[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales Court of Appeal (Civil Division) Decisions
You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> IPcom GmbH & Co Kg v HTC Europe Co Ltd & Ors [2013] EWCA Civ 1496 (21 November 2013) URL: http://www.bailii.org/ew/cases/EWCA/Civ/2013/1496.html Cite as: [2013] EWCA Civ 1496, [2014] RPC 12, [2013] WLR(D) 456, [2014] Bus LR 187

[New search] [Printable RTF version] [View ICLR summary: [2013] WLR(D) 456] [Buy ICLR report: [2014] Bus LR 187] [Help]

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
PATENTS COURT
Mr Justice Roth
[2013] EWHC 2880 (Ch)

		Royal Courts of Justice Strand, London, WC2A 2LL
		21/11/2013

Lord Justice Floyd:

1. This appeal concerns the circumstances in which it is appropriate for a court in this jurisdiction considering combined patent infringement and revocation proceedings to grant a stay of its proceedings pending the outcome of co-pending opposition proceedings in the European Patent Office ("EPO"). By his order of 13 June 2013 Roth J, sitting in the Patents Court, declined to order such a stay on the application of the defendants in the action (collectively "HTC"). The stay was resisted by the claimant patentee, IPCom. HTC appeal, with the permission of the judge. We heard the appeal as a matter of urgency because the trial of the action is fixed to commence in the Patents Court in an eight day window commencing on 9 December 2013.
2. We were assisted by the able written and oral submissions of Mr Adrian Speck QC and Mr James Abrahams for HTC and Mr Henry Carr QC, Mr Brian Nicholson and Ms Kathryn Pickard for IPCom.
3. In reaching his decision to refuse a stay, Roth J was referred to and relied on the guidance given by this court in Glaxo Group Ltd v Genentech Inc [2008] EWCA Civ 23; [2008] FSR 18 ("the Glaxo guidance"). After the judge had announced his decision, but before he had given his reasons, the Supreme Court handed down its judgment in Virgin Atlantic Airways Ltd v Zodiac Seats UK Ltd [2013] UKSC 46; [2013] 3 WLR 299 ("Virgin"). In the course of their judgments in that case, Lord Sumption JSC and Lord Neuberger PSC both questioned the correctness of the Glaxo guidance. Those observations were drawn to the judge's attention and formed the basis of his decision to give permission to appeal to this court. In the course of this appeal, therefore, we heard submissions on whether the Glaxo guidance needed revision in the light of the Supreme Court's observations. Permission to appeal to this court would not normally be given where the question at issue involves a discretionary case management power, such as the grant or refusal of a stay, when the judgment involves no error of principle or is not plainly wrong.
4. The appeal arises in a case where IPCom, a patentee which does not itself manufacture or market products in accordance with its patent, seeks to cause HTC to enter into a licence rather than, on IPCom's case, to continue infringing by selling its mobile telephones. It is accepted that this is a case where no injunction will be granted: the patent is said by IPCom to be essential to the UMTS mobile telephony standard, and IPCom have undertaken to grant licences on fair, reasonable and non-discriminatory ("FRAND") terms. HTC characterises the case as just being about money, whereas IPCom maintains that its whole business depends on the exploitation of its patents, and a stay would prevent it moving towards a position where it can enforce them.

The procedural history

5. There is a complicated procedural history to this case which involves not only the present action ("the HTC action"), but proceedings in relation to the same patent between Nokia and IPCom ("the Nokia action"). The two actions have, to an extent, become entwined. The essentials are as follows.
6. EP (UK) 1,841,268 ("the patent") was granted on 17 March 2010. It relates to a method of controlling access to the random access channel or RACH in a wireless telecommunications network. During the nine month period allowed for opposition in the EPO, four companies launched oppositions, including Nokia on 18 August 2010 and HTC on 17 December 2010, the last day of the period allowed.
7. In addition to its opposition in the EPO, on 9 April 2010 Nokia commenced the Nokia action in the Patents Court. The Nokia action was an action for revocation of the patent. IPCom denied that the patent was invalid and counterclaimed for infringement. The pleadings also raised non-technical issues concerned with IPCom's obligation to grant licences on FRAND terms arising out of the inclusion of the patented technology in the UMTS standard. By an order made on 7 July 2010 the technical patent issues of validity and infringement were ordered to be tried first, with a trial on non-technical issues to follow if necessary. I will refer to this deferred trial as "the FRAND trial".
8. The technical patent issues in the Nokia action came to trial before me in April 2011. In my judgment dated 16 June 2011 I held the patent valid (subject to an amendment to the claims which I allowed). I also held that the patent was infringed by some Nokia devices. Certain other Nokia devices the subject of applications for declaratory relief were held not to infringe. I also held that the patent was essential to the UMTS standard. Although the parties have adopted a different nomenclature, I will refer to the claims held valid and infringed by Nokia as "the June 2011 claims". My June 2011 judgment was upheld by this Court on 10 May 2012. The Supreme Court subsequently refused permission to appeal. In July 2011 I gave procedural directions for the FRAND trial in the Nokia action. As will be seen, HTC later joined in to the proceedings to determine the FRAND royalty.
9. In the light of their success against Nokia, IPCom wrote to HTC inviting them to take a licence under the patent. They enclosed proceedings for infringement and asked whether HTC's solicitors were instructed to accept service. They were not, and so steps were taken to serve the proceedings out of the jurisdiction on HTC Corporation, the third defendant. The dispute about service out of the jurisdiction was compromised in April 2012. In August 2011 HTC attempted the so-called "Italian torpedo". This was a procedural step designed to obtain a stay of all proceedings in this country on the basis that the courts of Italy are first seised. The idea was to slow the litigation down to the pace of one of the slowest jurisdictions in Europe. For a full description of the Italian torpedo see Research in Motion (UK) Limited v Visto Corp [2008] EWCA Civ 153; [2008] FSR 20 at [12] to [14]. The Italian torpedo was eventually withdrawn, and on 20 January 2012 directions were given for trial of the technical issues in the HTC action. HTC were of course not bound by the decision in the Nokia trial.
10. In the meantime the EPO opposition proceedings had been continuing. On 25 April 2012 the Opposition Division ("OD") of the EPO decided that the patent was invalid for added matter. In accordance with EPO practice, none of the other objections to validity were considered. IPCom launched an appeal from that decision to the Technical Board of Appeal ("TBA").
11. On 11 May 2012 HTC applied to stay all aspects of the HTC action until the TBA handed down its written decision in the appeal.
12. On 30 May 2012 Roth J made an order in both the Nokia and HTC proceedings embodying agreed directions pursuant to which both HTC and Nokia were to participate in the FRAND trial. That order was made notwithstanding the fact that there had as yet been no finding either as to validity or infringement of the patent as between HTC and IPCom. HTC were understandably interested in participating in a trial which would enable them to know the cost of the licence to which they would be entitled under the patent if it was otherwise valid and infringed by them.
13. In addition to allowing the FRAND trial to go forward on this consensual basis, Roth J dealt with HTC's stay application. He granted a stay of the technical issues in the HTC action pending the TBA's decision in the opposition proceedings. HTC undertook that it would take a FRAND licence, and in the meantime restrict its commercial activities to one of the variants declared not to infringe in the Nokia action ("approved variants").
14. Despite the stay, matters did not stand still in relation to technical issues in the Patents Court. HTC was directed to produce product and process descriptions, and further disclosure was ordered of source codes. This was rendered necessary because of uncertainty over precisely what methods of controlling RACH access HTC was using. In order to prevent matters drifting, Norris J fixed the trial of the technical issues for a window commencing in December 2013 as by that stage it was known that the TBA's hearing would take place in March 2013. It is the normal practice of the TBA to announce its decision at the conclusion of the hearing.
15. On 7 March 2013 the TBA heard IPCom's appeal against the OD's decision revoking the patent. The result of the appeal was that the decision of the OD was set aside. The added matter objection was held to be overcome by an amendment put forward by IPCom. The TBA remitted the matter to the OD for consideration of the remaining issues: lack of novelty and lack of inventive step, issues which neither the TBA nor the OD had thus far addressed. The amended claims (which I will call "the March 2013 claims") were not the same as the June 2011 claims. They contained an additional feature. On 21 May IPCom applied in the Patents Court to amend the claims of the patent to bring them into line with the March 2013 claims.
16. The stay of the HTC action granted by Roth J came to an end with the handing down of the TBA decision. It had lasted about 9 months.
17. In the light of the TBA's decision, HTC next applied (on 23 April 2013) for an adjournment of the FRAND trial fixed for July 2013. Mann J granted that adjournment on 2 May 2013.
18. On 20 May 2013 HTC applied to adjourn the technical trial in the HTC action, now fixed for December 2013, and to stay all the technical issues until a final decision of the EPO on the remitted opposition proceedings, including any appeal.
19. Those applications (amongst others) came before Roth J on 12 June 2013. He refused HTC's stay application. The effect of that order was that the technical trial in the HTC action would proceed in December. In the light of that order the parties have continued to prepare for the technical trial. HTC contend that the judge was wrong to refuse the stay, and appeal from the judge's order.

The litigation system created by the EPC

20. The relevant background is the European Patent Convention ("the EPC"), an international treaty, and the Patents Act 1977 ("the 1977 Act"). The EPC contains provisions which enable any person to apply for revocation of a granted patent. Thus under Article 138 of the EPC a European Patent may be revoked by a national court of a Contracting State (e.g. in proceedings by way of direct action for revocation or counterclaim to an infringement action). Revocation in such circumstances will apply only to the designation of the Patent in that Contracting State.
21. A party who contends that a granted European Patent is invalid, and should be revoked, can also do so by means of the opposition procedure in the EPO. An opposition in the EPO must be launched within 9 months of grant, but there is no provision for when such proceedings must end. Experience has shown that opposition proceedings before the EPO are likely to take many years. There are numerous reasons for this. One reason is the EPO's serial approach to decision making. Where an opponent raises a variety of grounds of opposition, the practice of the EPO is to decide the issues in a given order. If, on consideration, a ground succeeds, it is rare for the OD to go on and decide the other issues in the sequence, as a first instance court in this country might consider itself obliged to do. As a result, if the OD's decision is overturned on appeal, the TBA will normally have to remit the case to the OD for it to consider the next issue. And so on. Some of that has happened in this case.
22. The EPO has been a victim of its own success. In the normal run of things an appeal from the OD to the TBA may take four years. Mr Carr showed us Cubist Pharmaceuticals Inc Decision T0612/09, a decision of the TBA on an appeal against a decision of the OD dated 19 January 2009 revoking a patent, some two years after the opposition proceedings had been commenced. The Board's decision remitting the case to the OD is dated 11 April 2013, more than four years later, and six years from the commencement of the proceedings. The OD's attempt to decide more issues than necessary were dismissed as obiter dicta by the TBA. The Board's discussion of whether it should remit the case in these circumstances is revealing:

"25. As already mentioned, the present appeal was filed in January 2009. The oral proceedings were held on 10 April 2013 and the decision issued on 11 April 2013. The appeal proceedings thus lasted a little over four years. While it gives the board no pleasure to say so, four years is currently the average time taken to dispose of the appeals in its list of pending cases.

26. Against that background the board, in considering the respondents' argument that better justice would be done by the board itself reconvening at a later point in time, has to ask when that point in time would, or should, be? Quite clearly it could not, at the very earliest, be before any of the board's currently scheduled oral proceedings in other cases, and thus not in 2013. If, on the one hand, it should be before oral proceedings have taken place in all, or some, of the other cases in the board's list of pending cases, then clearly there would be a possible argument that the board was giving unfair preference to this case, in which oral proceedings on the issue giving rise to the appeal have already taken place and that issue has been decided, over other cases which have also been pending for four years and in which oral proceedings have not yet taken place. If, on the other hand, the board were now to treat this case as a newly-filed appeal for the purpose of the undecided issues, the parties might, indeed in the currently prevailing conditions probably would, have to wait another four years for a final decision. In the board's opinion, neither of those solutions would lead to better justice, and certainly not necessarily to an earlier final decision, as the respondents argued.

27. The board understands that currently the time taken to dispose of opposition proceedings is (as happened in the present case) about two years so assuming that, after a remittal, the opposition division gives this case no particular preference, the parties would have a decision in half the time they might have to wait for a decision from the board. Of course, if one or more parties were then to appeal, a further long wait for a final decision might then ensue - just how long would depend on the length of the board's list of pending cases at that future point in time. But, in the absence of a further appeal, the likelihood must be that the parties will achieve a final decision sooner if there is a remittal and that is significant. Thus the board considers, not without regret, that the question of delay and the possible injustice delay may cause points, on balance, in favour of remittal."

23. A procedure which allows disputes over patent rights to take in excess of a decade cannot meet the needs of industry, particularly in rapidly moving areas of technology. Although such a procedure may technically comply with Article 6 of the European Convention on Human Rights, which guarantees a trial before an independent tribunal within a reasonable time, the opportunity for successive appeals and remittals means that there is in practice no final determination of the parties' rights for many years. Given their procedures, the Boards have a difficult task in seeking to achieve justice and finality.
24. Although the ODs and TBAs have shown sensitivity to acceleration of their proceedings in cases of commercial urgency, the serial nature of their procedures is not well adapted to arriving at early decisions even in cases where expedition is granted. The present case is an illustration of that fact. Even with expedition, the OD and TBA in this case have resolved only one of the issues in the oppositions since the opposition period closed on 17 December 2010. When the OD sits again on the currently predicted date of 22 Jan 2014 to consider lack of novelty and inventive step, the proceedings will have passed their third anniversary. An expedited appeal might take another year. By contrast, all the issues of infringement and validity in the Nokia action were considered by two instances in this jurisdiction in just over two years from service of the claim form.
25. Whilst the delays which occur in the opposition procedure of the EPO cannot be described as part of the architecture of the system, they form an important part of the background to any application for a stay of national proceedings pending the outcome of proceedings in the EPO.
26. It is inherent in the scheme provided by the EPC and the Act that the twin routes to revocation of a patent may be pursued at the same time. This is visible in a number of places in the Act. Thus section 77(3) provides that:

"(3) Where in the case of a European patent (UK) -

(a) proceedings for infringement … have been commenced before the court or the comptroller and have not been finally disposed of, and

(b) it is established in proceedings before the European Patent Office that the patent is only partially valid,

the provisions of section 63 or, as the case may be, of subsections (7) to (9) of section 58 apply as they apply to proceedings in which the validity of a patent is put in issue and in which it is found that the patent is only partially valid."

27. That section recognises that a decision of the EPO may be made at a time when infringement proceedings are before the national court, and provides that the effect of any amendment to the patent made by the EPO is the same as if the amendment had been ordered by the court.
28. It is important to keep in mind the range of possible consequences of the two sets of proceedings, and their impact on one another. It is these features of the system created by the EPC which make the considerations affecting the grant of stays in patent cases different from those in other types of case: see in this connection the observations of Lewison J in Glaxo v Genentech [2007] EWHC 1416 (Pat); [2007] F.S.R. 35 at [37] to [38]. Thus, in the EPO the patent can be revoked, maintained in an amended form or maintained as granted. Whilst revocation operates in rem, a finding that the patent is valid as granted in opposition proceedings does not operate as an issue estoppel between the parties thereto: see Buehler v Chronos Richardson [1998] RPC 609. Moreover, the EPO has no jurisdiction over the issue of infringement, which is within the exclusive jurisdiction of the national court. The jurisdiction of the national court is, however, territorial both in relation to infringement and in relation to validity. A decision by the national court to revoke a patent with effect for its own jurisdiction cannot be affected by a later decision of the EPO. The same is true of a finding of non-infringement.
29. One way in which the two jurisdictions interact can be seen in the example of the Nokia action. The English courts decided that the patent was valid with the June 2011 claims, infringed by certain Nokia devices and not infringed by the approved variants. The subsequent narrowing of the claims by the EPO to the March 2013 claims will leave many of the findings of fact and conclusions of law reached in the English judgments untouched. For example, the approved variants will still be outside the March 2013 claims, which are narrower than the June 2011 claims. The obviousness attacks against the patent, which failed in England, are unlikely to fare better against narrower claims. There will of course be new issues raised by the March 2013 claims, but these should be limited to whether the new feature adds subject matter, and whether the amendment takes the Nokia infringing devices outside the claim. These are relatively narrow further issues compared to the plethora of issues decided in the Nokia action. They may not even arise. There was a tendency in the submissions of Mr Speck for HTC to regard the fact of amendment by the EPO after an English judgment as throwing away the cost and expenditure of the English trial. It does not. There may be some additional expenditure caused by the amendment in the concurrent proceedings, but that is a consequence of the system, and is inherent in it.

Stays

30. Neither the EPC nor the Act contains any provision providing for a stay whilst the validity or final form of the patent is established in proceedings in the EPO. The practice in relation to the grant of stays pending the outcome of EPO proceedings is not uniform amongst member states of the EPC. In Germany, where the practice is to bifurcate the issues of infringement and validity, we were told that the courts will not stay infringement proceedings merely because there is a co-pending opposition in the EPO. The "infringement court" would apply a merits-based test to the validity attack, deciding whether there was a high likelihood that the patent would be revoked. On the other hand the national "validity court" will always stay proceedings where there are co-pending opposition proceedings at the EPO. In the light of what I have said about the speed of opposition proceedings in the EPO, under the German practice it seems it could take a defendant many years to overturn a decision that he had infringed a patent by establishing that it is invalid.
31. We were also shown Article 33(10) of the as yet un-ratified recent Agreement for the Unified Patent Court. Article 33(10) of the Agreement provides, in relation to infringement and validity proceedings, that:

"A party shall inform the Court of any pending revocation, limitation or opposition proceedings before the European Patent Office, and of any request for accelerated processing before the European Patent Office. The Court may stay its proceedings when a rapid decision may be expected from the European Patent Office."

32. In the light of what I have said thus far, one can understand why the contracting states thought that the discretion to stay proceedings should only arise if an EPO decision is expected rapidly.
33. It is common ground that the Patents Court enjoys a jurisdiction to stay its proceedings pending the outcome of co-pending EPO proceedings. The jurisdiction is recognised in section 49(3) of the Senior Courts Act 1981:

"Nothing in this Act shall affect the power of the Court of Appeal or the High Court to stay any proceedings before it, where it thinks fit to do so ..."

34. The jurisdiction is obviously a very wide and unfettered one, and must be exercised having regard to all relevant circumstances.

Domestic Case law

35. The cases prior to Glaxo v Genentech are comprehensively explained in the decision of Lewison J (as he then was) at first instance in that case: [2007] EWHC 1416 (Pat) at [23] to [36]. The principal observations in the cases were these:

"The judge, as I see it, had to exercise his discretion whether or not it is in the interests of justice that the English proceedings should be stayed until the decision of the European Patent Office by balancing the various procedural considerations which may result from either course. The fundamental one, as I see it, is the length of time the proceedings in the European Patent Office are likely to take." per Dillon LJ in Pall Corp v Commercial Hydraulic (Bedford) Ltd [1989] RPC 703 at 711.

"… the Patents Court will stay the English proceedings pending a final resolution of the European proceedings, if they can be resolved quickly and a stay will not inflict injustice on a party or be against the public interest. Unfortunately that is not always possible as a resolution of opposition proceedings in the EPO takes from 4-8 years" per Aldous LJ in Beloit Technologies Inc. v Valmet Paper Macinery Inc [1997] RPC 489 at 503.

"… absent any other consideration, a stay is the appropriate course to adopt. But when there are other considerations, it is for the court to weigh up the pros and cons and see where the justice of the situation lies." per Laddie J in Hunt Techology Ltd v Don & Low Ltd [2005] EWHC 376 at [7].

"Of course in principle the preferred option is to stay UK proceedings if there are corresponding EPO proceedings. And it may in some circumstances be the case that an interim injunction could serve to hold the fort whilst these proceed. But all must depend on the circumstances and particularly the timing. Normally, although a stay is in principle the preferred course, it would be wrong to prevent the patentee from enforcing his patent here if the EPO opposition will not be concluded reasonably soon – as all too often it sadly is not. Take this case: the action started here in May 2002 and was finally over by November 2005. The EPO proceedings are still running and could be still doing so at the end of next year. Business needs to know where it stands – and a patentee is entitled to enforce his patent without undergoing the risks inherent on the cross-undertaking in damages – especially if the period involved could involve years." Per Jacob LJ in Unilin Beheer BV v Berry Floor and others [2007] EWCA Civ 364; [2008] 1 All ER 156. ("Unilin") at [25].

36. To these I would add what Lewison J himself said in Glaxo:

"Equally I do not accept Mr Alexander's submission that there is an independent condition to be satisfied, namely that the stay must be a short one (how short is short?). I have no doubt, however, that the length of any stay and any prejudice that would flow from it is a highly material (sometimes the most material) factor. In my judgment the correct approach for me to adopt is that there is a presumption, although not a strong one, in favour of a stay, and that it is for the party resisting the stay to overcome the presumption. In the end, I must decide where the balance of justice lies."

37. Glaxo v Genentech was a case in which there was no allegation of infringement. Glaxo, and its partner Genmab, had brought proceedings for revocation in both the EPO and the Patents Court. They had a commercial product and wished to clear the way for its introduction onto the market. Lewison J refused a stay, and his decision was upheld on appeal. In this context, the Court of Appeal, (Ward, Mummery and Jacob LJJ) gave some general guidance on the Patents Court's discretion to grant stays on the grounds that there are parallel proceedings pending in the EPO contesting the validity of the patent. They made clear that the guidance:

"is not intended to fetter the discretion of the court nor should it be interpreted as having that effect."

38. I should now set out the guidance:

"First, the discretion, which is very wide indeed, should be exercised to achieve the balance of justice between the parties having regard to all the relevant circumstances of the particular case.

Secondly, it is the discretion of the Patents Court, not of the Court of Appeal. The Court of Appeal would not be justified in interfering with a first instance decision that accords with legal principle and has been reached by taking into account all the relevant, and only the relevant, circumstances.

Thirdly, although neither the EPC nor the 1977 Act contains express provisions relating to automatic or discretionary stay of proceedings in national courts, they provide the context and condition the exercise of the discretion.

Fourthly, the possibility of the duplication of proceedings contesting the validity of a patent granted by the EPO is inherent in the system established by the EPC. In practice national courts exercise exclusive jurisdiction on infringement issues and they have concurrent jurisdiction with the EPO on validity issues. As Mr Daniel Alexander QC appearing for GSK said, the Contracting States and the UK Parliament contemplated that the national Patents Courts should be able to determine the same issues of patentability as the EPO. The resultant legislation allowed the determination by the national court and the EPO to proceed at the same time. Indeed, there is nothing in the EPC or the 1977 Act to prevent the commencement of revocation proceedings in the Patents Court on the very date of the grant of the patent by the EPO.

Fifthly, this setting indicates that, in present conditions, one factor affecting the discretion will usually carry more weight than any other. That is the length of time that it will take for the respective proceedings in the national court and in the EPO to achieve some certainty on the issue of the validity of the patent in suit so that business knows where it stands. The length of the stay of proceedings, if granted, is, in general, the most significant factor in the discretion. Both the parties' legitimate interests and the public interest are in dispelling the uncertainty surrounding the validity of the monopoly rights conferred by the grant of a patent and the existence or non-existence of exclusive proprietary rights on a public register. A decision in the revocation action in the Patents Court will dispel some of the uncertainty. If the likelihood is that proceedings in the Patents Court would achieve this resolution significantly sooner than the proceedings in the EPO, it would normally be a proper exercise of discretion to decline to stay the Patents Court proceedings. They should be allowed to proceed to a decision that would supply some certainty in the public interest and the parties' legitimate interests.

Sixthly, there are no grounds justifying the application by the Patents Court of a presumption that the duplication of legal proceedings in it and in the EPO is, without more, a ground for a stay of the proceedings in the Patents Court, as the EPC system allows for parallel proceedings contesting the validity of the patent in both the international court (which is what the EPO in substance is) and in the national court.

Seventhly, the Patents Court judge is entitled to refuse a stay of the national proceedings where, as here, the evidence is that some commercial certainty would be achieved at a considerably earlier date in the case of the UK proceedings than in the EPO. It is true that it will not be possible to attain certainty everywhere until the EPO proceedings are finally resolved, but some certainty, sooner rather than later, and somewhere, such as in the UK, rather than nowhere, is, in general, preferable to continuing uncertainty everywhere. Thus, in this case some degree of commercial certainty could be achieved at the trial of the UK revocation action in February without unfairly prejudicing Genentech's legitimate interests in the protection of its patent.

Eighthly, much weight should be given to an assertion by a commercial party that it has a good reason for resisting a stay. Normally a party is the best judge of its interests. Contentions of a competitor that there is no commercial need for early resolution of validity should be viewed with suspicion. Detailed arguments of the sort advanced here are unlikely to carry weight and a judge would be justified in dealing with them shortly.

Finally, other considerations in the particular case may affect the balance of justice, such as the additional costs in the duplication of proceedings, the order in which the proceedings were commenced and so on, but, in general, the other factors, through relevant, are of lesser importance than achieving some commercial certainty somewhere sooner. The judge will receive evidence and submissions on other relevant factors, but should be wary of over-elaboration of the issues by the parties in their evidence and legal submissions. Although due consideration must, of course, be given to the evidence and the arguments, the actual exercise of the discretion does not require the judge to deliver a judgment dealing in detail with all the points taken by the parties. A global assessment of the relevant material, supported by valid reasons, is normally sufficient to justify the decision to refuse or to grant a stay."

39. Before coming to the observations made on this guidance by the Supreme Court in Virgin, it is necessary to explain how the landscape has been adjusted by the substantive decision in that case. To do that it is necessary to explain by way of background the prior decision of this court in Unilin.

Unilin

40. One problem thrown up by the existence of parallel proceedings in the national court and in the EPO is the following: can the patentee still obtain damages for infringement when the proceedings in the national court reach a conclusion and the patent is found both valid and infringed as between the parties to the action, and, later, the EPO revokes or amends the patent? This issue arose in Unilin.
41. In Unilin the claimant had successfully brought infringement proceedings against the defendant, and no further appeals lay against the judge's orders for compensation by way of an inquiry as to damages or an account of profits. The defendant applied for a stay of the subsequent inquiry or account on the grounds that opposition proceedings in the EPO had not been completed and the patent might ultimately be revoked or limited in such a way as to exclude the defendant's products. The claimant contended that the issue of infringement was res judicata and nothing which could happen in the EPO could affect the claimant's right to compensation.
42. The judge granted a stay but this court reversed his decision relying on two earlier Court of Appeal cases, Poulton v Adjustable Cover and Boiler Block Co. [1908] 2 Ch 430 and Coflexip SA v Stolt Offshore MS Ltd (No 2) [2004] EWCA Civ 213. The result was that parties to the national court judgment were estopped from contending that the patent was revoked, even when a subsequent decision of the EPO said that it had been. Jacob LJ, at [88], said that the consequence was as follows:

"It means that businessmen in this country know they can use the rather speedy court system here to get a conclusion one way or the other. If the patent is revoked, the way is cleared; if it is upheld and held infringed then compensation will be payable for past acts. And an injunction will run unless there is a later revocation by the EPO. Subject to that last point, the effect of all this is that one does not have to wait to find out who has won until the slowest horse in the race gets there."

43. Although the question of the grant of a stay arose in the context of a party wishing to proceed with an enquiry as to damages, the Unilin principle, if correct, would be of relevance at an earlier stage when considering whether to stay the steps towards a trial on liability. On the assumption that Unilin is correct, the trial on liability would be capable of delivering some certainty not only to the alleged infringer but also to the patentee. Thus a judgment that the patent is valid and infringed together with an order for an inquiry as to damages would mean that the patentee would be entitled to damages for past acts. That conclusion could not be affected by any subsequent decision of the EPO, even though such a decision might lead to a discharge of the injunction.

Virgin

44. The res judicata principle affirmed in Unilin no longer represents the law. In Virgin the Supreme Court unanimously held that the revocation or amendment of the patent in the EPO applied retrospectively, and that a party held to have infringed a valid patent could rely on the subsequent revocation of the patent as an answer to the claim for damages. This is a significant alteration to the landscape. Lord Sumption JSC expressed the conclusion in this way at [35] to [36]:

"In my opinion Poulton is no longer good law, and Coflexip was wrongly decided. It follows that Unilin was also wrongly decided because it proceeded on the premise of the law stated in Coflexip. The point with which Unilin was actually concerned, namely whether there is a different rule for European patents arising from the scheme of the relevant legislation, has been argued before us but it does not arise, because the anomaly in English law to which that point is directed does not exist. Accordingly, where judgment is given in an English court that a patent (whether English or European) is valid and infringed, and the patent is subsequently retrospectively revoked or amended (whether in England or at the EPO), the defendant is entitled to rely on the revocation or amendment on the enquiry as to damages.

Once the enquiry is concluded, different considerations will arise. There will then be a final judgment for a liquidated sum. At common law, that judgment could be challenged on the ground that the patent had later been revoked or amended only by way of appeal, and then only if an appeal is still open. I doubt whether an implied statutory right to reopen it could be derived from the scheme of the Patents Act 1977, but that is a question which will have to await a case in which it arises."

45. At an early point in his judgment, at [7], Lord Sumption JSC made some observations on the interaction between the EPO's revocation procedure and the practice of the Patents Court in accordance with the Glaxo guidance:

"The effect of these provisions is that the English courts have the same jurisdiction to determine questions of validity and infringement in the case of a European patent as they have for domestic patents, but that concurrent jurisdiction over questions of validity is exercisable by the EPO. There is, however, an important difference between the legal effect of a decision in the two jurisdictions. Both are decisions in rem. They determine the validity of the patent not only as between the parties to the proceedings, but generally. But the English court's jurisdiction over the question of validity is purely national. A decision of an English court declaring a patent invalid, or (which will normally follow) revoking it, will have effect in the United Kingdom only, whereas a corresponding decision of the EPO, which was the authority by which the patent was granted, will have effect in all the states for which the patent was granted. These considerations make it highly desirable to avoid inconsistent decisions if it can be done. National procedures for achieving this differ from one contracting state to another. In England, there are established procedures for staying English proceedings in which the validity of a European patent is in issue, if there are concurrent opposition procedures in the EPO. However the value of these procedures is somewhat diminished by the current practice of the High Court, which is based on dicta of Jacob LJ in [Unilin], at para 25 and on the subsequent decision of the Court of Appeal in Glaxo Group Ltd v Genentech Inc ... Their effect is that the primary consideration on an application for a stay is the probable duration of the proceedings before the two tribunals. If the question of validity would be likely to be resolved quicker in the English court than in the EPO, it would normally be appropriate not to stay the English proceedings. The consequences of this practice are particularly serious in a case where the English courts "win" the race to judgment if, as the Court of Appeal decided in Unilin, the effect is to bind the parties to a decision of the English court that the patent is valid notwithstanding that the EPO which granted it subsequently decides that it should be revoked or amended ab initio."

46. Lord Sumption returned to this topic at the end of his judgment. At [38] he said this:

"I add a brief observation on the procedural implications. If I had concluded that the defendant was estopped from relying on the revocation or amendment of the patent once the court had adjudged it to be valid, that would have had important implications for the question whether English proceedings should be stayed pending a decision in concurrent opposition proceedings in the EPO. On that footing, it would in my opinion have been essential to stay the English proceedings so that the decision of the EPO would not be rendered nugatory by the operation of the law of res judicata. On that hypothesis, it would have been difficult to defend the guidance given by the Court of Appeal in Glaxo Group Ltd v Genentech Inc … to the effect that the English court should normally refuse a stay of its own proceedings if it would be likely to resolve the question of validity significantly earlier. The effect of that guidance is to put more litigants in the impossible situation in which successive decisions of the Court of Appeal placed the parties in this case. As it is, the problem has not gone away, even on the footing that those decisions are overruled. In the first place, a similar problem may well arise if the patent is revoked by the EPO after a judgment has been given for a liquidated sum. Second, that problem is aggravated by the fact that a decision of the English court on validity is directly effective only in the United Kingdom, whereas the EPO's decision, being the decision of the authority which granted the patent, is directly effective in every country for which the patent was granted. Third, even if the EPO opposition proceedings are concluded in time to affect the English proceedings, the uncertainty and waste of costs involved do little credit to our procedures. This is not a suitable occasion, nor is the Supreme Court the appropriate tribunal to review the guidance, but I think that they should be re-examined by the Patents Court and the Court of Appeal."

47. Lord Neuberger PSC at [69] agreed with what Lord Sumption had said about Glaxo v Genentech.

Glaxo in the light of Virgin

48. The judgments of the Supreme Court in Virgin invite the Patents Court or this court to review the Glaxo guidance. Neither side suggested that we could not or should not do so, or that we were bound only to make such adjustments as were strictly necessitated by the judgment in Virgin. I begin by looking at the matters to which Virgin draws our attention.
49. Firstly, it is plain from paragraphs [7] and [38] of Lord Sumption's judgment that a principal concern was that the guidance should take account of the fact that the refusal of a stay should not be allowed irrevocably to deprive a party of any part of the potential benefit of pursuing proceedings in the EPO in addition to the national proceedings. If the likely effect of refusing a stay is to allow the patentee to obtain a binding judgment on infringement and validity here, obtain an order for an inquiry and thereby trump the decision of the EPO, there is potential for injustice to the alleged infringer. To some extent, however, the decision of the Supreme Court in Virgin not only identifies the illness but provides the cure. The alleged infringer will not be prevented from raising the subsequent revocation of the patent in answer to the claim for damages, at least until the court makes an order for payment of a liquidated sum. What is left is the possibility that, further into the future, the patentee will obtain a judgment for a liquidated sum. Such a judgment could not be undone except by way of appeal, and any sums actually paid over would be unlikely to be recoverable if the patent were revoked. That possibility will need to be weighed in the balance, along with any undertakings designed to eliminate its effect. If any such irrevocable benefit remains, then, depending on its nature, it may be a powerful factor in favour of some form of stay.
50. Secondly, also at [7] in Lord Sumption's judgment, it is clear that he considers that the guidelines should be designed to "avoid inconsistent decisions if this can be done". Avoiding inconsistent decisions is of course not easily achieved in a system which expressly allows for concurrent jurisdiction, a characteristic of the system which Lord Sumption expressly recognises both there and at [33]. As it is not suggested that the national court could stay the EPO, the elimination of any risk of inconsistent decisions would require a mandatory stay of all national patent proceedings, something which I am sure Lord Sumption would have said if he had meant. A mandatory stay of all national validity proceedings pending the resolution of validity in the EPO would mark the end of the ability of the national court to give judgments "clearing the way" for those who have products to launch and businesses to conduct in this jurisdiction, provided only that the EPO proceedings had not reached their conclusion. It would also make a significant hole in the ability of a party to enforce a patent for a very large proportion of its life.
51. I do not believe that Lord Sumption was suggesting such a radical change. I read Lord Sumption's observations as taking on board and accepting the difficulty of avoiding inconsistent decisions in the system created by the EPC. If the risk can be eliminated, so much the better, but failing that, matters should be arranged so that the impact of inconsistent decisions does not cause injustice.
52. Thirdly, again at [7] in the judgment, the Glaxo guidance is said to have the effect that:

"the primary consideration on an application for a stay is the probable duration of the proceedings before the two tribunals."

53. I have already explained how the likely duration of proceedings before the EPO forms an important part of the background to an application for a stay. But Lord Sumption is of course right that likely duration should not be seen as the only or even the principal reason for granting or refusing a stay. Delay must be considered in combination with other factors. A delay which causes neither side any real prejudice may justify a stay.
54. Two specific criticisms remain, namely the second and third points made by Lord Sumption at [38] in his judgment (quoted above). The second point is that the decision of the English court on validity is directly effective only in the United Kingdom, whereas the EPO's decision is directly effective in every country for which the patent was granted. The third is that the uncertainty and waste of costs do little credit to our national procedures.
55. Neither side made much of the second point. It is clear that Lord Sumption had in mind the fact that the decision of the EPO revoking the patent is directly effective in every country for which the patent is granted. This is of course a factor to bear in mind. But a large number of oppositions fail. In those cases the decision of the EPO has no effect at all in any of the member states. Even in cases where the patent is upheld or upheld with some amendment, the need for full national infringement proceedings remains. In such proceedings the defendant will be free to challenge validity in the national court as well. In the long period before the EPO decides, one cannot predict that it will resolve the dispute between the parties: it remains no more than a possibility.
56. In the same way, it must be borne in mind that national proceedings have the potential to deliver some degree of certainty. A defendant who is held not to infringe the claims of a granted patent by the national court knows that will remain so whatever the EPO decides. Claims cannot be broadened in opposition proceedings. Equally, the decision of the national court that the patent is invalid cannot be undone by the EPO. Moreover a rapid decision of a respected national court may promote early settlement of the dispute as a whole, bringing to an end proceedings everywhere.
57. The third point in [38] is of course important, as wasted costs are an unattractive feature of any litigation system. A system which allows for concurrent proceedings is more likely than others to involve some wasted costs in some cases. The court should seek to arrange matters to avoid wasted costs. Again, however, unless stays are to become mandatory, the court should recognise that in some cases the risk of wasted costs will be outweighed by other factors. Costs are, for example, not necessarily wasted if the steps in the national proceedings can produce some commercial certainty for the parties or advance the case towards settlement.
58. I turn to the submissions made by the parties on other features of the guidance. Mr Speck first submitted that the guidance should be abandoned altogether. He submitted that the effect of the Glaxo guidance was that a stay will very rarely be granted except by consent. He advocated a return to the position which he says prevailed before Glaxo, namely that there was a presumption in favour of a stay. In a related submission, he criticised the sixth guideline which, he submits, does away with any presumption in favour of a stay.
59. I have already accepted that the Glaxo guidance requires some revision in the light of Virgin, but I do not consider it should be swept away altogether. I do not accept that its effect is to prevent a stay except where the parties consent. The test remains one directed to finding where the balance of justice lies. I also reject the premise involved in Mr Speck's argument that Glaxo somehow did away with the presumption in favour of a stay. It cannot have done so because the presumption was expressly recognised and relied on by Lewison J at first instance at [41] and recited by the Court of Appeal at [30] before they said at [78] and again at [89] that he had correctly applied the legal principles relevant to a stay. I suspect Mr Speck's real complaint was with the ease with which, both before and after Glaxo the court has been prepared to treat the presumption as overcome.
60. I do not accept Mr Speck's submission about the sixth guideline, which is that there is no presumption that duplication of proceedings without more is a ground for a stay. Put in slightly different language, I understand this guideline to be saying that the mere fact that there are both national proceedings for revocation and EPO proceedings is not a ground for a stay. As so understood, it is as Mr Carr submitted, obviously right.
61. I would add that I would avoid for the future the use of the expression "duplication of proceedings". Even when national proceedings are confined to validity, concurrent proceedings here and in the EPO do not duplicate each other, because they can achieve different things. Unless the party attacking the patent agrees to be bound by the outcome in the EPO (a course which is seldom if ever taken), a dismissal of the opposition does not resolve the issue of validity between the parties. A national court is able to hold the patent valid as between the parties. In this respect the concurrent proceedings are unlike co-pending claims for breach of contract or claims in tort.
62. Mr Carr was inclined to accept that if there were no additional factor the court could and probably should grant a stay. One does not have to have concurrent validity proceedings just because the EPC permits them. So a stay is indeed the default position. It is for the party opposing a stay to demonstrate why it should not be granted. The guidance should more clearly reflect that position.
63. Mr Speck's alternative submission was that the operation of the Glaxo guidance should be confined to cases where a party was seeking to clear a patent out of the way to launch a product. I do not think that would be a satisfactory outcome. The guidance was intended as general guidance and should remain so. Any attempt to characterise cases in terms of the type of prejudice that a party will suffer, or the type of commercial certainty that may be achieved by the national proceedings is fraught with difficulty.
64. Mr Speck also made submissions about the fifth guideline. He submitted that it gives undue precedence to delay independently of prejudice. He also submitted that it fails to focus on the type of certainty which the decision in the national court will deliver. Mr Carr submitted that the fifth guideline merely emphasises the importance of delay as a factor, which it always will be. It did not mean that if you have a lengthy delay there would never be a stay.
65. I have already recognised that delay should not be seen as the only or even the principal reason for granting or refusing a stay. Delay must be considered in combination with other factors. A delay which causes neither side any real prejudice will normally result in a stay. Again, the guidance should reflect that position.
66. Mr Speck made no specific criticism of the seventh guideline on its own, namely that some certainty somewhere, and sooner rather than later, is better than continuing uncertainty anywhere. His submission was that, when taken in combination with the eighth guideline, that much weight should be given to assertions by a commercial party that they will suffer prejudice by refusal of a stay, the wrong balance was struck. He submitted that the court should look critically at both sides' contentions and decide accordingly.
67. I do not think that the eighth guideline was inviting the court to accept uncritically unlikely assertions of the need for commercial certainty. Roth J expressly said that he did not interpret it in that way, in his May 2012 judgment at [24], and he was right not to do so. The important point to draw from the eighth guideline is that the hearing of a stay application is not to become a mini-trial of the minutiae of the reasons advanced by either side for adopting one course or the other. It is clear that in the Glaxo case there had been an attempt to deconstruct Glaxo's reasons for wanting an early determination here.
68. In the light of the observations in Virgin and the arguments on this appeal I would recast the Glaxo guidance as follows:

"1.The discretion, which is very wide indeed, should be exercised to achieve the balance of justice between the parties having regard to all the relevant circumstances of the particular case."

2. The discretion is of the Patents Court, not of the Court of Appeal. The Court of Appeal would not be justified in interfering with a first instance decision that accords with legal principle and has been reached by taking into account all the relevant, and only the relevant, circumstances.

3. Although neither the EPC nor the 1977 Act contains express provisions relating to automatic or discretionary stay of proceedings in national courts, they provide the context and condition the exercise of the discretion.

4. It should thus be remembered that the possibility of concurrent proceedings contesting the validity of a patent granted by the EPO is inherent in the system established by the EPC. It should also be remembered that national courts exercise exclusive jurisdiction on infringement issues.

5. If there are no other factors, a stay of the national proceedings is the default option. There is no purpose in pursuing two sets of proceedings simply because the Convention allows for it.

6. It is for the party resisting the grant of the stay to show why it should not be granted. Ultimately it is a question of where the balance of justice lies.

7. One important factor affecting the exercise of the discretion is the extent to which refusal of a stay will irrevocably deprive a party of any part of the benefit which the concurrent jurisdiction of the EPO and the national court is intended to confer. Thus, if allowing the national court to proceed might allow the patentee to obtain monetary compensation which is not repayable if the patent is subsequently revoked, this would be a weighty factor in favour of the grant of a stay. It may, however, be possible to mitigate the effect of this factor by the offer of suitable undertakings to repay.

8. The Patents Court judge is entitled to refuse a stay of the national proceedings where the evidence is that some commercial certainty would be achieved at a considerably earlier date in the case of the UK proceedings than in the EPO. It is true that it will not be possible to attain certainty everywhere until the EPO proceedings are finally resolved, but some certainty, sooner rather than later, and somewhere, such as in the UK, rather than nowhere, is, in general, preferable to continuing uncertainty everywhere.

9. It is permissible to take account of the fact that resolution of the national proceedings, whilst not finally resolving everything, may, by deciding some important issues, promote settlement.

10. An important factor affecting the discretion will be the length of time that it will take for the respective proceedings in the national court and in the EPO to reach a conclusion. This is not an independent factor, but needs to be considered in conjunction with the prejudice which any party will suffer from the delay, and lack of certainty, and what the national proceedings can achieve in terms of certainty.

11. The public interest in dispelling the uncertainty surrounding the validity of monopoly rights conferred by the grant of a patent is also a factor to be considered.

12. In weighing the balance it is material to take into account the risk of wasted costs, but this factor will normally be outweighed by commercial factors concerned with early resolution.

13. The hearing of an application for a stay is not to become a mini-trial of the various factors affecting its grant or refusal. The parties' assertions need to be examined critically, but at a relatively high level of generality."

The issues in the action

69. HTC challenges the validity of the patent. No more need be said about that. However, the parties' positions in relation to infringement received some attention at the hearing. HTC's position has been that it has used and is using one or other of the approved variants. It appears to be accepted, however, that at least some of HTC's shipments in the past have not used such variants. However, on 30 May 2012 they felt able to give the undertaking to Roth J only to use one of the approved variants as a quid pro quo for the stay granted at that date. When it served its product and process descriptions ("PPDs") in relation to its current range, HTC indicated that all the devices operated in a manner similar to the B1 variant. Subsequently they moved to the D variant, and served amended PPDs. Once it appeared that, notwithstanding the Nokia judgment, IPCom did not accept that the variants did not infringe, HTC sought declarations of non-infringement in respect of all variants in the series, including B1 and D as well as C, B2, E, F and G.
70. Paragraph 24 of Roth J's order of 13 June 2013 required IPCom's statement of case on infringement to identify which of HTC's devices identified in their PPDs was contended to be an infringement. In that document IPCom identified the A and B1 devices, but not the D device as direct infringements of the patent. IPCom have very recently said that they will now allege that the D device is a direct infringement. If matters proceed it will be a matter for the trial judge to decide whether, in the light of the procedural history, IPCom should be allowed to do so.
71. At the end of its statement of case, IPCom alleged that the D device infringed the patent indirectly on the basis that it was capable of being updated over the air to operate in accordance with the standard (and thus the patent). The claim for indirect infringement has now been compromised on the basis of undertakings by HTC that they will not update any mobile device to perform in accordance with the standard, or to any RACH access method held between the parties to infringe the patent. Whether that undertaking would prevent the updating of a phone to operate, for example, in accordance with the B1 method is dependent on whether B1 infringes. It follows that there is a live and substantial dispute between the parties as to whether the A and B1 devices infringe. In addition, a decision on HTC's declaration of non-infringement may further inform the content of the undertaking which HTC have now given in relation to updating phones. It cannot be said to be common ground that HTC's current sales of the D device do not infringe.

Roth J's May 2012 judgment

72. Because much of Mr Speck's argument is based on a supposed inconsistency of approach between the judge's decision to grant a stay in May 2012 and his decision to refuse one in the judgment under appeal, I should summarise the reasons which the judge then gave for granting the stay.
73. HTC's position was that the costs of the technical trial would be wasted if either the patent was not restored or emerged with amended claims. IPCom's position was that there was a potentially huge delay. IPCom's business model depended on being able to licence its patent portfolio and enforce its patents where no licence was entered into. A stay would remove any incentive for HTC to resolve the dispute. The impact would not be limited to HTC, but would have an impact in relation to the industry as a whole. The cases against HTC and Nokia were important test cases for other potential licensees.
74. The judge recognised the force of IPCom's points, particularly the point that IPCom's business model would be imperilled in the absence of any threat of injunction or imposition of FRAND terms. Two points were set against that, however. Firstly, as matters then stood, there would still be a determination of FRAND binding on HTC. The FRAND trial was still going ahead despite the stay. Moreover, at that stage at least, HTC's undertaking to use only the approved variants was seen as an undertaking not to continue infringing. The judge said this at [25]:

"Without those two features I can see a strong case against a stay."

75. Having recognised that a patentee in the position of IPCom could have valid grounds for resisting a stay, the judge went on to draw comparisons with the case of a patentee manufacturer seeking to resist an incursion into its market, or a competitor seeking to clear the patent out of the way so that it could market its product without fear of infringement. Those were examples of parties' commercial objectives being frustrated if the proceedings were stayed. IPCom's interest was in the protection of its income stream. The judge was satisfied that this would be adequately protected by the setting of the licence terms.
76. Having discussed the issues relating to the likely delay, and balanced the relevant factors against the background as at that date the judge concluded that it was an appropriate case for a stay.

The judgment under appeal

77. The position at the date of the hearing in June 2013 was, as the judge explained, that (i) the opposition to the patent was still pending (ii) the FRAND trial had been vacated, and (iii) the technical trial was listed to commence in December 2013. He pointed out that, although he had previously granted a stay pending the outcome of the appeal to the TBA, the discretion to grant a stay had to be exercised afresh in the light of all the current circumstances.
78. The judge rejected a submission that the fifth Glaxo guideline (length of time to achieve some certainty on the issue of validity) did not apply because IPCom was a non-practising entity. He considered that the commercial circumstances of both parties, including the fact that IPCom was a non-practising entity, were relevant. This echoed the findings in his May 2012 judgment that IPCom had good grounds for contending that the protection of its business model required it to be allowed to pursue the national proceedings.
79. The judge distinguished the circumstances which had faced him in May 2012. At that time the OD had decided the patent was invalid. If that decision had been upheld on appeal, any trial of the technical issues would have plainly been a waste of money. The present situation involved waiting for two further decisions of the TBA as it was accepted that a further appeal was likely in the EPO. Even if accelerated, the EPO decision would come at least 8 months after the technical trial. Even that decision might not be the end of the road in the EPO. Accordingly the potential total delay now sought was much longer than that envisaged in 2012.
80. A further significant distinction from 2012 was that there were now live issues of infringement in relation to the variants proposed to be used by HTC in the period of the stay. This represented a change of position by IPCom, but not one which HTC contended was not open to it. This meant that IPCom faced not only delay in recovering damages for past infringement but also faced continuing infringement which it was powerless to prevent by injunction. In that connection he said this, consistently with his findings in May 2012:

"…where a patentee's whole business is based on the exploitation of its patents, extended uncertainty over a patent's validity in circumstances where it contends that the subject-matter of the patent is being continually used is likely to be of significant consequence. In my judgment, IPCom's assertion that there is significant commercial benefit in the relative certainty offered by determination against HTC in this court of validity of the TBA Patent is a very relevant factor."

81. The judge acknowledged that there was a risk of wasted costs in the event that IPCom prevailed on the issues of novelty and inventive step in this country but lost in the EPO. He took heart, however, from the fact that the issues had been investigated once in the Nokia action. The narrowing of the claims by the TBA could not have added strength to those objections.
82. The judge also acknowledged an irony in HTC's position. It was HTC who contended that the patent was invalid. If they were right, that was an end of the matter. It was only if the result was the opposite of that which they contended for that there would be any possibility of wasted costs.
83. The judge accepted that there was a risk of yet further amendments to the claims in the EPO, and that those amendments might have an impact on the technical trial. But he concluded that that factor was outweighed by the risk of prolonged delay and the relative commercial certainty that would be achieved by a technical trial at the end of the year.

The arguments on the appeal

84. Mr Speck commences his argument by submitting that the judge based his judgment on the Glaxo guidance which the Supreme Court has subsequently criticised. I have dealt with the submissions about the guidance above and concluded that the Glaxo guidance does require some revision. But it remains the case that we should not interfere with the patent judge's discretion unless it fails to accord with legal principle or has not been reached by taking into account all the relevant circumstances and only the relevant circumstances. The fact that Virgin has necessitated a change in the Glaxo guidance does not necessarily mean that the judge's exercise of the discretion was flawed.
85. Mr Speck also submitted that the factors which favoured the grant of a stay in May 2012 were still present and favoured the grant of a stay in 2013. I will deal with that submission when I have considered whether the judge's exercise of discretion appears flawed.
86. Mr Carr submitted that even in the light of any necessary revisions to the Glaxo guidance, the judge's exercise of discretion was correct.

Factors in the present case

87. Both sides assume there will be at least one further round of decisions of the OD and the TBA. The current estimate is that the OD hearing will be in January 2014 and an appeal, if expedition is granted, would be in February or March 2015. If the trial date is not fixed in the interim, an 8 day trial would then be at least a year away. In theory one could, as has been done here, fix the trial in the interim, but this would lock up court time for other users. If that is not done the effect of a stay would be to postpone the technical trial from December 2013 to early to the middle of 2016. If there is a further round of EPO hearings, the delay may be significantly longer.
88. The judge did not expressly consider whether the refusal of a stay could irrevocably deprive HTC of any part of the benefit of the concurrent jurisdiction. As the law then stood, it might have done, because IPCom might have obtained an enforceable award of damages before the EPO had its final opportunity to declare the patent invalid. However, IPCom had already, at the May 2012 hearing, recognised the unattractiveness of the notion of attempting to obtain what have been described as Unilin damages, by undertaking to repay any damages if it subsequently transpired that the patent was invalid. An undertaking of that kind is still offered. There is no question of IPCom acquiring any irrevocable benefit.
89. Thus, in paragraph 31 of his judgment, the judge identifies the fact that the extended delay meant that, on its case, IPCom was facing delay in recovering damages for past infringement before the change to approved variants, and also facing continuing infringement which it could not prevent by injunction, and damages which would also be delayed. However he recognised as well that if the March 2013 claims turned out to be invalid, IPCom would have no such rights.
90. In that paragraph he also considered that, for a patentee whose whole business depends on the exploitation of its patents, continuing uncertainty over the validity of the patent in circumstances where it was being continually used was likely to be of significant consequence. His evaluation was that this consideration was of greater weight than the procedural complications which might result if the patent was subsequently amended or revoked.
91. I do not think there is anything in the remaining revisions I have made to the guidance which is of any real materiality to the judge's evaluation of these factors. I do not think that this court would be justified in interfering with the judge's exercise of discretion. There can be no general rule that a patentee who has precluded himself from obtaining an injunction has also precluded himself from obtaining any other form of relief or determination until the opposition processes of the EPO are complete. Everything is acutely dependent on the circumstances of the individual case, which are for the patents judge to assess.
92. The judge would also have been entitled to take account of the fact that the technical trial provides the potential for HTC to achieve certainty that their devices do not infringe valid claims in the UK, either because the March 2013 claims are invalid or because their products fall outside their scope. Even though this is not something which HTC stressed in their evidence and arguments, it remains the fact that it is their contention, and the trial provides them with an opportunity to demonstrate that they are right.
93. I also do not accept Mr Speck's alternative arguments based on a comparison of the factors facing the judge in May 2012 and June 2013. In his May 2012 judgment the judge was weighing the prejudice to IPCom in the context of a delay pending one further instance in the EPO with the reassurance of a pending FRAND determination, and no infringement in the meantime. He expressly rejected at [32] an argument that he should consider a longer delay because of the practice of the EPO of serial decision making, thus opening up the possibility of further hearings before the OD and TBA. He rejected the argument based on the longer delay on the basis that he was only considering, and indeed had only been asked to consider, a stay until the decision of the TBA on the pending appeal, and his decision was without prejudice to the arguments on a further application for a stay. The judge was right to say in the decision under appeal that he was exercising an entirely fresh discretion.
94. It is not for this court to say whether the judge was right or wrong to grant the stay in May 2012. All I can say is that there is nothing in that judgment which would justify this court in interfering. The same is true of his judgment under appeal which related to different circumstances.
95. I would accordingly dismiss the appeal.

Lady Justice Rafferty

96. I agree.

Lord Justice Patten

97. I also agree.