IBM United Kingdom Limited v LzLabs GmbH & Ors.

This is the fifth CMC in a claim that arises out of the development by the first defendant (LzLabs) of software, known as the ‘Software Defined Mainframe’ or ‘SDM’, which is said to enable LzLabs’ customers to take applications developed for IBM mainframe computers and run them on x86-based computer architectures without the need for source code changes or recompilation.

The claimants’ case is that in developing the SDM, the defendants disassembled, decompiled or otherwise reverse engineered IBM mainframe software, licensed by the claimant to the second defendant (Winsopia), so as to replicate it within the SDM or, alternatively, that they copied it.

The claimant is a licensor of the IBM mainframe software within the UK.

The defendants are:

On 9 August 2013 IBM licensed the IBM mainframe software to Winsopia pursuant to an IBM customer agreement (ICA). Subsequently, further licence agreements were entered into by the parties.

The claimant’s case is that Winsopia used the IBM mainframe software to develop the SDM to run IBM software systems without an IBM mainframe or or the IBM mainframe software stack by reverse assembling, reverse compiling or reverse engineering the software.

In December 2020 the claimant requested an audit of Winsopia under the terms of the ICA and other licence agreements, but Winsopia refused on the ground that it exceeded the ambit of the claimant’s contractual audit rights.

By notice dated 24 February 2021 the claimant terminated the licence agreements for contractual breach; alternatively at common law.

On 5 October 2021, the claimant issued these proceedings, in which it seeks: (i) a declaration that Winsopia’s licence has been lawfully terminated; (ii) an injunction restraining Winsopia from making any further use of the IBM mainframe software, including from offering any services relying on the SDM that contains or uses any part of the IBM mainframe software (and D1, D3, D4 and D5 from procuring the same); and (iii) an account of profits and/or damages.

The defence is that the claims are speculative, based on inference and bound to fail. SDM is a software platform which enables mainframe users to use their legacy applications in current computing environments, such as x86, ARM and Linux, without the need to rewrite and all recompile those applications; it does so by providing a thin compatibility layer, which provides an interface between the users’ applications and current computing environments. They claim injunctive declaratory relief and damages.

The defendants’ case is that SDM was developed by LzLabs following an extensive research and development process spanning almost 10 years, using strict processes and policies which applied both to LzLabs and the developers whom it engaged, and to Winsopia, to ensure that no IBM material was used other than in compliance with the terms of the ICA. In developing the SDM, LzLabs employed a clean room process and did not use Winsopia’s IBM mainframe.

At a case management conference on 21 October 2022 before Waksman J, directions were given to a trial on liability, fixed for 9 April 2024 with an estimate of 28 days (7 TCC weeks), including 4 days of judicial reading.

The court has before it a number of applications on both sides. The first is an application made by the defendants seeking guidance from the court in relation to its disclosure. There were originally three limbs to the application but, happily, one of those limbs has now been resolved, leaving two:

As part of the clean room process established by the defendants, provision was made in its code of conduct since early 2014 for calls and online meetings to be recorded and archived. The result is over 8900 hours of recorded MP3 audio and Webex video files in 9203 recording files; over 400 of the calls exceed two hours and they average over 35 minutes in length. As a result of the nature of the calls and the operation of the code of conduct, lawyers were often in attendance and therefore it is the defendants' case that significant numbers will contain privileged legal advice.

The starting point is the agreed DRD document that set out the basis on which disclosure would be given. In relation to the MP3 call recordings, the following was set out:

“These are MP3 call recordings of primarily weekly and ad hoc conference calls between the first, second and third defendants. There are 503 MP3 call recordings from 22 April 2014 to 29 April 2020, which amount to approximately 2 GB in total.

D1-D3 consider that the most proportionate and cost effective way of transcribing the recordings will involve automated machine learning transcription assisted by manual review for quality, relevance, privilege and confidentiality.

Following initial testing, the defendants note that manual review is necessary, for example, to determine who is speaking. The defendants have compiled an index of all call recordings, listing their dates, durations and custodian attendees insofar as they can be ascertained from initial review, from which the claimant has been invited to select a sample group of up to 51 MP3 files which the defendants will then transcribe, review and disclose as appropriate, subject to any necessary redactions. Failing any such nomination from the claimant, the defendants will proceed to choose the sample themselves across a range of dates and custodians.”

The DRD statement in relation to the Webex videos is:

“Webex is an online communications platform used by D1-D3. Meetings were recorded from around May 2018 until around April/May 2020. It was used for calls between D1 developers and D3 employees as well as for customer calls.

The recordings are video based (in MP4 and arf format) and the stored files have a total volume of around 477 GB. The defendants have identified in the region of 8,800 recordings and continue to investigate if others subsist… the defendants' data forensic firm has converted these files to MP4 for ease of review. The defendants have compiled an index of all identified recordings and the claimant has been invited to select a sample of up to 10% of those recordings which the defendants will then review and make available for inspection as appropriate. Failing any such nomination of files from the index by the claimant, the defendants will proceed to choose the sample themselves across a range of dates and custodians. Review will employ automated transcription software assisted by manual review for quality, relevance, privilege and confidentiality. Following initial review, it is clear that manual review will be necessary to determine, for instance, who is speaking and whether the communications are privileged. For the avoidance of doubt, any recordings which are calls between D1 or D3 and a customer will be removed as they are irrelevant, and any text transcripts referring to confidential and/or privileged communications, including customer information, will be redacted.”

The defendants’ position is that the likelihood of probative documents existing in these call recordings is low because it is said recorded calls are very unlikely to be a ‘backdoor’ for conveying ICA Programs, not least because software cannot be provided verbally. The defendants therefore selected (as they said they would do in their DRD) a sample of 100 hours, a significant proportion of which were found to be privileged or irrelevant. Reviewing all of them would be both prohibitively expensive and take up enormous time. The sample is small relative to the total number of recordings, but not insignificant. A full 100 hours of audio/video have been reviewed, with specific focus on inter-company meetings where this could be identified.

The claimant’s position is that these recordings and video communications are potentially of critical importance to the issues to be determined at trial. The audio recordings consist, or ought to consist, of a running record of all discussions which took place between the defendants, and which would include discussions involving employees of the second defendant in the course of their analysis on the mainframe and would also reveal the identity of the person who was provided with access to the mainframe. Such discussions are liable to show, amongst other matters, whether or not the clean room procedure was followed and, more generally, communications between the defendants in relation to the development of the SDM. The claimant contends, that, whilst not a perfect solution, the defendants should at least, and in the first instance, produce automated transcripts of those recordings, which can be done rapidly and at little cost, which would enable searches to be conducted and any cross-check to be focused on documents of particular interest.

I note that an index has been provided by the defendants to the claimant of all of the relevant calls, although it is accepted that more information has been provided in respect of the MP files rather than the Webex video calls.

The review that was carried out by Ms Scott, as set out in her 14th witness statement, is that of her sample, which consisted of 241 recordings, 135 were found to have been relevant to at least one of the issues for disclosure; 104 were documents which were potentially confidential and identified as potentially relevant; four were identified as wholly covered by privilege; and five were considered to be covered by privilege in part.

What this indicates to the court is that the recordings would appear to contain material that is potentially relevant, in that it appears to be responsive to issues identified by the parties; also, the recordings would appear to contain material, much of which is unlikely to be privileged.

The difficulty that arises is the time taken to carry out the review and therefore the costs and also the likely probative value of any documents so produced. Ms Scott has compared a selection of automated transcripts with corrected transcripts based on the audio in which she has identified and exhibited to the court the number of mistakes that are made by an automated transcription service.

Mr Stewart KC, leading counsel for the defendants, has explained that it is not sufficient for the transcripts alone to be produced; it is necessary to listen to the audio in order to carry out appropriate corrections and to identify the relevant individuals speaking. It is also then necessary for there to be a manual review in order to identify confidential documents and/or privilege. Further, it is said that at the very best all this produces is secondary evidence.

In response, on the issue of confidentiality, Mr Saunders KC, leading counsel for the claimant, has indicated that that could be dealt with by disclosure of all the relevant documents into tier 1 of the confidentiality ring which would then enable a further review to be carried out, if necessary, in order to change the designation of the documents.

The court considers that, based on the limited review carried out by Ms Scott to date, the documents are potentially relevant and contain material that is unlikely to be covered by privilege. Therefore the starting point is that there should be some disclosure of these documents, particularly as the parties included such documentation as relevant for disclosure in the DRD.

The court considers that the first exercise should be for the defendants to obtain automated transcripts of all of the recordings. Once that has been done, key search terms should be applied in order to exclude documents that contain nothing of relevance. That may well exclude documents that are simply subject to incorrect transcription, but the court considers that at this stage that is an appropriate risk that the parties should take. Once the search terms have returned a pool of responsive documents, the documents can then be reviewed for privilege and relevance, and should then be disclosed to the claimant.

When the documents have been disclosed, it will be open to the claimant to identify any gaps in the disclosure given or, alternatively, to sit down and discuss with the defendants whether there is an alternative method that might produce a better pool of documents. This seems to the court to be a reasonable and proportionate approach that will at least produce a decent sample of documents that will indicate whether or not there are any significant probative value documents within the pool.

The starting point is the order of Waksman J made in February 2023 at paragraph 4(b), in which he directed the defendants to disclose the Git repository for the current and previous versions of CPX, together with the previous versions of CPX which are not contained within the Git repository insofar as they still exist.

CPX is a tool which was used to ‘scrub’ or delete IBM proprietary elements from the software running on the second defendant’s Mainframe as part of the process of transportingthose elements to the SDM.

The question of the ambit of disclosure in relation to the CPX has already come before the courts on a number of occasions. On 22 February 2023, the judge made the relevant order to which I have just referred. There was a further dispute before the court in March at the CMC at which the court was asked to consider whether it should order the defendants to carry out a reasonable search for the CPX data or whether it should disclose all versions of CPX in their possession and control. The court ordered that the defendants should disclose all versions of CPX in their possession and control, as sought by the claimant.

The basis of the application by the defendants is that it has carried out searches for the current and historic versions of the CPX tool which they were ordered to disclose. They have disclosed 325 versions of CPX, including their complete Git repository, dating from 2018. In addition, the defendants have disclosed 146 further archive files of historic CPX versions in unstructured repositories stored on the Winsopia mainframe, which date back to 8 March 2017. In addition, the defendants searched two backup cassette tapes from January 2017 (in a “3592” tape format) and no further versions of CPX were found.

The current application concerns 35 archived backup tapes for disaster recovery of unknown date but probably more recent than January 2017.

The defendants submit that they cannot review or restore them because they are in an older 3590 tape format. They have not looked at those back-up tapes but, based on the researches carried out and the review of the two earlier back-up cassette tapes, they consider that it is unlikely that those tapes will contain further versions of CPX. On that basis it is submitted any further searches and disclosure of the back-up tapes would be both unreasonable and disproportionate.

The claimant relies on Waksman J's earlier rulings which expressly rejected the points now asserted by the defendants in its application. Mr Saunders submits that it is clear that the defendants are required to disclose all versions of CPX, whether contained in a Git repository or not, and that carrying out a reasonable search for previous versions of CPX is not sufficient for the purpose of complying with the February order. There is now no dispute that the defendants can in fact locate the previous versions of the CPX and, in particular, it is submitted that there must be versions of the CPX that pre-date 2017 which have not yet been disclosed.

In terms of the significance of this issue, the claimant submits that it is clear that information contained on early back-up tapes is likely to be of importance to the dispute generally. Based on their enquiries with Mr Alepin, their expert, the claimant understands that the back-up tapes are likely to contain information that would be found on the mainframe, namely files that include artefacts of SDM development work, tools such as those used to develop CPX, and others used to carry out analysis of IBM programmes, three test case repositories, libraries and other files relating to tests used to ascertain specifications and operational details of IBM programmes on the mainframe, and tests developed on the mainframe for testing on the SDM system conducted by Winsopia.

The defendants submit that it is very difficult or impossible to restore the relevant back-up tapes. However, the claimant has made enquiries of third party providers which have confirmed that it would be possible to recover the relevant data and that such a service could be undertaken for a relatively modest cost. Once the tapes have been recovered, it would then be possible for the defendants to give disclosure of them.

I consider that the order made by Waksman J in February and subsequently reviewed in March is one that should be maintained. For the reasons that have been set out in the earlier rulings, it is clear that the CPX material is critical to the central issues between the parties which concern whether or not there has been reverse-engineering by the defendants in order to produce the SDM. Although the defendants have carried out a significant amount of work to both locate and disclose historic CPX versions so far, it seems to me that it is not an unreasonable or disproportionate exercise for them to go on and restore the 35 archived back-up tapes that they have already located and disclose those as originally ordered.

This is the claimant’s application to relax the requirements for disclosure of documents in the ‘Cognitive Support Portal’ or CSP.

This relates to bug reports by IBM customers, communications between IBM support and IBM customers following such reports, and APAR fix descriptions. Responsive documents to the request are contained in a centralised reporting system, the CSP, belonging to IBM Corp. CSP is a live platform used by IBM customers and IBM support personnel in order to report, and communicate in order to resolve, bugs. IBM service personnel can access, search and display records from all customers using a tool known as ‘Support Search’, but that tool does not allow them to print records.

In the DRD, the claimant agreed that it would provide a responsive request to a model C request, requiring it to disclose such documents. It submits that, whilst the CSP is web-based, using the tools which are available to extract and provide records would involve a disproportionately labour-intensive process of printing the records on a result-by-result basis. For that reason, the claimant has put forward an alternative by way of a solution to that dilemma by offering for the defendants to provide suitable reviewers to attend at a mutually convenient location, which could be one of the solicitors' offices, to carry out keyword searches of the CSP as they see fit and to extract manually the records that they wish to extract, using an IBM employee who will carry out the relevant searches and extract the records on behalf of the reviewers.

The application is opposed by the defendants who submit that the claimant has agreed to carry out these searches and they are relevant to pleaded allegations by the claimant that a number of programmes could not be produced as the defendants have produced them without reverse-engineering. Mr Stewart explained to the court that the CSP repository contains an archive stretching back many years of IBM's mainframe troubleshooting and debugging interactions with customers. It is said that it is likely to show what IBM routinely instructed its licensees to do, prior to the date of the ICA, in order to diagnose and trouble-shoot problems with ICA programmes. It is also likely to show that these included the same sorts of steps that IBM now complains were breaches of the ICA when done by Winsopia. Further, it is also likely to show that users of IBM mainframes commonly encountered problems with ICA programmes that required their interaction with customer applications to be debugged, traced and analysed closely.

The defendants’ position is that the claimant accepted in its DRD that it would conduct searches of responsive communications between IBM and the customer insofar as stored on an IBM centralised problem-reporting system. That is what it should be required to do.

Given the difficulties that have been identified by the claimant, the defendants have proposed alternative keyword searches in order to cut down on the number of hits. At annex A to their skeleton argument, the proposed keyword searches currently produce about 12,000 hits, but with some minor tweaks to take out word searches that produce too many hits, that could be reduced to about 9,000. There are then a number of search terms proposed that have an unknown number of hits but which the defendants say should be used to carry out a search.

The court is satisfied that the defendants have established the potential relevance of these documents. It is clear that the extent of any searches needs to be reasonable and proportionate. The suggested keyword searches now proposed by the defendants in the reduced list seem to me to be a sensible starting point and the court considers that the claimant should carry out these searches across its CPS so as to identify documents which can then be reviewed for relevance and/or provided to the defendants.

As discussed in relation to the other disclosure orders that the court has made today, if in fact when the search terms are applied they produce a number of hits that is wildly in excess of the 12,000 or so that is now identified by the defendants, then it is of course open to the parties to have a discussion and consider how to resolve that; alternatively for the parties to make short written submissions to the court for determination in writing.

In summary, the claimant agreed to this in the DRD, there is no good reason to change the nature of the disclosure at this stage, and the defendants' amended proposals seem to be a reasonable starting point.

This is an application by the defendants dated 2 May 2023 pursuant to paragraph 17 of the Practice Direction 57AD in relation to the claimant's disclosure.

The defendants seek an order that the claimant should serve a witness statement providing answers to various questions set out by the defendants in correspondence, specifically:

From that recitation of the orders sought it can immediately be seen that focus is very sharply on the role played by Mr Anzani in relation to key issues identified by the defendants in their defence relating to the decision to terminate the ICA and the audit, of which Mr Anzani was a central figure.

Mr Stewart has set out in his skeleton argument the key concerns that have given rise to this application. The defendants are concerned at the paucity of the claimant’s disclosure. In particular, they are concerned with the lack of documents relating to the role of Mr Anzani and the role of Mr Paul Knight, who was Winsopia's customer relationship manager between 2013 and 2018, and who eventually left the claimant in 2020.

The central issues with which the defendants are concerned in relation to this application are the claimant’s decision to carry out the audit, its decision to terminate and the issue of limitation: when the claimant first knew of the matters relied on in relation to both the audit and termination, namely, the allegation that the defendants were using reverse engineering in order to implement the SDM.

The defendants submit that there is an absence of clarity in the claimant's approach to documents held by Mr Anzani and those held on IBM Corporation systems. They are concerned about unilateral changes that the claimant has made to search terms in its disclosure certificate, the result being, it is said by the defendants, that IBM's provision of voluntary disclosure of documents from IBM Corporation and Mark Anzani has resulted in limited documentation being disclosed, despite his involvement. He was the IBM group's key internal point of contact in relation to LzLabs, he was the person who initiated the audit request to Winsopia on 3 December 2020 and he held himself out as vice president of IBMZ, IBM Systems Group. He is said by the defendants to have orchestrated a campaign against LzLabs by contacting the defendants’ customers about the SDM and he was the one that met with the fourth and fifth defendants during 2020 to discuss the SDM and alleged IP infringement issues. The defendants also point to the removal of key words related to both Mr Anzani and Mr Knight, as explained in the DRD and the disclosure certificate. All of those matters, it is said, give rise to a clear indication that there has been a failure adequately to comply with the order for extended disclosure by the claimant and that in those circumstances the court should order the claimant to make a witness statement explaining its approach to disclosure.

The application is opposed by the claimant, who submits that there is no basis for making such an order in the present case and it would serve no useful purpose. Mr Saunders submits that the jurisdiction on the part of the court to make the order sought depends on the defendants making out a case that there has been, or may have been, a failure adequately to comply with an order for extended disclosure. However, the evidence of Ms Scott in her 13th witness statement merely raises suspicions about the claimant's disclosure in relation to IBM Corp and Mr Anzani's documents.

Further, the claimant has sought to respond to the defendants' queries in relation to those matters in correspondence. The seventh witness statement of Mr Pantlin explains that there has been no breach of the claimant's disclosure obligations; Mr Pantlin has overseen the disclosure exercise and is satisfied that the claimant has complied with its disclosure obligations in respect of IBM Corp and Mr Anzani.

The starting point for dealing with this matter is to consider the court's jurisdiction. Paragraph 15 of Practice Direction 57AD provides at 17.1:

“Where there has been or may have been a failure adequately to comply with an order for Extended Disclosure the court may make such further orders as may be appropriate, including an order requiring a party to –

(1)

serve a further, or revised, Disclosure Certificate;

(2)

undertake further steps, including further or more extended searches, to ensure compliance with an order for Extended Disclosure;

(3)

provide a further or improved Extended Disclosure List of Documents;

(4)

produce documents; or

(5)

make a witness statement explaining any matter relating to disclosure.”

There has been some consideration by the courts in relation to these matters. In the case of Berkeley Square Holdings Ltd v Lancer Property Asset Management Ltd [2021] EWHC 849, Robin Vos, sitting as a deputy judge of the High Court, set out guidance on the approach to be taken to applications which sought to remedy non-compliance with extended disclosure. In particular, he made the point that it was necessary for the party seeking the remedy to identify that there was a likelihood of relevant documents existing.

In Sheeran v Chokri [2021] EWHC 3553 (Ch) Meade J set out the appropriate approach by the court in such matters to be that first, the court should consider whether there has, or is likely to have been, a failure to comply with the extended disclosure obligations of a party; secondly, if that threshold is passed, the court should consider whether it is reasonable and proportionate in all the circumstances of the case to make the relevant order.

Having considered the witness statements provided by each party in respect of this matter, the court considers that there is no demonstrated failure on the part of the claimant to comply with its obligations of extended disclosure. However, the court does have some concern that there should be further searches carried out in relation to both Mr Anzani and Mr Knight.

The DRD document sets out the relevant issues for disclosure:

The Model C requests for disclosure in relation to issue 21 are:

Those issues are clearly relatively wide.

The custodians that are identified as being the subject of the relevant searches include Mr Anzani, in respect of whom it is noted that:

“Mark Anzani's documents (including his emails) are not within the control of the claimant. Notwithstanding the foregoing insofar as Mr Anzani acted for or on behalf of the claimant in respect of any matters that fall within the list of issues for disclosure, he will be asked to provide the claimant with all documents in his possession or control in relation to such. Insofar as Mr Anzani communicated with any of the above identified custodians, those communications will be searched.”

The claimant's extended disclosure certificate dated 31 March 2023 contains further explanation in relation to Mr Anzani and Mr Knight.

In relation to Mr Anzani, as against issues 17, 19 and 21, the claimant states this key word was varied to Anzani for issue 17 and issue 19 and removed for issue 21. It is explained that including the words "Mark" or "Anzani" produced responsive hits of 193,157. When the disjunctive key word search was done with Anzani, that returned 67,766 responsive hits and 3,433 responsive hits with the inbox limitation, of which 935 responsive hits related to issue 17 and issue 19.

The claimant state the inclusion of the key word "Anzani" for issue 17 and issue 19 and the removal of the key word for issue 21 was a proportionate decision that was necessary to balance the inclusion of relevant key words and the objectives of promoting the cost effective and efficient conduct of the disclosure process.

A similar statement is made in relation to Mr Knight in relation to issues 19 and 21. The key word was not modified for issue 19 but was removed for issue 21.

I consider that the general complaints that are raised by the defendants as set out in Ms Scott's witness statement do no more than raise a general suspicion that there does not seem to have been very much disclosure given. That is not sufficient to engage paragraph 17 of PD57AD. However, in relation to Mr Anzani and Mr Knight, I consider that the defendants have made out a good case that the claimant's decision to limit the key word searches in respect of those two individuals has produced a very limited number of responsive hits, which suggests that the key word search is too narrow. That, it seems to me, gives the defendants an argument under paragraph 17 of the practice direction that there may have been a failure adequately to comply with an order for extended disclosure.

I emphasise that that is not a finding by the court that the claimant has intentionally or negligently failed to comply with its disclosure obligations; merely that the evidence put before the court by the defendants indicates that there may have been an inadvertent failure to comply with the order for extended disclosure by virtue of the limitation on the key word searches made in respect of Mr Anzani and Mr Knight.

Therefore, the court orders that the imposed limitations in relation to searches using the term "Anzani" and searches using the term "Knight" should be removed and the search should be re-run. That is a proportionate approach taken by the court to the legitimate concern raised by the defendant that the searches carried out to date may not be sufficient to comply objectively with the extended disclosure ordered by the court and set out in the DRD document. The parties can address the court as to when that can be done.

In summary, the court refuses the defendants’ application to order a further witness statement but will order the claimant to re-run the searches as above.

This is the claimant's application for an order for additional searches to be carried out by the defendants and for disclosure to be given against additional key word searches using search terms identified by the claimant.

Since the issue of the application, the scope has been narrowed by the claimant so that what is now sought is an order to require the defendants:

The genesis of this application is that this case poses significant challenges for the parties in terms of identifying reasonable and proportionate search terms that will produce useful responsive hits that are manageable, reasonable and proportionate in the context of this case, against a background where it has become evident that the parties, for perhaps obvious reasons given the nature of the case, are mistrustful of each other.

Both parties rely upon the obligation to cooperate. Each party considers that they have been helpful and made sensible suggestions but that the other party has failed to respond adequately and/or to cooperate.

It is clear that this is a challenging part of the case, but it is also clear to the court that there must be a sensible way forward. The claimant has sought to suggest that the search proposals put forward can be adequately narrowed if the defendants provide a breakdown of responsive hits to those search terms across various repositories and email accounts of custodians.

The difficulty, it seems to the court, is that that will not necessarily give the claimant useful information. It will identify the overall numbers of responsive hits but it will not indicate whether those hits are of any use. And I say that having regard to what the defendants have correctly identified as being very general search terms such as "legal" or "lawyer" which quite clearly will produce responses, many of which -- arguably most of which -- will in fact relate to privileged material which will simply require the defendants to review thousands of documents and claim privilege.

Further, there are search terms such as "pizza box" which likewise throw up a huge number of responsive hits, many of which will simply be of no relevance to the issues that are raised in this case.

What is required is for the parties to meet and to have a sensible discussion about the search terms that are likely to provide responsive hits on the real issues in the case. That can be done by identifying a term such as "legal" or "lawyer" but adding to it an additional term that is more likely to produce a responsive hit that is relevant to one of the issues in the case.

This will require the parties to sit and think about appropriate search terms, rather than just producing generic terms and then seeking to reduce the numbers of responsive hits. Rather than numbers, the parties should concentrate on producing relevant hits, responsive hits. I am not going to make an order in the terms either originally sought by the application or in the revised terms set out in the claimant's skeleton at paragraph 88. However, I will require the parties' lawyers to meet within seven days and discuss the proposed search terms that are currently set out in the annex both to the claimant's application and in the annex to Ms Scott's 14th witness statement; to discuss a sensible way forward by limiting the proposed search terms to specific custodians who are likely to produce relevant material and/or to a refinement of the terms. Not only is that likely to be more reasonable and proportionate in terms of the burden placed on the defendants, but also it is more likely to be better focused and helpful to the claimants, who want relevant documents that can be used in their assessment of the merits of the case and preparation for any amendments and/or other presentation of their case.

This is the defendants’ application for further information in relation to the claimant’s plea of deliberate concealment, which is in itself in response to the defendants’ limitation defences.

The Amended Defence raises the issue of limitation as follows:

61.

Pending disclosure, it is to be inferred that IBM and/or IBM UK has at all material times since at least 2013 (alternatively, by no later than May 2017) been aware of the existence of LzLabs, the nature of the work in which it was engaged, and the relationship between LzLabs and Winsopia.

83.

Further and in any event, by reason of clause 1.11.4 of the ICA, it is denied that IBM UK is entitled to bring a legal action, regardless of form, arising out of or related to the ICA (or any of the other Agreements or any transaction under them) more than two years after the cause of action arose.

87.

Further or alternatively, insofar as (i) the breaches alleged in paragraphs 23 to 27 and the claim for breach of contract against Winsopia, or (ii) the claims in tort against LzLabs and LzLabs UK, relate to acts which were committed by Winsopia prior to 21 September 2015, any such action is barred by sections 5 and two of the Limitation Act 1980, respectively.

The Reply denies that the claims are time barred and raises a plea of deliberate concealment:

4.6

The claims are not time barred. The Defendants have not properly pleaded the factual basis on which the time bar is said to arise, with the result that its plea in this regard does not permit a full response. Without prejudice to that: ICA clause 1.11.4 does not apply to these claims; and, by s32, the primary limitation period under the Limitation Act 1980 does not apply.

52.

In any event, the Defendants’ acts constituted deliberate concealment within the meaning of s32 of the Limitation Act 1980 and so the limitation period did not arise until the Claimant discovered the Defendants’ deliberate concealment. At trial the Claimant will rely on each and every example of deliberate concealment within the meaning of s32(1)(b) and s32(2). Pending disclosure, the Claimant is not able to plead each and every example of deliberate concealment but relies on the following particulars:

52.1

The Defendants committed and procured repeated breaches of the ICA in circumstances where they would be unlikely to be discovered for some time because they were carried out by the Defendants at the Defendants’ premises by the Defendants’ staff without the Claimant’s knowledge.

52.2

The Defendants took substantial steps deliberately to conceal their wrongdoing including inter alia, (i) the formal implementation of the Winsopia Clean Room process while in reality repeatedly breaching the ICA and (ii) refusing to comply with its audit obligations under the ICA.

The RFI in question was raised on 29 November 2022, seeking responses to questions directed at establishing when the claimant became aware of facts relevant to the cause of action:

The claimant does not resist all of the requests for further information but Mr Saunders has drawn attention to the order that has already been made by Waksman J which provides that by 23 June 2023 the claimant must provide particulars of its case on deliberate concealment. He concedes that this does not include, at least in terms, a direction that the claimant should provide particulars as to the date of knowledge of the relevant facts relied on but accepts that that is something that the court could direct.

The defendants rely upon contractual and statutory limitation defences. In response to that, the claimant in its reply has raised: (i) issues as to the construction of the contract in relation to the contractual limitation period; (ii) an issue of estoppel by representation; and (iii) deliberate concealment, namely that facts relevant to its cause of action against the defendants were deliberately concealed. The claimant pleads that those relevant facts were not discovered, and could not have reasonably have been discovered, until shortly before the claim form was issued, outside the six-year limitation period that might otherwise apply.

The defendants are entitled to understand the nature of the claimant's case, both as to the facts that it alleges were deliberately concealed, and also as to the date on which the claimant asserts that it had knowledge of those facts, so as to enable it to commence these proceedings.

It may well be that in answer to some of the requests, the claimant can do no more than provide the nature of its case; but in relation to some of the specific allegations that have been identified in the pleading, it must be able to identify when it became aware of the relevant factors; for example when it became aware of its allegation that the clean room procedures were a sham and when it became aware of alleged direct access by individuals to the IBM mainframes containing the IBM software.

Therefore, those particulars need to be set out by the claimant. The question that then arises is by what date. Given that Waksman J has already considered the issue of deliberately concealment and has ordered that proper particulars be provided by 23 June 2023, the sensible course of action is for the claimant to provide one set of particulars or further information that deal with both the allegation of deliberate concealment and also with the date of knowledge.

Therefore, I will order that the further and better particulars ordered by Waksman J should include a response to the following parts of the RFI dated 29 November 2022:

That will enable the defendants then to understand the nature of the case as a whole on deliberate concealment that they have to meet.

This is the defendants’ application for further information in relation to the technical case pleaded by the claimant. The material requests are requests 22, 27, 32, 36, 37, 38, 55, 58.2, 58.3, 58.4, 63, 65, 66, 68, 69, 72, 73 and 75 of the defendants’ Part 18 RFI dated 12 November 2021 set out in Annex 1 to the 11th witness statement of Ms Scott.

The court orders that by 23 June 2023 the claimants shall respond to those requests for information:

The reasons for that order are as follows.

The defendants have identified specific allegations that are currently part of the claimant's pleaded case. Those allegations include allegations of a technical nature relating to the claimant's case that various aspects of the claimant's software have been used by the defendants in such a way that it is inferred that they must have been re-engineered. The defendants are entitled to ask for further information in relation to those detailed pleaded allegations and the requests for information on their face appear to be proper requests for information. That appears to be accepted by Mr Lavy KC, leading counsel for the claimant.

The issue that has arisen is whether some of the requests are now unnecessary, either because they have been overtaken by events, or whether they have already been dealt with by agreement or by orders of the court.

Without going through an audit trail of each request and response, the court notes that the original response by the claimant to many of these requests was that pending disclosure by the defendants of the source code to the SDM, the claimant was unable to provide the information. That was an adequate response at the time but the claimant now has the source code to the SDM. The defendants are entitled to know the nature of the case made against them. So, therefore, the starting point is that the defendants are entitled to that information.

In terms of the timing of the further information, again the starting point is that Waksman J has already ordered the claimant to provide further and better particulars of its case by 23 June 2023. This court has already ordered that further particulars should be provided in relation to the case of deliberate concealment by that date. Therefore, that would be an appropriate date for the further information to be provided by the claimant.

I acknowledge the claimant's concern that it is likely to change and/or amplify and/or refine its technical case following consultation with its experts and review of disclosure, including disclosure of the source code by the defendants. I will certainly hear the parties on general amendments to the timetable that are likely to be required.

Leaving that aside, the claimant must know at this stage whether it is intending to pursue these technical allegations; if it is, it should be in a position by the end of June, to provide a response. That does not mean that the claimant would be shut out from seeking to amend that technical case at a later date, but I am concerned that the defendants are entitled to know the case against them at this point in time because otherwise the case cannot move forward.

For those reasons I will order a response to the requests to be answered by 23 June 2023.

This is the claimant’s application for further information under CPR Pt 18 pursuant to request dated 22 November 2022.

The claimant's pleaded case is that in breach of its obligations under the ICA, the second defendant used the IBM mainframe software or parts thereof and/or permitted the same to be used for the purposes of inter alia development and/or operation by the first defendant of the SDM and/or providing to the first defendant or persons acting on its behalf information obtained by decompiling, dissembling or other reverse engineering such software.

The defendants deny those allegations. At paragraph 18 of the Amended Defence and Counterclaim it is pleaded that on 4 December 2013 LzLabs and Winsopia entered into an agreement for the provision of services by Winsopia, including at paragraph 18.3:

Prior to June 2015, software development services in respect of an “agent” program, which was intended to run on a customer's IBM mainframe and to communicate with a compatibility layer network appliance developed by LzLabs, but which was not ultimately taken forward.”

At paragraph 65 of the pleading it is specifically alleged that the second defendant, Winsopia, did not permit the ICA programs to be used by the first defendant or any persons acting on its behalf to any extent at all. Further, it is alleged that the second defendant did not use or authorise the use of the materials identified by the claimant other than within the customer's enterprise as defined in the ICA.

There matters might have rested, but Ms Scott stated in paragraph 59 of Annex 2 to her seventh witness statement dated 17 October 2022 as follows:

“Between 2014 and 2015, the second defendant provided software development services to the first defendant in respect of an agent program to be run on the designated machine for testing purposes. It would allow communication with a terminal which could be accessed in a premised basis remotely via VPN by the first defendant's personnel. This terminal was known as 'the appliance'.”

The information the claimant seeks is in relation to that statement, which is not necessarily inconsistent with the Amended Defence but in respect of which they seek the following further information:

In my judgment, this is information which is the proper subject of a request for further information. It arises out of matters pleaded in the Amended Defence in response to a direct allegation made by the claimant. Therefore it relates to an issue in the case. The detailed requests arise out of the witness statement of Ms Scott which provide further flesh on the bones of the pleaded case in the Amended Defence relating to the agent program.

It may well be the case that it turns out to be a ‘red herring’ in that it relates to a period between 2014-2015 and the defendants' case is that this development was simply not taken forward. Nonetheless, the claimant is entitled to have a response to the requests made in order to provide clarity as to the defendants' position.

In line with the other orders that I have already made today, I will order that this response be provided by 23 June 2023.

This is the defendants’ application dated 29 November 2022, seeking to restrain the claimant from making any further reference to proceedings before the Western District of Texas between Neon Enterprises Software LLC and IBM Corporation, case number 1:09-CV-00896 AWA (the “Neon Proceedings”) in any witness statement, skeleton argument or other submission in these proceedings.

The background to this issue is that there were separate proceedings, the Neon proceedings, which took place in Texas against IBM Corporation. The allegations in the proceedings included allegations by IBM Corp that Neon had unlawfully reverse engineered IBM code, on the instruction of John Moores, the ultimate beneficial owner of Neon and Winsopia, the second defendant in these proceedings.

The US proceedings were settled by way of a confidential settlement. Mr Moores agreed to an injunction but the US court made no findings of any wrongdoing and no final determination of the claims.

In the particulars of claim as initially drafted by the claimant in this case, the claimant referred to the Neon proceedings and to the injunction that was granted against Mr Moores and various other parties. Paragraphs 11 and 12 of the pleading stated that although the claimant did not have any knowledge of the extent of involvement of Mr Moores and other persons subject to the terms of the injunction, it reserved the right to seek to join Mr Moores (and those other persons subject to the injunction) to these proceedings in the courts of England and Wales.

That pleaded case, at paragraphs 11 and 12 of the particulars of claim, was struck out by Eyre J. There was no appeal against that part of his judgment and indeed those paragraphs 11 and 12 have been deleted from the current claim. Therefore as things stand the claimant's pleaded case makes no allegation against Mr Moores or others subject to the injunction in the Texas courts.

The issue has arisen because in a CMC skeleton produced by the claimant last October 2022 reference was made to the Neon proceedings in the context of disclosure that was being sought by the claimant from the defendants. Documents from the Neon proceedings were included in the bundle for that CMC. It is those references in the skeleton and those documents with which the defendants take issue.

Mr Stewart submits that the ongoing references to the Neon proceedings and Mr Moores, and the inclusion of documents in the bundles put before the courts are prejudicial and inappropriate in circumstances where the relevant allegations in relation to the Neon proceedings have been struck out.

The claimant opposes the application. Mr Lavy submits that the court has no legal basis for making the order. He accepts that there is jurisdiction for the court to make various civil restraint orders, but submits that it would not be appropriate in this case. Further, he submits that there is nothing vexatious in the claimant's actions to date; in particular, the references in the skeleton and bundles for the October 2022 CMC were for proper purposes of considering disclosure. Finally, he submits that it would be wrong to restrain the claimant so as to fetter the right of the claimant's counsel to pursue the case as they see fit, including through skeleton arguments and oral submissions.

The starting point is that the court has very wide powers to control evidence that is placed before it, including a power to determine what evidence should be admitted. Even if evidence is in principle admissible, the court is still in a position to exclude it from any part of the proceedings if not necessary for the purposes of determining the issues before it. Likewise, the court has very wide powers to preclude a party or a legal representative from making applications in any particular set of proceedings or indeed, more widely, to make any application or bring any proceedings in front of this or indeed any other court.

However, as Mr Lavy quite properly pointed out, those orders are usually only made in extreme cases and the facts in this case do not support the making of such a wide order. In my judgment the relief that is currently sought by the defendants goes too far and is not justified on the current facts.

Firstly, the claimant is not seeking to make any allegation in its pleaded case that would include any reference to the Neon proceedings. Therefore there is no basis for any order that prevents it from doing so. If any application is made in the future, the court will consider it on its merits.

Secondly, the claimant is not currently seeking to rely on any witness statement that makes any reference to the Neon proceedings. Again, if at any stage such a witness statement is put forward, it would be open to the defendants to seek to strike out that part on the grounds that it offends the Practice Direction PD 57AC, or trespasses on the issues struck out by Eyre J, or indeed on any other grounds. However, currently there is no application to adduce the evidence of such a witness statement and the issue does not arise.

Thirdly, there is no application by the claimant to reply upon the Neon proceeding documents. It may well be that there comes a time when the claimant does seek to do that. If so, the court would hear full argument on whether or not those documents contained any evidence that was of probative value or were otherwise admissible documents in this litigation. But currently there is no attempt by the claimant to rely upon those documents or to admit them as evidence.

Therefore the premise on which the defendants seek the injunction is not established. The court can indicate that the mere fact that an individual might have been involved in a reverse engineering case in another jurisdiction against another party of itself is not likely to be of any probative value. However, the court does not consider that it would be appropriate to tie the hands of either party in this case against making any applications that might raise the issues that were raised in the Neon Proceedings in the future.

For all of those reasons, the court refuses the defendants’ application for the injunction.