FEDERAL COURT OF AUSTRALIA Motorola Solutions, Inc. v Hytera Communications Corporation Ltd (Discovery) [2019] FCA 1887 File number: NSD 1283 of 2017 Judge: PERRAM J Date of judgment: 7 November 2019 Date of publication of reasons: 15 November 2019 Catchwords: PRACTICE AND PROCEDURE – application for discovery – where documents sought said to be relevant to ownership and subsistence of copyright in, and originality of, applicant's source code – where source code in part derived from earlier code – consideration of necessity and likely effects of discovery being ordered Legislation: Copyright Act 1968 (Cth) ss 14, 35 Federal Court Rules 2011 (Cth) r 20.11 Cases cited: Data Access Corporation v Powerflex Services Pty Limited [1999] HCA 49; 202 CLR 1 IceTV Pty Limited v Nine Network Australia Pty Limited [2009] HCA 14; 239 CLR 458 Interlego AG v Tyco Industries Inc (1988) 12 IPR 97 JR Consulting & Drafting Pty Ltd v Cummings [2016] FCAFC 20; 116 IPR 440 Neilson v Overseas Projects Corporation of Victoria Ltd [2005] HCA 54; 223 CLR 331 Date of hearing: 7 November 2019 Registry: New South Wales Division: General Division National Practice Area: Intellectual Property Sub-area: Patents and Associated Statutes Category: Catchwords Number of paragraphs: 26 Counsel for the Applicant: Mr C A Moore SC with Mr A R Lang Solicitor for the Applicant: Herbert Smith Freehills Counsel for the Respondents: Mr C Dimitriadis SC with Mr T Cooke and Mr B Mee Solicitor for the Respondents: Shelston IP Lawyers ORDERS NSD 1283 of 2017 BETWEEN: MOTOROLA SOLUTIONS, INC. Applicant AND: HYTERA COMMUNICATIONS CORPORATION LTD First Respondent HYTERA COMMUNICATIONS (AUSTRALIA) PTY LTD ACN 165 879 701 Second Respondent JUDGE: PERRAM J DATE OF ORDER: 7 NOVEMBER 2019 THE COURT ORDERS THAT: 1. The Respondents' interlocutory application for discovery dated 25 October 2019 ('Discovery Application') be dismissed. 2. The Respondents pay the Applicant's costs of the Discovery Application. 3. The matter be listed for a case management hearing at 9:30am on 18 November 2019. AND BY CONSENT, THE COURT ORDERS THAT: 4. By 29 November 2019, the Applicant provide verified discovery of the documents identified in Annexure A. 5. By 29 November 2019, subject to any privilege claim, the Applicant produce to the solicitors of the Respondents a copy of each of the documents discovered pursuant to order 3, such production to be given in electronic format in accordance with the electronic discovery protocol agreed between the parties. Note: Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011. Annexure A Definitions Document has the meaning set out in the Dictionary in Schedule 1 of the Federal Court Rules 2011 (Cth), which includes the definition of 'document' in Part 1 of the Dictionary to the Evidence Act 1995 (Cth). Motorola Works has the meaning given to that expression in paragraph 31A of the Applicant's Second Further Amended Statement of Claim dated 3 September 2019. Category All Documents recording any transfer of ownership of the Australian copyright subsisting in any of the Motorola Works, or any portion thereof, including but not limited to those referenced in Paragraph C.3(d) of the Motorola Amended Particulars of Copyright Claim filed on 2 September 2019. REASONS FOR JUDGMENT PERRAM J: 1 This application for discovery by the Respondents ('Hytera') was dismissed with costs at a hearing on 7 November 2019. These are the reasons for that conclusion. The application itself was advanced by means of an interlocutory application dated 25 October 2019. The parties prepared an application book. The book contains much which is commercially sensitive. To avoid the necessity of having to make suppression orders over some of the book's contents the course I took was to treat all the affidavits in the book as having been read (there were no objections) and with the documentary exhibits within it being treated as if in evidence. I indicated that after the reasons for judgment were published I would return the book, with its sensitive contents, to the parties' representatives. For their part counsel conducted the hearing by not reading aloud any of the forbidden material so that no part of the transcript needed to be suppressed. In addition to this material, Hytera also tendered 'amended particulars of its copyright claim' and 'amended further particulars regarding patent and copyright infringement and additional damages' which had been provided to Hytera by the Applicant ('Motorola'). I treated these as Exhibits 1 and 2 respectively. Exhibit 1 contains much which is confidential as well and I indicated that I would return Exhibit 1 to the parties on the same basis. Because it is surplus to my requirements I will also be returning Exhibit 2. I record these matters lest there subsequently be an issue as to what was before the Court. For completeness, I have included a schedule at the end of these reasons setting out what was in evidence in more detail. 2 Hytera eventually sought discovery of four categories of documents. There were other categories in dispute but these were resolved consensually between the parties. The remaining four categories were 1, 2, 3 and 9. Category 1 3 By paragraphs 31A-31C of the second further amended statement of claim Motorola alleges that the 'Motorola Works' are original literary works in which copyright belonging to it subsists. At paragraph 31D it then alleges that, inter alia, certain Hytera devices embody a reproduction in material form of the whole or a substantial part of one or more of the Motorola Works. The Motorola Works are defined in the particular to paragraph 31A to be three suites of software which are then further defined in the particulars. The three suites are the Mobile Firmware, the Portable Firmware and the DSP Firmware. The definition of the Motorola Works is, in fact, a little broader than this for it is also alleged that the programs which make up the three suites are themselves literary works in which its copyright inheres. For the purposes of these reasons, that aspect of Motorola's definition of the Motorola Works may be largely, although not entirely, disregarded as the issues arising from it are the same as those which arise in relation to the three suites themselves. 4 The three software suites were issued together under the title 'Release 1' or, less prosaically, 'R01.00.01'. The three suites of software are, in fact, the firmware for Motorola's MOTOTRBRO range of portable and mobile radios. The three suites of software came into being as a result of Motorola's digital mobile radio project. This project was given an internal title within Motorola. It is not clear to me whether that name is itself subject to the confidentiality orders which thus far have been made in this proceeding and hence ought not to be uttered so, in these reasons, I will call it Project Alpha. 5 It is apparent that some of the components of the Project Alpha firmware were not created specifically for Project Alpha but already existed having been created for other earlier projects. The number of these other projects and their purposes is presently unimportant. To make it easier to follow Hytera's argument and to render the projects' names anonymous (lest they too be confidential) I will refer to them as Project Bravo and Project Charlie. The bottom line is that some components of the Project Alpha firmware are derived from lines of code generated in the course of Project Bravo and Project Charlie or, if not directly copied from that source code, are modifications of it. 6 Motorola alleges its copyright in the source code for its Project Alpha firmware has been infringed by Hytera by its acts of, inter alia, importing and selling various devices (all of which relate, in one way or another, to mobile radio handsets) which are alleged to embody a reproduction in material form of a substantial part of the firmware. This allegation is particularised at annexure 2 to the particulars. This is a long and difficult document. At its grittiest (which is very) it contains allegations that identified lines of source code contained within the Motorola firmware are infringed by identified lines of code within the source code for the firmware in Hytera's devices. So for example, there is an allegation that lines 80-82 and 103-116 of an identified file in Motorola's firmware have been copied in lines 41-44 and 80-83 of identified files in Hytera's firmware. I interpolate here both for completeness and accuracy that in fact the lines of code referred to are not on either set of devices. They are instead lines of source code which, once assembled, are installed on the devices. But that does not matter for present purposes. 7 Hytera does not deny that copyright subsists in the firmware but it does put the subsistence of that copyright in issue at paragraphs 31A, 31B and 31C of its second further amended defence. It does deny the allegation at paragraph 31D of the pleading that its devices embody a reproduction in material form of a substantial part of the firmware. 8 Hytera is interested in pursuing an argument along the following lines: assuming against it that some lines of its source code are a copy of some lines of Motorola's source code, it wishes to submit that the lines of Motorola code are not a substantial part of the firmware (a reproduction of a substantial part of the firmware being taken to be a reproduction of the whole by s 14 of the Copyright Act 1968 (Cth) ('Copyright Act')). A straightforward way of testing that contention would be to examine the lines of source code in Motorola's firmware alleged in each case to have been reproduced and then to compare them, from both a quantitative and qualitative perspective, with the firmware overall to see if the lines of source code relied upon are, in fact, a substantial part of the firmware. 9 However, Hytera submits that in carrying out such an assessment it is relevant to know how original the lines of source code are, citing Data Access Corporation v Powerflex Services Pty Limited [1999] HCA 49; 202 CLR 1 at 32-33 [83]-[84]; IceTV Pty Limited v Nine Network Australia Pty Limited [2009] HCA 14; 239 CLR 458 at 473-474 [30]-[32] and 508-512 [154]-[171]. That some components of the Motorola firmware had not been written for the purposes of Project Alpha but were taken from earlier parts of Project Bravo and Charlie was indicative that those parts were not in that sense original (where 'original' in the context of copyright law essentially means not copied from somewhere else). Ex hypothesi these components were copied from somewhere else and hence those parts could not be original with the consequence that they ought not now to count in the calculus of assessing whether Hytera has taken a substantial part of the Motorola firmware. 10 To put the argument in more concrete terms by means of an example raised during argument: if I produce an annotated version of Hamlet with footnotes, corrected spellings and with a rearrangement of the act and scene structure to reflect recent discoveries in Elizabethan theatre, this opus will certainly be a literary work in which a copyright, belonging to me, subsists. But if some other author also produces an annotated, restructured and footnoted edition of Hamlet which I allege involves a substantial reproduction of my version, then it would be unsound to include in the substantiality analysis any of the words written by Shakespeare (which, depending on the length of the footnotes, might be expected quantitatively to be most of my annotated work). It would be true that 95% of my annotated Hamlet included the play Hamlet but I could not claim substantial reproduction of my annotated Hamlet just because the other author included Shakespeare's words as well. 11 This is the line of territory which Hytera now wishes to traverse: to investigate how much of the source code for the Motorola firmware has its origins in source code written at an earlier time for the purposes of Projects Bravo and Charlie and to pursue erected thereon a full frontal attack based on an absence of substantiality. As it happens, Motorola has kept detailed records of the coding history for the development of the Project Alpha firmware and, although an examination of these records is not something I would at all recommend, nevertheless there does not appear to be too much doubt that these records are such as would feasibly permit to be carried out the kind of epic deconstruction which Hytera now desires to embark upon. 12 I do not think, however, that this would be a useful endeavour. I accept the internal logic of the Hamlet example but the difficulty is that the analogy between Shakespeare and the authors of the software in Projects Bravo and Charlie is imperfect. All of the authors of Project Bravo and Charlie are employees or contractors of Motorola. Let it be supposed that the proposed discovery did reveal that lines 100-150 of a particular file in the Motorola firmware did in fact come from Project Bravo so that lines copied from them and found in Hytera's firmware could be said not to involve a reproduction of an original part of the Motorola firmware in Project Alpha. All that would flow from that forensic success would be an amendment by Motorola of its case so as to rely upon the corresponding element of the source code in Project Bravo. This would be in circumstances where Motorola owned that copyright in that source code too. 13 Hytera had two answers to this. First, repeated iteration of this process might well reveal that Project Bravo was not the true headwaters of the source code either which might be revealed by further discovery to lie upstream in Project Charlie or even beyond, perhaps in Project Delta. Secondly¸ it should not be assumed that Motorola's putative amendment application would succeed. 14 I am unpersuaded by either of these propositions. As to the first, all that would then occur is that the process would be repeated until the eventual root of the source code was fossicked out, perhaps somewhere in Project Lima. But the authors of Project Lima would still be employees or contractors of Motorola and the same result would occur, that is to say, Motorola would still own the copyright. As to the second, I do not see why Motorola's amendment application would fail. It has advanced the proposition that copyright in the source code for its firmware subsists. No doubt, like all commercial software, it is made from various elements, parts of which have been updated from earlier parts. However, it is uncontroversial that such a software program is itself a new literary work in which copyright subsists: JR Consulting & Drafting Pty Ltd v Cummings [2016] FCAFC 20; 116 IPR 440 ('JR Consulting') at 496 [304]. There is nothing therefore delinquent about Motorola's pleading that copyright subsists in its firmware as Hytera sought to insinuate (Hytera initially submitted that copyright did not subsist in an individual release of software which was updated from earlier software but it withdrew that submission during the hearing). Hytera's argument about substantiality is all its own work and does not follow from the way Motorola has played its hand. I do not see any legitimate basis upon which I could refuse to allow Motorola dynamically to amend its pleadings to respond to Hytera's iterative contentions that there was no substantial reproduction of the source code in software W because elements of it came from earlier software X. Where Motorola owns software X (and software Y and Z for that matter) it would be perverse not to permit it to point that out. 15 That being so, the effect of allowing this discovery expedition to set out will be sixfold. First, it will be unduly onerous for Motorola's records of the history of the firmware's production are complex. Secondly, most likely the present discovery application will only be the first in a cascading series of discovery applications as Hytera iteratively seeks to uncover the fons et origo of each line of code which Motorola alleges to have been copied. The scale of this enterprise given the size of the firmware should not be underestimated. Thirdly, the iterative nature of what is contemplated by Hytera follows because each step in that process will engender concomitant pleading amendments by Motorola. Fourthly, this will produce an eventual pleading which is complex. Fifthly, I am not especially confident that this enterprise will be completed by the time the trial commences in May 2020. Sixthly, once the expedition returns home nothing will have been achieved by it for Motorola will still own the copyright in the now-finally-unearthed fons. 16 Hytera developed two variants of the argument I have just rejected. The first of these was as follows: the issue of whether the root of a particular piece of source code was an original work was not the really the correct question. Rather, granted that a particular part of the source code in Project Alpha could be found in Project Bravo, the real question was whether the alteration which had occurred between Project Bravo and Project Alpha was sufficient to mean that the passage which appeared in Project Alpha was original. Senior counsel for Hytera submitted that there had to be some element of material alteration or embellishment from the earlier version of the work which sufficed to make the totality of the later work an original work: Interlego AG v Tyco Industries Inc (1988) 12 IPR 97 ('Interlego') at 122. But in this case the suggestion is not that the three software suites resulting from Project Alpha are a reproduction of source code derived from earlier projects. Were that the case then one might well inquire into whether it was the three suites, on the one hand, or their predecessor, on the other, which constituted the original work (although there would remain the difficulty that even on that hypothesis Motorola would remain the owner of the copyright). But that is not what has happened here. What is said instead is that portions of the three suites can be found in the earlier works. The application of the Interlego principle to those portions is the wrong inquiry where what is required is an inquiry into the overall software suites. That situation seems to me to be squarely covered by JR Consulting. It is true, I acknowledge, that the Motorola Works are also alleged to include component programs within the three software suites and here the argument may have more traction, i.e. because one could meaningfully ask what the relationship between the program components in Project Alpha and Project Bravo was—was it a copy or was it original?) But this analysis would still lack utility because success for Hytera would only drain back to an earlier copyright work of which Motorola would still be the owner. 17 The second variant concerned the possibility, flirted with in oral argument, that archaeological excavation of Motorola's source code back to the early Neolithic might reveal that there was no copyright in it all. Here, as I understood this audacious argument, one began with the principle in Interlego that the copying from an earlier version of a passage of code would only be original where there was some sufficient change to what had gone before. Thus, there would be no copyright in a program contained within the three suites if one could see that it was copied from Project Bravo perhaps with some minor alteration but not of such a kind which would to lead to the conclusion that the segment of source code in the three suites was thereby original. This would then take one to the corresponding source code in Project Bravo. Here the same inquiry might, if permitted, show that the segment had in turn been copied from Project Charlie with minor but, again, not sufficient alteration to mean that it was an original work either. Working from this point recursively it might then appear that the program in question had been built up over time by means of a gradual accretion of non-material alterations where each new layer of alteration changed the source code but not as in such a way as to give rise to an original work. As one descended into the full history of the Motorola source code, the size of the unearthed source code would shrink as each layer of non-substantial alteration was carefully brushed away and it might finally appear that the root of the source code was itself so insubstantial that no copyright in it could inhere (this last step is necessary otherwise Motorola's ownership of the copyright in that stub renders the argument futile). Consequently, it would follow that there was no copyright in that original root nor, because of the insubstantial nature of each accreted layer of alteration, in the resulting substantial source code in the relevant portion of the three software suites. 18 Whilst I admire, without wishing in anyway further to encourage, the undoubted ingenuity of the persons who fashioned this argument, I reject it for a number of reasons. First, the utility of the undertaking is to be weighed against the chances that some useful piece of the code's early progenitor was so insubstantial that copyright in it cannot inhere. But this seems to me most unlikely. If the three suites contain a program whose purpose is to check to see if it is possible to synchronise a timing signal one may accept that this function may have been performed at earlier, and perhaps simpler, times by less sophisticated software but I struggle to see that any of those progenitors are at all likely to be so insubstantial as not to constitute copyright works. Secondly, assuming in its favour that Hytera is right and somewhere in the distant past such a forlorn stub existed, the chances that it has developed across time by means of an unbroken process of alteration where each new version is so insubstantially different from the last that at no point does an original work ever emerge seems to be vanishingly unlikely and perhaps impossible. Thirdly, assuming in Hytera's favour that its discovery campaign is not entirely quixotic, what emerges is a proposal to allow a significant discovery undertaking where the chances of anything useful emerging from it are, at their very highest, exceptionally low. I do not think, in this case, what is proposed is proportionate, sensible or, to be frank, even useful. I reject both variants of the argument. 19 For these reasons, discovery of these materials would be time-consuming, trial date threatening, expensive and ultimately pointless. It would not facilitate the just resolution of the proceeding as quickly, inexpensively and efficiently as possible as required by r 20.11 of the Federal Court Rules 2011 (Cth). That is why I refused it. Category 2 20 The point of category 2 is to find out whether the employees of Motorola had in their employment contracts some provision which meant that the copyright in the code they produced was vested in them rather than Motorola. This would be contrary to s 35(6) of the Copyright Act which provides that where a literary work is made an author in pursuance of the terms of a contract of employment the employer is the owner of the copyright. Section 35(3) allows this to be modified by contract. This question may well be governed by foreign law in this case but Hytera did not seek, in its pursuit of category 2, to demonstrate what the foreign law was and I proceed on the assumption that it is the same as Australian law: Neilson v Overseas Projects Corporation of Victoria Ltd [2005] HCA 54; 223 CLR 331 at 372 [125], 416-417 [267] and 411 [249]. Essentially what it is sought therefore is to go through Motorola's employment contracts to see if, by chance, any of them contains a clause in which Motorola has expressly agreed that the copyright in code created for it by its own employees belongs to those employees. This strikes me as distinctly unlikely and not worth expending resources on. For that reason, I refused category 2. In doing so, I note that Motorola was prepared to consent to some of this category (and category 3). That does not alter my conclusion. Category 3 21 Category 3 was the same as category 2 except that it was concerned with independent contractors. There is no default position in that case under the Copyright Act as there is in the case of employees. It strikes me as unlikely that Motorola would agree with contractors writing code for it that the copyright in that code should remain vested in the contractor unless Motorola also obtained at the same time an exclusive licence to use the code. If this were not so, it would be liable for using the very software whose creation it had commissioned. It is difficult to understand any rational reason Motorola would act in this way. It seems likely that this could only occur by errors of a clerical kind. 22 When would this undertaking therefore have utility? Its only point would be in the process described under category 1 of showing that the original root of the source code was owned by a contractor (as part of Hytera's substantiality analysis). This would only be useful if: (a) there was a contractor where such a drafting error had occurred (i.e. leaving out an exclusive licence); and (b) the lines of code produced by that contractor happened to be lines of code now alleged by Motorola to have been infringed by Hytera. 23 The prospects of (a) and (b) are individually low. The chances of both are remote. I do not think that the effort that would be involved justifies granting this category. It is not necessary for the fair trial of this case. It was for that reason that I refused it. Category 9 24 This category related to Motorola's aggravated damages case. That case is that Hytera appears to have a quantity of Motorola's documents which appear on their face to be confidential and whose presence in Hytera's possession is something it will need to explain at some point. Hytera wishes to have Motorola give discovery of the circumstances in which documents marked in various ways to be confidential contain open source code, have been released into the public domain, or are given to third parties including Hytera. In my view, this approach to the matter is upside down and, at least this stage, neither useful nor necessary. Hytera is yet to put on its evidence. When it does it will be obliged to explain how it came by Motorola documents marked apparently as confidential—a matter which it is reasonable to assume is within the knowledge of Hytera. 25 If it wishes to prove that it received the documents from third parties, or that it was open source software or that Hytera in fact obtained the documents from Motorola then it is at liberty to give evidence to that effect. It does not need discovery to do this. It was for that reason I refused to give discovery of this category. Result 26 Having rejected all categories which remained in dispute, I dismissed Hytera's application with costs. There was a category which was agreed and I made orders for it by consent. I certify that the preceding twenty-six (26) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Perram. Associate: Dated: 15 November 2019 Schedule Evidence at Interlocutory Hearing No Description Date 1 Hytera's Interlocutory Application for Discovery 25 Oct 2019 2 Fifth affidavit of Katrina May Crooks (Crooks #5) 25 Oct 2019 3 Annexure KMC-7 to Crooks #5 25 Oct 2019 4 Confidential Annexure KMC-8 to Crooks #5 (Confidential Material – External Advisors Only) 25 Oct 2019 5 Confidential Annexure KMC-9 to Crooks #5 (Confidential Material – MSI Source Code) 25 Oct 2019 6 Seventeenth Affidavit of Sue Maree Gilchrist (Gilchrist #17) 1 Nov 2019 7 Confidential Annexure SMG-92 to Gilchrist #17 (Confidential Material – External Advisors Only) (Confidential Material – MSI Source Code) 1 Nov 2019 8 Confidential Annexure SMG-93 to Gilchrist #17 (Confidential Material – MSI Source Code) 1 Nov 2019 9 Confidential Annexure SMG-94 to Gilchrist #17 (Confidential Material – MSI Source Code) 1 Nov 2019 10 Sixth Affidavit of Katrina May Crooks (Crooks #6) 5 Nov 2019 11 Confidential Annexure KMC-10 to Crooks #6 (Confidential Material – MSI Source Code) 5 Nov 2019 12 Eighteenth Affidavit of Sue Maree Gilchrist (Gilchrist #18) 6 Nov 2019