course of the integration process there was to be no involvement with the system itself, that is, he said, “with the computer codes”. Mr McDiarmid confirmed, as the draft report records, that he was told by Mrs Mason that the performance measures in question were in use at RFS before the FTCCMS software came into operation. With this reassurance, he continued to ascertain the rules behind the performance measures. He also saw Mrs Mason’s response as consistent with a distinction which could fairly be made between the business logic and rules relating to credit risk and the actual software implementing it.

[208] Mr McDiarmid said he recalled being told by Ms Horvat that there were some tolerances in play in the delinquency calendar that needed to be factored in. Tolerance values determine how a customer’s arrears should be aged. For example, a customer with a high rating may not be classified as being more than one month overdue if their arrears are less than $9 (being the tolerance amount). Mr McDiarmid searched for documentation setting out RFS rules on tolerances. He was unsuccessful. Nor could he find anyone who knew the RFS tolerances from memory. This led Mr McDiarmid to comment in his report that “actual values [of tolerances] to be determined from CMS code”. The further questions (3 and 4) asked in the BRD report arose because of a lack of documentation in other areas of the business logic behind RFS credit policy.

[209] Mrs Mason said in evidence that when she saw that Mr McDiarmid had asked for information to be obtained by code inspection “my heart dropped”. She said that ethically she would not allow inspection. Mr McDiarmid said, however, that he was not suggesting by the questions that he would check the code himself, rather, that he would ask an RFS IT team member to ascertain the business rules and values from the code.

[210] While the BRD report was being finalised, Mr McDiarmid wrote a functional specification. It contained all the information required by the programmer who was to write the source code for Lending. It set out, for example, the values and meanings for the payment calendar and delinquency calendar. There is no reference to CMS source code. However, it may be inferred that, to the extent that values and business rules could not be ascertained from documents, they were derived from CMS source code, albeit by RFS personnel entitled to access the code.

[211] Mr Malcolm Jenkins, the manager of FPF’s Information Service department, confirmed that because the business requirement was to retain the RFS business rules, FPF had to understand and document what those rules were. He said that instances where the FTC/CMS codes appears to have been referred to were for the purpose of identifying a particular value, such as a tolerance, which none of the business users could recall. He said that is because the tolerances will have been determined by the business earlier and then embedded in the software program. He went on to say, however, that he was advised, particularly by Mrs Mason, that the payment calendar, contractual delinquency calendar and credit rating used business rules of RFS of longstanding pedigree that in fact predated CMS. Accordingly, no CMS intellectual property was in issue. He said that, in any event, RFS business logic was implemented in Lending by developers working under his supervision without access to FTC-CMS codes.

[212] Mr Jenkins acknowledged that on conversion to Lending, the payment calendar became available for use by screen users just as it had been available to RFS staff previously. However, he saw it as simply customer data in a summarised form.

[129] The Judge concluded that the functional specifications for the payment and delinquency calendars in Lending incorporated code values and meaning used in CMS, and that in some instances information was obtained through access to CMS source code.[94] But he was satisfied that the information obtained in that way was limited to the business rules of F Credit.
[130] On appeal Karum challenged these findings, arguing that the Judge had understated the extent to which CMS code and structure were taken during FPF’s “Performance Indicators” project and the originality and commercial value of what was taken. As to the first of these, Mr Kennedy emphasised that FPF wanted to retain the same codes so that customer services representatives need not be retrained, and that FPF faced time and cost constraints which led it, as Mr Jenkins conceded, to migrate “virtually all” the CMS “attributes” to Lending, so that the payment and delinquency calendars were replicated in their entirety, along with the underlying aged debt component.
[131] Mr Kennedy drew attention to the Business Requirements document referred to above, which specified that the Performance Indicators project aimed to “ensure continuity of RFS contract performance measures by carrying over [Farmers]-style Aged Debt, Payment Summary Calendars, Contractual Delinquency Calendars and Credit Ratings”. To this end, staff responsible for developing Lending liaised closely with employees who had access to CMS source code, asking them technical questions which were answered by reference to the code. Staff developed a parallel testing procedure for CMS and Lending which was designed to ensure that Lending generated the same outcomes. All of this was said to represent a “careful dissection and disassembly of the design” of the CMS components.
[132] As to the significance of what was taken, counsel emphasised the evidence of Kenneth Maliga, a very experienced consultant and expert witness for Karum, who stated that the use of a “textually represented delinquency calendar to display the payment history” was unique and very powerful. Mr Maliga considered that the payment calendar was a “great idea” and a major selling point; indeed, it had a “wow” factor.
[133] FPF admits that it studied the functionality of CMS to ensure that it would continue in Lending. FPF identified the outputs that CMS produced when applied to its business data and ensured those outputs were replicated. That process involved checking what rules applied within CMS to determine application of the various codes. FPF argued that while CMS source code may have been examined during this exercise, contrary to best practice, nothing turns on it, for none of the code was copied and it was checked only to examine tolerances at play which pre-dated CMS; further, questions asked by the Lending developers were not answered by reference to the code, which was, in any event, meaningless given that Lending used a different language and structure. So far as the CMS payment calendar is concerned, FPF argued that there is nothing original about displaying payment history in a 24character string; that is mere functionality, and insubstantial at that.
[134] The evidence for Karum was that the payment and delinquency calendars in CMS and Lending are identical. One of its witnesses, Dale Williams, who was one of the authors of CMS and remained a Karum employee, produced the following, undisputed, table:

2014_38900.png

[135] Manifestly the codes and their meanings are the same; for example, L signifies that the debtor made less than half the payment due, and P that the debtor made something more than half but less than full payment.
[136] The Judge reasoned that payment calendars which present data in a useful manner are a common feature of credit management systems, as is the manner in which payments are analysed.[95] He accepted expert evidence for FPF that the meanings ascribed to the codes represent all possible logical outcomes of whether a satisfactory payment had been made, and evidently accepted that this was mere functionality. He also found that the particular code values in CMS required no significant skill and judgment, and further that RFS had used them for credit authorisation rules written into CMS in 2001.[96]

(g) Delinquency calendar

[137] The Judge described the CMS delinquency calendar as follows:

[203] The contractual delinquency calendar also analyses a customer’s payment history for the preceding 24 months. It does so by analysing payments by reference to arrears. A numerical code is assigned to each month according to the length of time the payment is overdue. A payment which is up to 30 days past due (i.e. one payment has been missed) is assigned the code “1”, a payment overdue between 31 and 60 days is “2” and so on. There are also codes which indicate other information about the status of the account such as a credit balance or the existence of a dispute. As with the payment calendar, a business user familiar with the code can tell at a glance the payment record of a customer and status of the account over the preceding 24 months.

[138] The following table shows the delinquency codes used in CMS:

The customer’s Contractual Delinquency Calendar contains twenty-four months of account history. Each digit represents a month, listed in reverse chronological order. The very first digit is the Memo Contractual Delinquency field, indicating the present state of delinquency.

Contractual delinquency is calculated as the number of payments billed minus the number of customer payments. Each number in the calender represents how many payments the customer was past due for that particular month. To change the past delinquency of an account, type the desired history line in this field.
* Current
1 0 – 30 days past due
2 31 – 60 days past due
3 61 – 90 days past due

4 91 – 120 days past due

5 121 – 150 days past due

6 151 – 180 days past due
7 181 – 210 days past due
8 211 – 240+ days past due
9 Profit and Loss Account
- No activity this cycle (zero balance last cycle and this cycle)
C Credit balance on account at billing
D Dispute caused by one of the following codes during billing:
01 – Complaint
12 – Repossession
Q Current due to Quick Cure rule (client option)
R Reactivated from a Profit and Loss status
W Zero balance due to small balance Write Off function

[139] As with the payment calendar, Karum claimed that FPF had copied the CMS code values, the method which determines when a given code is applied, and the sequence in which values are created. Mr Williams depicted the similarities as follows, emphasising that the delinquency calendar is displayed to a user onscreen in the same manner in Lending as it was in CMS:

2014_38901.png

[140] FPF rejects the inference that it copied these values from CMS. Mr Galbraith conveniently summarised it, the delinquency calendar in Lending is a 24 character string made up of the values *, 1–9, -, C, R, and W, representing the last 24 months of payment history, with the most recent month first. Values 1–8 represent whether the customer is one month etc past due with these values derived from contractual aging. Not all of these values are attributable to CMS. Notably, F Credit had used numeric codes 1 to 3 to represent contractual aging, and 4 signified that the account had been referred to a collection agency. Four codes were mapped from F Credit to CMS to reflect F Credit rules, although different codes were then applied.[97] The codes R (reactivated after having been written off) and W (write off small balance to zero) are the only codes in the CMS calendar used in Lending which did not have equivalent codes in F Credit.
[141] Further, FPF submitted, the underlying logic of the two calendars is not the same. Karum attempted to show that the structure used by Lending was substantially copied, but the attempt did not survive the criticisms of Mr Jenkins and another FPF witness, Michael Harvey. By way of example, Mr Jenkins said that Lending sets up working fields in a different way; unlike CMS it does not store the delinquency calendar as a single row of 24 characters but creates a string of characters only when required, collating data into a temporary table. The Judge accepted this evidence, finding that FPF had copied functionality only, not the logic of CMS.
[142] The Judge found that a delinquency calendar is an elementary component of a credit management system and such calendars are invariably based on contractual aging; further, the use of numbers to represent days outstanding – such as 30 days (1), 60 days (2) and 90 days (3) – is also standard.[98] He accepted that the code values used in CMS were largely incorporated into Lending, but accepted that some of these, including numbers 1 to 3 for 30, 60 and 90 days and provision for write-offs and credit balances, had existed in F Credit and were mapped into F Credit, while others, such as C, R and W were obvious.
[143] The Judge concluded that, as with the CMS payment calendar, FPF had replicated only functionality:

[224] Although the delinquency calendar is more complex and its values and meanings less obvious than the payment calendar, I do not think it warrants any difference in approach. I consider FPF was entitled to replicate its functionality provided it did not copy the CMS software in order to do so. There was no particular skill and judgment involved in the identification of the codes used. For the most part, they were in common use, some had their equivalents in F Credit and, of course, they were known to business users. There was nothing secret about them.

[225] If there has been a breach of confidence or infringement of copyright, it is in the use of CMS processing logic or source code. I am satisfied that isolated instances of access to CMS source code were for the purpose of and restricted to ascertaining or clarifying RFS business rules. There is no evidence that CMS source code was copied in Lending. Mr Harvey identified material differences between the logic used in the two programmes. It is consistent with the view I have reached, based on my assessment of FPF witnesses, that the Lending program was developed from the Functional Specifications and not from copying the logic of the CMS system.

(h) Aged debt

[144] The Judge’s findings with respect to aged debt were brief:

[226] Karum argues, as a separate though related issue, that Lending copied the CMS software by which “aging buckets” were used to determine the values in the CMS delinquency calendar. Unlike payment and delinquency calendars, the aging buckets and logic driving their operation are not visible to a user. Karum says that the reports generated by FPF in the course of Operation Orpheus show that FPF accessed the CMS source code and replicated the aged debt logic in Lending.

[227] However, I accept the evidence of Mr Jenkins who explained that, for the purpose of managing Q card arrears, FPF had earlier (in 2004) developed an aged debt bucket system. He said that in order to achieve aged debt functionality for the RFS Farmers Card contracts, the code was written by an FPF developer employing the basic logic of the earlier development. His evidence was not challenged in cross examination. Mr Harvey confirmed that the methods by which the CMS and Lending programs perform their function are materially different.

[145] When a customer is in arrears, the lender must identify the amount in arrears for each monthly payment cycle. Payments are assigned to the instalment or “bucket” containing the oldest debt, with any surplus being poured into the next bucket, and so on. This reflects the terms of the lender’s contracts with customers. Manifestly, contractual aging is an integral feature of any credit management system. The witnesses agreed that it is normal to track payments on a monthly or 30-day basis. Obviously F Credit had done this before CMS.
[146] Karum asserts that the CMS aged debt component’s “buckets” and calculations are the foundation of its delinquency and payment calendars, and that FPF copied that component. It is said that the Judge overlooked concessions made by Mr Jenkins in evidence; he acknowledged that FPF now uses CMS source code to calculate aged debt buckets for FPF’s Q Card accounts.
[147] FPF responded that contractual aging is rote and all it did was to discern the specific business rules, which were industry standard, and confirm they corresponded to the terms of the lending contracts, so that nothing protected was copied; further, at code level aged debt functionality works differently in Lending.
[148] FPF also contended that Mr Jenkins did not concede that FPF employed CMS source code. We agree. He did acknowledge that the process used to calculate aged debt for credit limits on Q Card accounts was the same as in CMS, but it was not in dispute that as part of its Performance Indicators project FPF had checked the CMS business rules and sought to replicate the same functionality.[99] The important question was whether anything had been copied at a lower level. As to that, Mr Jenkins also said that FPF had separately developed an aged debt bucket system, in 2004, and that was the source of the aged debt component in Lending. The Judge recorded that Mr Jenkins was not cross-examined on this evidence.
[149] Mr Kennedy argued that the Judge was mistaken about this last point. We accept that Mr Jenkins was asked to comment on a general statement that FPF migrated more and more CMS functionality to Lending as time went on, and he accepted that the attributes referred to included the calendars and “CMS aging”. But that evidence did not confront his account of how the aged debt bucket system was developed in Lending. We note too that Karum did not challenge before us a corrected logic map produced by FPF which demonstrated that the logic used in Lending differed from that in CMS; CMS holds all buckets on the same row, while Lending holds them on individual rows denoting age.

(i) Did FPF infringe copyright in CMS?

[150] The parties joined issue on a number of sub-issues: whether the trial Judge misunderstood what Karum meant by “logic”, leading him to conflate it with source code; whether Karum claimed copyright in business rules of FPF or its predecessor businesses; whether it claimed copyright in functionality or ideas or industry standard concepts and rules; and whether anything copied was substantial in a copyright sense.
[151] Karum began by arguing that the Judge interpreted “logic” in an unduly restrictive manner which enabled him to dismiss Karum’s claims in their entirety. In particular, the Judge appeared to conflate or at least closely associate the concept of logic with source code. In the two passages cited to support this submission,[100] however, the Judge did nothing of the sort. He reached a conclusion that the logic of CMS and Lending are quite different by reference to what we have described as the structure of the two programs, not by comparing the source or object code which are the programs’ literal form of expression.
[152] We observe at this point that in his written submissions Mr Galbraith catalogued at some length the respects in which the structures of the CMS and Lending payment and delinquency calendars differ. Karum did not confront this convincing analysis directly, preferring for the most part to present its case at a somewhat more abstract level. It emphasised the process that FPF followed, relying on that both to sustain an inference of structural copying and to distinguish the UK cases in which similar non-literal claims have failed. It also argued that the calendar (and aged debt) components as a whole expressed their authors’ creativity, and so ought to enjoy copyright protection. We observe that a notable feature of Mr Maliga’s evidence, in particular, was the bald assertion that because effort was required the product was protectable intellectual property.
[153] Karum’s approach requires that we pay close attention to considerations that might confine or preclude copyright protection. It disclaimed any attempt to seek copyright protection for FPF’s business rules, and further accepted that FPF was entitled to identify and maintain business rules originally established by F Credit, implementing them in Lending. Mr Miles instanced such a rule: stop credit when overdue amount = X. Karum also acknowledged that FPF could migrate credit rating values which existed before Farmers utilised CMS.
[154] Karum argued, however, that FPF migrated not only the business rules but also the methodology by which they were applied; namely, the calendars. The codes and meanings in the Calendars were not business rules; they were system-specific methods used to convey base data to users. Mr Miles submitted that the Judge was wrong to characterise what were copied as business rules. He contended that FPF copied CMS modules precisely because copying saved it a good deal of time and money; he quoted an internal memorandum in which FPF recognised that customers’ credit profiles had to remain unchanged on migration and stated that the logic of CMS was “know[n] and relatively simple to implement compared with developing an alternative measure”.
[155] We are not persuaded that the Judge was wrong to conclude that all that FPF took from its examination of source code for the CMS payment and delinquency calendars was the business rules, otherwise poorly documented, of the predecessor businesses, in order to achieve the necessary functionality in its new system. FPF needed a precise understanding of the existing performance measures. Although reflected in CMS, performance measures fell into the category of business rules.
[156] Karum also contended that the Judge erred by failing to distinguish design from functionality or ideas. Mr Miles characterised functionality as the commercial purpose served by a program or component, and defined the functionality of payment and delinquency calendars respectively as that of giving users information about a customer’s payment history and delinquency history, while an aged debt module allowed a customer’s missed payments to be tracked and aged. At that level, counsel emphasised, Karum did not seek copyright protection. It was submitted that the United Kingdom cases – Navitaire, Nova, and SAS Institute – could all be distinguished because the defendants in those cases only replicated functionality and did not have access to source code.
[157] We do not agree that the United Kingdom cases can be distinguished. FPF’s access to source code is a distinction without a difference; Karum accepts that source code was not copied and the programming languages work differently. With respect to functionality, there is no material difference between this case and Navitaire, which concerned a “ticketless” airline reservation system. The airline concerned, easyJet, had used Navitaire’s software, but later engaged another developer to create a system that offered its users identical functionality. The underlying code and system architecture were dissimilar. Navitaire complained that easyJet had exploited knowledge of its software to copy commands, the screen displays and reports, and the “look and feel” of the website.
[158] Pumfrey J held that copyright protected the embodiment of functionality in software, not the functionality itself, and characterised the things copied – in effect, the user interface – as functional effects which were not protected. He observed, as we do, the plaintiff’s difficulty in identifying with precision something non-literal about its software that merited protection.
[159] With respect to commands, Navitaire claimed copyright both in individual commands and the collection as a whole. Pumfrey J held that copyright did not subsist in commands, which singly and collectively lacked the necessary qualities of a literary work.[101] He observed that:[102]

In the end, the question is merely whether a written artefact is to be accorded the status of a copyright work having regard to the kind of skill and labour expended, the nature of copyright protection and its underlying policy.

[160] Pumfrey J neatly described the challenge with which the plaintiff was left:[103]

The problem, therefore, if one is to arrive at a finding of infringement is to settle on a level of abstraction that describes something that is not merely inherent in the nature of the business function to be performed by the software, is taken by the defendants, represents the skill and labour of the designers and programmers but goes wider than the details of the command set and the screen displays ...

[161] He rejected the plaintiff’s attempts to locate something protectable in the links between screens and commands, concluding that they did not exist separately and could not be considered separately.
[162] Nova Productions concerned coin-operated video games which simulated the game of pool.[104] The defendants were said to have infringed, in their versions of the game, the plaintiff’s copyright in its own version. The games were visually similar and some features were common: notably, a power meter allowed the player to calibrate force, the power meter pulsed, a sight line indicated shot direction, the cue rotated around the cue ball to help the player’s aim, values were associated with pockets, and the table was shown in plan view. Copyright was claimed both as a graphic work (the screen shots) and as a literary work (the computer program). With respect to the latter, the Court of Appeal held, per Jacob LJ, that a written work consisting of the specifications of the functions of a computer program would attract protection as a literary work, but the functions themselves would not.[105] The plaintiff’s appeal failed.
[163] SAS Software concerned software used for statistical analysis.[106] The defendant sought to replicate the functionality of the plaintiff’s program as closely as possible, using the same language. Various claims were made, but the only one that concerns us is a claim that the defendant had indirectly copied elements of the plaintiff’s program. The Court of Appeal, per Lewison J, accepted that functionality falls on the ideas side of the line between idea and expression, as does a combination of several functionalities.[107] Further, if expression is dictated by technical function then the expression is not original, even if it involved skill and judgement.[108] And, as the High Court in SAS Software held:[109]

... the distinction between expression and ideas ... is one between different kinds of skill, judgement and labour. Skill, judgement and labour in devising ideas, procedures, methods of operation and mathematical concepts is not protected by the copyright in a literary work. What is protected by copyright in a literary work is the skill, judgement and labour in devising the form of expression of the literary work.

[164] Karum contended that the calendars and aged debt are not mere functionality; rather, they comprise design elements of CMS because they reflect the skill, care or judgment of the authors. We reject these submissions. The evidence of Mr Williams and another CMS author, Paul Hart, confirms that these elements required an impressive amount of programming knowhow and labour, but that is not enough. The codes used are not distinctive and in any event they are not literary works, whether taken singly or together, for the reasons given by Pumfrey J in Navitaire. The meanings associated with the codes are mere functionality, as Rodney Hansen J found. Further, we accept FPF’s submission that most of the same logic was used in F Credit, unsurprisingly so since there was nothing novel about it.
[165] In reaching these conclusions the Judge plainly rejected much of the expert evidence for Karum, in which it was said that system logic and “product data” (codes, flags and indicators) were intellectual property and features of CMS such as the code values used were unique and powerful. Speaking generally, we think he was right to do so. Although a number of Karum’s witnesses were expert, few could claim independence.[110] Some of the expert evidence adopted inappropriate legal assumptions about the status of, say, “system logic” or “product data” as intellectual property, or presented as hyperbole (such as the suggestion that while numeric codes are commonplace a combination of numbers and letters is “very powerful”), or supported aspects of Karum’s claim that failed at first instance and were not pursued on appeal, or was shown at trial to be mistaken in its understanding of Lending or FPF’s migration process. If accepted, this evidence would have allowed Karum to capture as its intellectual property functionality which simply implemented the business rules of F Credit and FPF. For reasons explained above, courts for the most part have resisted the attempts of legacy system owners to leverage intellectual property rights so as to inhibit competition from second comers.
[166] It follows inevitably from what we have said that if anything original was copied it was not a substantial part of the work. This conclusion follows whether CMS is treated as a single program or a series of elements, each capable of protection.[111]

(j) Conclusion on copyright infringement

[167] We are not persuaded that the Judge was wrong to find that FPF did not infringe copyright in CMS.

Breach of confidence

[168] This claim includes the payment and delinquency calendars and aged debt, but it extends to special codes and intercept codes, in respect of which breach of confidence is the only claim made. We begin with the latter items.

(a) Special codes

[169] Special codes are values which identify specific issues with cardholder accounts. Some merely present information (for example, that the customer is an employee). Others trigger an activity within the software that implements some business decision. For instance, an account which has been identified as forged (special code 02) has its statements suppressed and is not targeted for receipt of promotional materials.
[170] Karum says that the CMS special codes are unique. A detailed specification issued in 1994 records that they are assigned to particular functions. For example, code 07 means “Closed” and the specification details how the code is entered by an operator and explains that the software switches the code off should it conflict with another code, such as LBF (Legal Bankruptcy Follow-Up).
[171] Karum accepts that FPF merely mapped some codes to existing functional requirements in Lending and was entitled to do so. It makes no claim to a number of special codes developed by Farmers and added to CMS after the base software was delivered in 1994. It does say that FPF appropriated eight special codes which can be traced back to the base software. Those codes are 02 (Fraud), 04 (Collection Agency), 07 (Closed), 10 (Terms and Conditions Required), 14 (No Finance Charge), 24 (Do Not Write Off), 27 (Send Write Off Statement), and 54 (Permanent Pull).
[172] The following table compares special codes used by Farmers in F Credit and in CMS. The eight codes in issue are in bold:
Special Codes
F CREDIT
CMS
2 Fraud
G1 Fraud / G2 Fraud Recover Card
02 Forged
4 Collection agency
YA
04 Agency
5 RFS Closed
G3 No Further Credit or Inactive Budget Account - unclear
05 No resell
7 Closed
G6 Closed Account
07 Acct Closed
10 Terms and conditions required
A4, AS, C5, D5 - Temporary accounts
10 Need new app
14 No Finance Charges
Y1 Freeze Interest; Y2 Freeze Interest; Y8 Freeze Interest- Promotional Acct; EV VIP Customer
14 No Fin Chg
16 Purchase Order
B8 Purchase Order Required
No number 16 in doc
24 Do not write off
CU Special arrangement
24 Do not C/O
27 Send write off statement
27 P&L STMT
34 SIO filed
D2 Summ Inst Order
34 “unused’
40 Suing
C2 Sued Paid; D7 Summons issued
40 “unused”
41 Voluntary Return
DB Voluntary return
No number 41
42 Insur Prod
E2; HS Credit Card Sentinel
No number 42
44 Pull one month
Y9 Pull statement one month only
No number 44
48 NO CMSCS Lttrs
Z8 Hold Payment Demand Letters
No number 48
51 Cust Set Limit
Z6 Credit Limit set by Customer or No Further Credit
No number 51
53 No Agency Referral
YD
No number 53
54 Permanent Pull
HP Permanent Pull
54 Cntr Hold
[173] At trial Karum extended its claim beyond the codes themselves to the “system logic”, meaning the way in which the codes behave. It pursued this allegation faintly on appeal, saying that its claim extended to “the conditions and interrelationship of the codes”.
[174] FPF says that codes are routine and utterly unremarkable in credit management systems. A code has no substantive content at all; it is an arbitrarily chosen flag or sign which simply states a fact or a decision about the customer or its account. Such fact or decision is the product of business rules, not the “trade secrets” or system logic of the software provider. By way of example, a code may signify fraudulent behaviour and ban any further transactions; it does so because that is a requirement of the business.
[175] Of the eight codes which are the subject of the claim, FPF says that all but one were mapped into CMS from existing codes in F Credit and then mapped further into Lending. For example, the code values G1 and G2 (Fraud) in F Credit were mapped to code value 02 (Forged) in CMS, and in turn mapped to override code FRD (Fraud) in Lending. The exception is special Code 27 (Send Write Off Statement), for which data was mapped from an existing account status of “In Write Off” designated by the credit rating “H”.

(b) Intercept codes

[176] Customer statements generated by a credit management system are printed or forwarded as an electronic “statement print file” to a mailing service bureau – in FPF’s case, Datamail – for printing and despatch to customers. Intercept codes cause the bureau to send the statements or to withhold them for various reasons. For example, statements for customers whose recorded addresses are known to be inaccurate are withheld.
[177] Datamail printed Farmers’ statements, which required an agreed interface file so that Datamail’s system could interpret the data being sent to it. Karum says that FPF used the interface file format which CMS had developed for this purpose, and also utilised the CMS intercept codes. By these means it ensured that the interface with Datamail did not change on migration to Lending.
[178] The claim concerns the codes 54 (Permanent hold), 18 (Bankrupt), 12 (Repossession), 02 (Fraud/Forged) and 98 (Regular mail). The intercept codes used by F Credit and CMS are illustrated in the following table:
Intercept Codes
F CREDIT
CMS
54 Permanent hold
040 Pull statement (HP, Y9)
54 Center Hold
18 Bankrupt
08
18 Bankruptcy
12 Repossession
DA, DB
12 Repossession - special code 12
02 Fraud/Forged
140 Frauds (G1, G2)
02 - Forged - special code 2
98 Regular Mail
92-95 (various different sorts of regular mail) - 98 is pre-sort first class)
[179] FPF concedes that it used these codes, but it says they are trivial, having no meaning in themselves, and argues that some of them served different purposes in the base CMS Software. Code 98 (Regular Mail) depended on ZIP codes and imperial measures, so was of no use in New Zealand. In Lending, 98 now represents the default – that is, “print”. Of the other four codes, it says that the relevant functions all existed in F Credit and it simply mapped them to a CMS code value Farmers made significant use of intercept codes, and these codes were mapped to the CMS intercept codes on conversion from F Credit.
[180] FPF concedes that it retained the existing interface file layout with Datamail, so that intercept code values still appear in the Datamail interface file for Farmers statements (but not for new FPF accounts). There is one minor difference; Datamail formerly printed statements subject to intercept codes and sent them to Farmers; now it does not print them. It says that Karum failed to prove it created the interface file, and that the file cannot be confidential given its purpose as an agreed protocol with a third party, Datamail.

(c) The breach of confidence claim

[181] At trial, Karum’s breach of confidence claims were that FPF had failed to maintain the confidentiality of the CMS software, and had failed to comply with the clean room representations, and had incorporated parts of the CMS software into Lending.[112] Only the last of these is now pursued. The claim with respect to appropriation of special and intercept codes does not depend on FPF’s Performance Indicators process for migrating functionality to Lending, and it is not suggested that FPF accessed CMS source code in relation to these items. (Of course, Karum does make that claim in relation to the calendars and aged debt.)
[182] The claim is brought in equity, for breach of confidence. (There was some dispute about this before us, Mr Miles suggesting that it was brought in contract under SLA 2005, but we think it is clear from the pleading and the judgment below that the claim was pleaded and argued in equity, although the licence was called in aid; it was said to prove the necessary quality of confidentiality.) The elements of the cause of action are not in dispute: the information must have the necessary quality of confidence, it must have been communicated in circumstances importing an obligation of confidence, and it must be used in an unauthorised manner to the detriment of the person communicating it.[113] The first and last of these elements require elaboration.

(i) The quality of confidence

[183] The Judge found, and it is not in dispute, that at source and object code level the CMS software was confidential; further, it was communicated in circumstances importing a duty of confidence.[114] He accepted that there had been some inconsequential breaches of confidence in source code.[115]
[184] The Judge observed that SLA 2005 provided in cl 2.3 that:[116]

Client acknowledges and agrees that the Software and the systems, ideas, methods of operation, and information contained therein are proprietary “trade secret” information of CMS, the use and disclosure of which must be continuously controlled and that the Software is protected by the Copyright Act and treaties of the United States. Without limiting the generality of the foregoing, Client will not make the Software or any part thereof, including screen-shots, available or accessible to third parties (not being members of the Client Group).

[185] He reasoned, however, that the terms of SLA 2005 were not conclusive, and he found that the special and intercept codes were not confidential. For the reasons outlined earlier, FPF was entitled to ascertain the business rules of F Credit which led to the special codes being activated, and to give effect to those same rules in Lending. The special codes were not confidential. Nor were the intercept codes; they were arbitrary, and they had been shared with Datamail since the inception of CMS.
[186] We have already examined the evidence about the calendars and aged debt. We record that the Judge did not consider them confidential either. The calendar values were not confidential, nor were the things that they stood for. FPF replicated functionality, but it was entitled to do that. With respect to aged debt, he found that FPF did not use CMS information; rather, it developed its own system.
[187] Karum argues that SLA 2005 must be considered conclusive, and that in any event the information concerned is manifestly confidential in nature. FPF responds that the claim is brought in equity and it is for the Court to decide what is confidential, viewed objectively and in light of public policy considerations which restrict monopolies over information.[117] It adds that cl 2.3 must be read consistently with cl 2.2, which asserts ownership of trade secrets and provides:

CMS shall own all right, title and interest in and to all intellectual property rights, including patent, trademark, copyright and trade secrets rights, related to the Software and, subject to Section 2.4, any and all services provided by CMS in connection with Client’s use of the Software. Client agrees to execute and deliver such confirmations, assignments and other documents as CMS may request to confirm such ownership.

[188] We do not accept FPF’s claim that cl 2.2 is confined to trade secrets: it includes all intellectual property rights in CMS. However, we do accept that it is ultimately a question for the court whether any particular item of information has a sufficient quality of confidence to justify intervention in equity. Naturally the answer will be informed, and sometimes supplied, by the parties’ agreement, but the agreement here is not dispositive. SLA 2005 provided that nothing in cl 2.3 applied to the extent that the things listed there had become generally available to the public or had been disclosed to FPF on a non-confidential basis. It may be, for example, that information found in CMS is ubiquitous in credit management systems, or was held independently by Farmers or FPF: in that case the information cannot possess the necessary quality of confidence or sustain a remedy. Further, the parties agree that FPF was entitled under SLA 2005 to examine CMS to extract business rules and identify functionality employed in F Credit.
[189] We observe again that Karum’s argument relied heavily on the process that FPF followed; for example, counsel contended that CMS aged debt was replicated, and confidential information misused, because FPF dissected and analysed that module before writing its own program. But process does not take Karum very far. It is not in dispute that FPF used the eight special codes and five intercept codes, or that it examined the calendars to identify functionality and business rules. Karum has not shown that the functions served by these codes were new to F Credit, or that FPF is wrong to say that almost all of them were mapped from existing F Credit codes. FPF also used certain values found in the payment and delinquency calendars, as we have noted earlier. FPF was entitled to examine CMS for the purpose of identifying existing business rules and functionality, and that entitlement extended to testing both systems to ensure nothing was lost in translation.
[190] We accept FPF’s submission that rules as to what a code does (indicating that certain behaviours are to happen if a customer is being pursued for arrears, for example) are business rules which cannot be confidential to CMS. The same is true of rules which cancel a given code in the event that another code is applied. Mr Miles invited us to distinguish between business rules and the means by which they are implemented in software, but the distinction is of no assistance; Karum has identified nothing tangible lying between business rules, which are not confidential, and the object and source code, which are (but which FPF did not misuse).
[191] Karum argued that the codes themselves were confidential, for they were restricted to FPF operators and employees, all licensed users; further, the aged debt component was not visible to operators. Mr Miles pointed out that confidential information need not be unique; protection may be afforded to information that is publicly available where it has been collated into a convenient and valuable form, and the “elegant simplicity” of the CMS design is one of its key advantages.
[192] We are not persuaded that the Judge was wrong to find nothing confidential about the codes in issue, which are arbitrary and trivial both singly and collectively. That is so even where a given code, such as 02, is associated with a specific business rule.

(ii) Detriment

[193] The parties joined issue on whether detriment is required in New Zealand law. We accept Mr Galbraith’s submission that equity intervenes to prevent unfairness; that in a commercial case detriment is the normal measure of unfairness; accordingly, a plaintiff in a commercial case must normally point to detriment if it is to get relief in equity.[118] Nothing turns on the point, however. Where FPF did only that which it might lawfully do Karum cannot justify intervention in equity.
[194] Karum argued that it has suffered detriment in two respects; it has lost control of its software and it has lost licence fees which FPF ought to have continued to pay so long as it used elements of CMS. The Judge did not accept that detriment had been suffered. We agree, for the reason that FPF acted within its rights. If it were otherwise, an inquiry into Karum’s losses would be required.

(d) Conclusion on breach of confidence

[195] We are not persuaded that the Judge was wrong to find that FPF did not breach its obligation of confidence to Karum, because such CMS information as it used was not confidential and because FPF was entitled to use the information to extract business rules and copy functionality.

Conclusion on intellectual property claims

[196] We end our analysis of the intellectual property claims by quoting the conclusions of the trial Judge:

[265] Contrary to the suspicions harboured by Karum, FPF was never motivated to appropriate CMS secrets. It was quite capable of designing and writing credit management programs no less complex and sophisticated than the CMS system. It had little to learn from the CMS system other than an understanding of the functionality that had to be replicated in Lending in order to achieve a seamless migration of RFS customer accounts onto the Lending platform. That functionality was a product of the business rules of [Farmers]/RFS. There was nothing secret or confidential about it.

[266] FPF took all reasonable steps to accommodate Karum and to comply with its contractual and broader legal obligations. Staff who worked on the business integration project were well aware of what they could and could not do. For the most part, there was compliance. Having regard to the scale, duration and complexity of the project, transgressions were relatively few. None involved a deliberate attempt to appropriate protected information. None resulted in detriment to Karum.

[267] Inspection of the Lending system persuaded Karum to the view that in key areas FPF had incorporated the logic of the CMS software. That view rested on a fallacy that pervaded Karum's case: that, to the extent that the business rules of [Farmers]/RFS were reflected in or embedded in the logic of the CMS software, they could not be replicated in Lending. That is not what the law of copyright or the obligation of confidence require. FPF could not copy the source code or the logic of the CMS progra[m]. But it was fully entitled to develop a progra[m] which emulated the CMS progra[m] in order to give effect to RFS business rules. In my view, ultimately, that is all it can be said to have done.

[197] Karum has not persuaded us that these conclusions were wrong.

Result

[198] Karum’s appeals and FPF’s cross-appeals against Judgments (No 2) and (No 3) and Ruling (4) are dismissed.
[199] Karum’s appeal against Judgment (No 4) is dismissed.
[200] Karum is ordered to pay costs for a complex appeal on a Band B basis together with usual disbursements. We certify for two counsel.



Solicitors:
Minter Ellison Rudd Watts, Auckland for Appellant
A J Park, Auckland for Respondent

[1] Fisher & Paykel Financial Services Ltd v Karum Group LLC (No 4) [2012] NZHC 3314, [2013] 2 NZLR 266 [HC Judgment]. The reported version of the case omits [24]–[127]; our references to those paragraphs are therefore to the neutral citation.

[2] Austin, Nichols and Co Inc v Stichting Lodestar [2007] NZSC 103, [2008] 2 NZLR 141 at [16].

[3] Austin, Nichols, above n 2, at [13].

[4] Rae v International Insurance Brokers (Nelson Marlborough) Ltd [1998] 3 NZLR 190 (CA) at 197–200.

[5] Fisher & Paykel Financial Services Ltd v Karum Group LLC (No 2) [2012] NZHC 240; Fisher & Paykel Financial Services Ltd v Karum Group LLC (No 3) [2012] NZHC 794; and Fisher & Paykel Financial Services Ltd v Karum Group LLC HC Auckland CIV-2006-404-6646, 11 October 2011 (Ruling 4).

[6] At [90]–[97].

[7] Fisher & Paykel Financial Services Ltd v Karum Group LLC [2013] NZHC 587. A decision has yet to be made on costs for the substantive High Court hearing.

[8] HC Judgment, above n 1, at [24]–[55].

[9] At [20]–[23].

  1. [10]These steps are detailed in the HC Judgment, above n 1, at [128]–[131] and in earlier judgments of Rodney Hansen J in Judgment (No 2), above n 5, and Judgment (No 3), above n 5.

[11] Ware v Johnson [1983] NZHC 155; [1984] 2 NZLR 518 (HC) at 537; New Zealand Motor Bodies Ltd v Emslie [1985] 2 NZLR 569 (HC) at 593–594; Wills v Williams CA38/99, 20 July 1999 at [49].

[12] Edgington v Fitzmaurice (1885) 29 Ch D 459 at 470; Buxton v The Birches Time Share Resort Ltd [1991] 2 NZLR 641 (CA) at 646–647; Manderson v Violich (1992) 5 TCLR 124 (HC) at 130–131; McKenzie Institute International v ARCIC (1997) 8 TCLR 329 (CA) at 331.

[13] HC Judgment, above n 1, at [65].

[14] See below at [71]–[81].

[15] The date should be a year later, 17 January 2005.

[16] Trustees Executors Ltd v QBE Insurance (International) Ltd [2010] NZCA 608 at [32].

[17] HC Judgment, above n 1, at [114]–[117].

[18] At [117].

[19] At [117]–[118].

[20] At [119].

[21] At [124].

[22] At [121].

[23] At [123].

[24] See above at [50](5).

[25] HC Judgment, above n 1, at [126].

[26] PAE (New Zealand) Limited v Brosnahan [2009] NZCA 611, (2009) 12 TCLR 626.

[27] Ellmers v Brown (1990) 1 NZ ConvC 95-076 (CA) at 190,572 and 190,577–190,578

[28] HC Judgment, above n 1, at [120].

[29] At [69]–[113].

[30] At [69]–[112].

[31] At [94].

[32] At [70]–[71].

[33] See above at [23] and [24].

[34] Henkel KGAA v Holdfast New Zealand Ltd [2006] NZSC 102, [2007] 1 NZLR 577 at [34].

[35] Copyright Act 1994, s 15(1). Writing includes any form of notation or code: s 2, definition of “writing”.

[36] Section 15(1); Agreement on Trade-Related Aspects of Intellectual Property Rights 1869 UNTS 299 (opened for signature 15 April 1994, entered into force 1 January 1995) [TRIPS], art 9(2).

[37] Section 14(1).

[38] Section 14(2).

[39] Henkel, above n 34, at [37].

[40] Henkel, above n 34, at [41].

[41] K Garnett, G Davies, G Harbottle (eds) Copinger and Skone James on Copyright (16th ed, Sweet & Maxwell, London, 2011) [Copinger] at [7–30(a)].

[42] Bleiman v News Media (Auckland) Ltd [1994] 2 NZLR 673 (CA) at 678, cited in Henkel, above n 34, at [44].

[43] H Laddie and others The Modern Law of Copyright and Designs (4th ed, LexisNexis, London, 2011) at [3.129]; Copinger, above n 41, at [7–57]; Copyright Act, s 2, definitions of “compilation” and “literary work”.

[44] Above at [10].

[45] Judgment (No 2), Judgment (No 3), and Ruling (4), all above n 5.

[46] Copyright Act, s 123.

[47] Copyright Act, s 124.

[48] Fisher & Paykel, above n 7.

[49] Judgment (No 2), above n 7, at [20], [32] and [35].

[50] At [21].

[51] The rights to CMS were ultimately transferred to Karum through a series of companies.

[52] At [21].

[53] At [22]–[24].

[54] At [25].

[55] At [28]–[40].

[56] At [38].

[57] At [41].

[58] There are many different types of programming paradigms: see Laddie, above n 43, at [36.12].

[59] Computer Associates International Inc v Altai Inc [1992] USCA2 1158; 982 F 2d 693 (2nd Cir 1992). A valuable description is also found in Data Access Corp v Powerflex Services Pty Ltd [1999] HCA 49, (1999) 202 CLR 1 at [4]–[5].

[60] Copinger, above n 41, at [7–57].

[61] IBCOS Computers Ltd v Barclays Mercantile Highland Finance Ltd [1994] FSR 275 (Ch).

[62] More sophisticated descriptions may be found in the texts; see for example Laddie, above n 43, at [36.9] and [36.12]. This definition suffices for our purposes.

[63] The term “module” is sometimes used to describe a subroutine or collection of subroutines.

[64] International Business Machines Corp v Computer Imports Ltd [1989] NZHC 159; [1989] 2 NZLR 395 (HC) at 411–413.

[65] Data Access, above n 59, at [54]. This is so notwithstanding that code does not convey ideas to a reader in the same way that words do.

[66] Copinger, above n 41, at [7–57].

[67] Altai, above n 59, at 698; Data Access, above n 59, at [54]; SAS Institute Inc v World Programming Ltd [2010] EWHC 1829, [2011] RPC 1 [SAS (HC)] at [229]. Algorithms are excluded from protection as mathematical methods: see TRIPS, above n 36, art 9(2).

[68] At [58].

[69] At 702.

[70] SAS (HC), above n 67, at [232].

[71] SAS Institute Inc v World Programming Ltd [2013] EWCA Civ 1482, [2014] RPC 8 [SAS (CA)] at [39].

[72] Whelan Associates Inc v Jaslow Dental Laboratory Inc [1986] USCA3 976; 797 F 2d 1222 (3rd Cir 1986).

[73] Notably, it was rejected by the Ninth Circuit Court of Appeals in Sega Enterprises Ltd v Accolade Inc [1993] USCA9 19; 977 F 2d 1510 (9th Cir 1992) at 1525.

[74] Altai, above n 59, at 705.

[75] SAS (HC), above n 67, at [207]; Copinger, above n 41, at [3–30(2)].

[76] Designers Guild Ltd v Russell Williams (Textiles) Ltd [2000] UKHL 58; [2000] 1 WLR 2416 (HL) at 2422–2423; SAS (HC), above n 67, at [206].

[77] See Navitaire Inc v easyJet Airline Co Ltd [2004] EWHC 1725, [2006] RPC 3 at [125]; Nova Productions Ltd v Mazooma Games Ltd [2007] EWCA Civ 219, [2007] Bus LR 1032 at [55]; SAS (HC), above n 67, at [232].

[78] Altai, above n 59, at 707; Navitaire, above n 77, at [125].

[79] Altai, above n 59, at 707. This is not to suggest that utilitarian works cannot be the subject of copyright: see Copinger, above n 41, at [3–30(4)]; SAS (HC), above n 67, at [235].

[80] Altai, above n 59, at 707–710; Navitaire, above n 77, at [129]. See also Copinger, above n 41, at [3–30(2)].

[81] As described in evidence by Paul Goldstein, Lillick Professor of Law at Stanford University and author of a US treatise: Paul Goldstein Goldstein on Copyright (3rd ed, Aspen, New York, 2014).

[82] TRIPS, above n 36, art 9(2).

[83] Navitaire, above n 77, at [129]–[130].

[84] SAS (HC), above n 67, at [229]; Copinger, above n 41, at [7–55]. Compare Laddie, above n 43, at [36.38].

[85] Nova, above n 77, at [32]–[33] and [44].

[86] SAS (HC), above n 67, at [249].

[87] Navitaire, above n 77, at [130]–[131].

[88] Directive 1991/250/EEC on the Legal Protection of Computer Programs [1991] OJ L 122/42.

[89] At [31]. We also refer to TRIPS, above n 36, art 9(2).

[90] See below at [155]–[167].

[91] Copyright Act, ss 18 and 19; HC Judgment, above n 1, at [164]–[169].

[92] Above at [15].

[93] HC Judgment, above n 1, at [190]–[195].

[94] At [213].

[95] At [215].

[96] At [216].

[97] “A–F” in CMS was mapped to “*” in CMS; “W” in F Credit was mapped to “9” in CMS; “.” in F Credit was mapped to “-” in CMS; and “*” in F Credit was mapped to “C” in CMS.

[98] At [221].

[99] See above at [128].

[100] HC Judgment, above n 1, at [217] and [225].

[101] Navitaire, above n 77, at [79] and [92].

[102] At [80].

[103] At [118].

[104] Nova, above n 77.

[105] At [51].

[106] SAS (HC), above n 67; SAS (CA), above n 71.

[107] SAS (CA), above n 71, at [45]–[46].

[108] At [33] and [46].

[109] SAS (HC), above n 67, at [207].

[110] See, for example, above at [132].

[111] We make this point because in its pleadings Karum seemed to characterise “the CMS software” as a single program, of which FPF was alleged to have variously misused “a substantial part or parts”.

[112] HC Judgment, above n 1, at [133].

[113] A B Consolidated Ltd v Europe Strength Food Co Pty Ltd [1978] 2 NZLR 515 (CA) at 520–521, adopting Coco v A N Clark (Engineers) Ltd [1968] FSR 415 (Ch) [Coco] at 419–421.

[114] At [140] and [143]–[144].

[115] At [220] and [225].

[116] At [140].

[117] Yates Circuit Foil Co v Electrofoils Ltd [1976] FSPLR 345 (Ch) at 384; Stephenson Jordan & Harrison Ltd v MacDonald (1952) 69 RPC 10 (CA) at 23; Potters-Ballotini Ltd v Weston-Baker [1977] RPC 202 (CA) at 205–206; G D Searle & Co Ltd v Celltech Ltd [1982] FSR 92 (CA) at 104; Scott & Associates Engineering Ltd v Finareva Renewables Inc 2013 ABQB 273 at
[90]–[92].

[118] Coco, above n 113, at 421 (in most cases, “detriment ought to be present if equity is to be induced to intervene”); K Garnett, G Davies, G Harbottle (eds) Copinger and Skone James on Copyright (15th ed, Sweet & Maxwell, London, 2005) at [20–21] (exceptions are most appropriately those cases involving personal or non-commercial information); Hunt v A [2007] NZCA 332, [2008] 1 NZLR 368 at [61] (basis of breach of confidence is that “He who has received information in confidence should not take an unfair advantage of it”) and [94] (detriment remains a factor to be considered).

NZLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.nzlii.org/nz/cases/NZCA/2014/389.html