Fontem Holdings 1BV and another v Ten Motives Ltd and another; Nicocigs Ltd v Fontem Holdings 1BV and another [2015] EWHC 2752 (Pat), is 2 October 2015 ruling by Mr Justice Norris, in the Patents Court, England and Wales.
In short, this was an application by the defendant -- the alleged infringer -- for summary judgment on infringement of a patent for an e-cigarette, or in the alternative an application to stay the action pending the outcome of European Patent Office (EPO) opposition proceedings. In a pretty short judgment for a patent case (just 46 paragraphs in length) Norris J summarised the law, considered the issue of infringement and briefly set out the particular circumstances of the new and rapidly-growing market for e-cigarettes.
Given the nature and speed of change in the marketplace, he said, there was a clear need for a much earlier determination of the validity and infringement issues in the national litigation than the earliest possible date for completion of the EPO's opposition proceedings. If however infringement were proven at trial, there remained the possibility of some sort of stay when it came to sorting out the appropriate relief.
The PatLit weblog covers patent litigation law, practice and strategy, as well as other forms of patent dispute resolution. If you love -- or hate -- patent litigation, this is your blog. You can contact PatLit by emailing Michael here
Showing posts with label infringement. Show all posts
Showing posts with label infringement. Show all posts
Friday, 9 October 2015
Monday, 24 August 2015
Microsoft v Motorola: did the US Court of Appeals for the Ninth Circuit have jurisdiction?
From our good friend Kevin Winters (trainee solicitor, Shepherd & Wedderburn) comes the following guest item on the jurisdictional dimension of the very lengthy ruling one of the most fascinating and complex pieces of patent litigation to emerge from the United States this year. This is what he writes:
Microsoft v Motorola: did the US Court of Appeals for the Ninth Circuit have jurisdiction?
In Microsoft v Motorola, Inc (here), the United States Court of Appeals for the Ninth Circuit affirmed the decision of the district court that Motorola had breached its obligations to offer reasonable and non-discriminatory (RAND) licences to its patents in good faith. In addition to the substantive arguments, the Court of Appeals had to decide whether the Ninth Circuit or the Federal Circuit had jurisdiction over the case.
The background
In late 2010 both Motorola and Microsoft sued in the U.S. International Trade Commission (ITC) and the Western District Court of Washington, alleging infringement of smartphone patents. During this litigation the parties entered into discussions about a cross-licensing agreement that would grant Motorola licences to Microsoft’s smartphones in exchange for licences to any of Motorola’s patents that Microsoft’s products may have been infringing.
In late October Motorola provided Microsoft with letters offering to license two patent portfolios, 802.11 and H.264 HEP, at 2.25% of the price of the end product incorporating the patents, regardless of the manufacturer: Microsoft was to pay Motorola 2.25% of the sale price of an Xbox game console or of any computer using Microsoft Windows. Both letters which were identical in material terms represented that the offer was in keeping with Motorola’s reasonable and non-discriminatory (RAND) commitments. After Microsoft received the letters it filed a diversity action in the Western District of Washington, alleging that Motorola had breached its RAND commitments to the IEEE and ITU. It alleged that Motorola’s letters constituted a refusal to license Motorola’s standards-essential patents (SEPs) on RAND terms. The next day Motorola filed suit against Microsoft in an attempt to enjoin Microsoft from using its H.264 patents. The cases were consolidated in the Western District of Washington.
Motorola also filed patent-enforcement suits with the ITC, seeking an exclusion order against importing Microsoft’s Xbox products into the USA, and with a German court, seeking an injunction against sales of Microsoft’s H.264-compliant products. To avoid the economic loss it would suffer if an injunction against the use of Motorola’s two German H.264 patents was granted – Microsoft’s European distribution centre for Windows and Xbox products was in Germany – Microsoft moved its distribution operations to the Netherlands. It also obtained an anti-suit injunction barring Motorola enforcing any injunction it might obtain in a German court against Microsoft’s use of Motorola’s 1h.264 SEPs until the district court could decide whether an injunction was an appropriate remedy for Motorola to seek. The Court of Appeals upheld the anti-suit injunction in 2012, while the German court ruled that Motorola was entitled to an injunction.
In the district court Microsoft altered its complaint to allege that Motorola’s filing of injunctive actions constituted a breach of contract, owing to the fact that the obligation to offer RAND licences to all seekers prohibited Motorola from seeking injunctive relief for violations of patents subject to that obligation. In response the court granted a joint motion to stay all patent-infringement claims in the consolidated cases until the RAND issues were resolved.
The district court issued a range of orders that, among other things, recognised that RAND commitments create enforceable obligations between Motorola and the SSO concerned and that Microsoft – as a standard user – could enforce those contracts as third-party beneficiary. It also issued orders that Motorola’s commitments to the ITU and IEEE required initial offers by Motorola to license its SEPs to be offered in good faith, but needn’t be on RAND terms, provided that a RAND licence is eventually issued. Further, the court decided that Motorola was not entitled to an injunction on its H.264 or 802.11 patents.
In November 2012 the district court conducted a bench trial to determine a RAND rate and range for Motorola’s H.264 and 802.11 patents. It concluded that the RAND royalty for Motorola’s H.264 portfolio was .555 per cent end-product unit, with an upper bound of 16.389 cents per unit, and the 802.11 portfolio was 3.71 cents per unit, with a range of .8 cents to 19.5 cents.
The case was then the subject of a jury trial on the breach of contract claim. While Motorola objected, Microsoft introduced the RAND rate’s determined at bench trial, along with testimony that Motorola and its parent company, Google Inc, had previously been investigated for failing to licence patents relating to smartphones, tablets and videogame systems on RAND terms. Microsoft sought damages in the form of attorneys’ fees, costs in defending the injunction actions that Motorola had raised and in relocating the distribution centre from Germany to the Netherlands.
In September 2013 the jury returned a verdict in favour of Microsoft for the amount of $14.52 million. The verdict form asked jurors the general question of whether Motorola “breached its contractual commitment[s]” to the IEEE and ITU and, the specific question, in determining damages, whether Motorola’s “conduct in seeking injunctive relief, apart from Motorola’s general course of conduct, violated Motorola’s dut[ies] of good faith and fair dealing with respect to Motorola’s contractual commitment[s].” The jury answered “yes” to all questions unanimously.
Motorola sought judgement as a matter of law both at the close of evidence and at the close of Microsoft’s case-in-chief. Having heard the jury’s verdict, the court denied Motorola’s motions concluding that (i) the evidence was enough for the jury reasonably to conclude that Motorola breached its duty of good faith and fair dealing by seeking injunctions against Microsoft, and (ii) the damages award was proper. Microsoft’s motion for entry of final judgement on the breach of contract jury verdict was granted.Motorola appealed the judgement on the breach of contract claim to the Federal Circuit which, on Microsoft’s motion, transferred the appeal to the Court of Appeals for the Ninth Circuit.
The question of jurisdiction
Motorola took issue with the decision of the Federal Circuit to transfer the case to the Ninth Circuit. It argued that the Federal Circuit was the correct forum for the case to be heard, owing to the dispute “arising under” federal patent law. The Ninth Circuit considered this argument and quickly dispensed with Motorola’s challenge to its jurisdiction.
The Ninth Circuit pointed out in the first instance that both the interlocutory appeal and the Federal Circuit’s decision to transfer the appeal to the Ninth Circuit were both in-keeping with the ‘law of the case’ doctrine. This doctrine requires that significant deference is given to prior court decisions that are higher in the court hierarchy, unless there are convincing reasons not to do so:
“…we may now decline jurisdiction only if “(1) the [earlier] decision[s] w[ere] clearly erroneous; (2) there has been an intervening change in the law; (3) the evidence on remand is substantially different; (4) other changed circumstances exist; or (5) a manifest injustice would otherwise result.”The court was of the view that Motorola’s argument that the district court “constructively amended” the complaint by holding a bench trial on the RAND rate implicitly invoked the “changed circumstances” exception. However the court pointed out that, given that it had decided upon jurisdiction of the appeal knowing that the RAND bench trial would take place, and the Federal Circuit’s decision to transfer the case was made after the bench trial, there was no evidence of changed circumstances.Furthermore it was pointed out in the court judgement that the fact that the district court decided to hold a trial on the RAND rate, did not affect the application of the law of the case doctrine: the Court of Appeals was aware of the district court’s intention to determine the RAND rate and it owed deference to the Federal Circuit’s decision to transfer the case – a decision that was taken after the district court had determined the RAND rate.The court also addressed the argument that there was a “clear error” or “manifest injustice” in the determination of the RAND rate. It pointed out that this did not transform the case so as to require the determination of a substantial question of federal patent law:“A complaint that alleges breach of contract and seeks damages sounds in contract; its nature “does not change because the contract is a patent license… Even if a court, in interpreting a contract and assessing damages, deems it appropriate to apply the law of patent infringement, that of itself does not change the complaint into one arising under the patent law.”While the court acknowledged Motorola’s argument that the Federal Circuit had exercised its jurisdiction in some breach-of-contract cases, it pointed out that those involved some determination of patent law. The case before the Ninth Circuit was a breach of contract action.
Friday, 17 April 2015
Self-sealed mailer patent comes unstuck
Everseal Stationery Products Ltd v Document Management Solutions Ltd & Others [2015] EWHC 842 (IPEC) is a decision of Judge Hacon in the Intellectual Property Enterprise Court, England and Wales, dating back to the beginning of this month.
Everseal had a patent for a product which still has some currency, even in the age of the internet: this product was a "mailer", this being a business communication which was folded, sealed by an adhesive around its periphery and sent through the terrestrial post. Claim 1 of Everseal's patent disclosed a mailer including an adhesive which self-sealed irreversibly under "finger pressure" namely pressure of 500 kPa or less. Claim 7 disclosed a mailer in which the adhesive was a natural rubber latex adhesive stabilised with ammonia.
DMS developed mailers which were referred to in the judgment as Mailer 1, Mailer 2 and Mailer 3. Everseal sued DMS for infringement of its patent; DMS counterclaimed for invalidity. The court had to decide whether (i) DMS's mailers fell within claims 1 and 7, and in particular whether DMS's adhesives sealed "irreversibly" and (ii) whether the patent was invalid for lack of novelty or inventive step over three pieces of prior art (Viking mailer, Faltin, and Johnsen).
Judge Hacon held for DMS. In his view:
* an adhesive self-sealed "irreversibly" if the recipient, receiving the mailer, could tell from both the adhesive structure and the appearance of the base material that it had previously been opened. If the fact that it had previously been opened was not betrayed by each of those criteria taken separately, the adhesive did not self-seal irreversibly when the mailer was first closed.
* it could be assumed that anyone who had previously opened the mailer without permission would have taken some limited effort to hide the fact that he had done so.
* the seals on Mailer 1 were irreversible within that meaning; mailer 1 infringed claim 1 of the patent.
* infringement of claim 7 would have required Everseal to show that the adhesive used for Mailer 1 was a natural rubber adhesive stabilised with ammonia, but there was no evidence to that effect.
* experiments carried out by DMS did not enable the court to reach any conclusion as to whether opening Mailers 2 or 3 resulted in the disruption in the mailer's base material so as to demonstrate that their adhesives sealed irreversibly, and Everseal had chosen not to conduct its own experiments.
* the burden rested on Everseal to prove infringement, but it had not done so regarding Mailers 2 and 3.
* turning now to the counterclaim, claim 1 of the patent lacked novelty over the Viking mailer which had an irreversible seal and had self-sealed under finger pressure. Claim 7 did not however lack novelty over the Viking mailer.
* where lack of novelty turned on whether an alleged prior use had particular features, it was sufficient to adduce evidence which proved, on the balance of probabilities, facts establishing that the prior use had those features. In contrast, where the court was invited to infer that prior use had one or more features on the basis of primary facts proved, the invention would lack novelty only if the inference was inevitable. Here the evidence -- that it was likely that the adhesive in the Viking mailer was a natural rubber latex stabilised with ammonia -- was not enough to establish lack of novelty regarding claim 7.
* the patent did not lack novelty over Faltin or Johnsen, since neither of them explicitly disclosed an irreversible seal.
* Claim 1 did however lack inventive step over Faltin and Johnsen, and over the Viking mailer if it did not anticipate the invention. Everseal had suggested no technical prejudice which would have deterred a skilled team, working prior to the claimed invention, from using an adhesive which self-sealed under finger pressure to create an irreversible seal.
* Claim 7 also lacked inventive step, the evidence being that a skilled person would have been expected to find a suitable formulation without difficulty.
This is a curious case, in that the litigation over this patent was commenced in the Patents County Court (PCC) in February 2010, before that court's rules were radically overhauled with effect from 1 October. The PCC then gave separate rulings on two preliminary points; the ownership of the patent was disputed; there were also a Tomlin Order, allegations of unlawful threats and two separate applications to amend the patent before the infringement action was finally heard by the IPEC. One might be tempted to wonder whether the sequential segmenting of issues is always more efficient than trying to hear as many things as possible all in one go.
Everseal had a patent for a product which still has some currency, even in the age of the internet: this product was a "mailer", this being a business communication which was folded, sealed by an adhesive around its periphery and sent through the terrestrial post. Claim 1 of Everseal's patent disclosed a mailer including an adhesive which self-sealed irreversibly under "finger pressure" namely pressure of 500 kPa or less. Claim 7 disclosed a mailer in which the adhesive was a natural rubber latex adhesive stabilised with ammonia.
DMS developed mailers which were referred to in the judgment as Mailer 1, Mailer 2 and Mailer 3. Everseal sued DMS for infringement of its patent; DMS counterclaimed for invalidity. The court had to decide whether (i) DMS's mailers fell within claims 1 and 7, and in particular whether DMS's adhesives sealed "irreversibly" and (ii) whether the patent was invalid for lack of novelty or inventive step over three pieces of prior art (Viking mailer, Faltin, and Johnsen).
Judge Hacon held for DMS. In his view:
* an adhesive self-sealed "irreversibly" if the recipient, receiving the mailer, could tell from both the adhesive structure and the appearance of the base material that it had previously been opened. If the fact that it had previously been opened was not betrayed by each of those criteria taken separately, the adhesive did not self-seal irreversibly when the mailer was first closed.
* it could be assumed that anyone who had previously opened the mailer without permission would have taken some limited effort to hide the fact that he had done so.
* the seals on Mailer 1 were irreversible within that meaning; mailer 1 infringed claim 1 of the patent.
* infringement of claim 7 would have required Everseal to show that the adhesive used for Mailer 1 was a natural rubber adhesive stabilised with ammonia, but there was no evidence to that effect.
* experiments carried out by DMS did not enable the court to reach any conclusion as to whether opening Mailers 2 or 3 resulted in the disruption in the mailer's base material so as to demonstrate that their adhesives sealed irreversibly, and Everseal had chosen not to conduct its own experiments.
* the burden rested on Everseal to prove infringement, but it had not done so regarding Mailers 2 and 3.
* turning now to the counterclaim, claim 1 of the patent lacked novelty over the Viking mailer which had an irreversible seal and had self-sealed under finger pressure. Claim 7 did not however lack novelty over the Viking mailer.
* where lack of novelty turned on whether an alleged prior use had particular features, it was sufficient to adduce evidence which proved, on the balance of probabilities, facts establishing that the prior use had those features. In contrast, where the court was invited to infer that prior use had one or more features on the basis of primary facts proved, the invention would lack novelty only if the inference was inevitable. Here the evidence -- that it was likely that the adhesive in the Viking mailer was a natural rubber latex stabilised with ammonia -- was not enough to establish lack of novelty regarding claim 7.
* the patent did not lack novelty over Faltin or Johnsen, since neither of them explicitly disclosed an irreversible seal.
* Claim 1 did however lack inventive step over Faltin and Johnsen, and over the Viking mailer if it did not anticipate the invention. Everseal had suggested no technical prejudice which would have deterred a skilled team, working prior to the claimed invention, from using an adhesive which self-sealed under finger pressure to create an irreversible seal.
* Claim 7 also lacked inventive step, the evidence being that a skilled person would have been expected to find a suitable formulation without difficulty.
This is a curious case, in that the litigation over this patent was commenced in the Patents County Court (PCC) in February 2010, before that court's rules were radically overhauled with effect from 1 October. The PCC then gave separate rulings on two preliminary points; the ownership of the patent was disputed; there were also a Tomlin Order, allegations of unlawful threats and two separate applications to amend the patent before the infringement action was finally heard by the IPEC. One might be tempted to wonder whether the sequential segmenting of issues is always more efficient than trying to hear as many things as possible all in one go.
Thursday, 27 March 2014
Adult TV channel blocking patents held invalid
Via Paul England (a professional support lawyer at Taylor Wesing and an occasional guest blogger with the IPKat) comes the following note on a mammoth judgment delivered yesterday by Mr Justice Arnold in the Patents Court, England and Wales, in Starsight Telecast Inc. and United Video Properties Inc. v Virgin Media Limited, Virgin Media Payments Limited and Tivo Inc. [2014] EWHC 828 (Pat). The ruling, which spans 316 paragraphs, is richer in terms of analysis of the facts than in establishing any new principle of law, but still has plenty in it in the manner in which it holds invalid a pair of patents for blocking adult TV channels. Explains Paul:
Starsight and United Video (collectively "Rovi") claimed that Virgin and TiVo (collectively "Virgin") infringed EP (UK) 1 763 234 which was an "Improved electronic television program schedule guide system and method" and EP (UK) 0 821 856 for "Merging multi-source information in a television system" by supplying two different classes of set-top box ("STBs"). Specifically, both of Virgin's STBs were capable of providing the user with a "TV Guide" and performing functions that allowed the user to "Hide Adult Channels" or have an "Alternative Adult Listing Display" by entering a PIN and accessing the STBs' respective parental controls menu.
In respect of the "Hide Adult Channels" function, claims 1 and 11 of the '234 patent were held not to be infringed, because entry of the PIN would not in itself change the availability of the adult channels, as was required by these claims. It would instead merely allow the user access to the relevant menu to edit the adult channel settings. In relation to the "Alternative Adult Listing Display", Virgin did not dispute infringement of claim 1. However, on the basis that this function would not operate by checking a record of schedule information in a database, as disclosed in the '234 patent, but would check whether the parental control option was turned on or off, the alleged infringement of claim 2 of '234 was disputed. Despite this argument, the judge found that claim 2 was infringed.But, Mr Justice Arnold also held the '234 Patent as a whole to be invalid for obviousness.
Claims 1 and 12 of the '856 patent were held not to be infringed by Virgin's STB as it only had one source and source device, as opposed to a "plurality of sources" as claimed by the '856 patent. In addition, the sequence of steps specified in the '856 patent that would be required to infringe were performed in a different order by Virgin's STB. Further, all the claims under the '856 patent were found to lack novelty and were held obvious over the prior art. '856 was therefore also invalid.As a final note, Mr Justice Arnold stated that even if the claims of the '856 patent were valid, Virgin would not have infringed them.The judgment, which followed eight days of hearings, contains one small item that attracted my attention: the notion, at [126], that a witness who was essentially a lone inventor might not be considered as being representative of a member of a skilled team who whom a patent is addressed.
Tuesday, 25 February 2014
No, you can't be served with any more pleadings, judge tells defendants
Last Friday Mr Justice Mann, sitting in the Patents Court, England and Wales, gave a ruling in a hearing relating to the service of further pleadings in a patent infringement action. The case, Electromagnetic Geoservices ASA v Petroleum Geo-Services ASA & others, is not reported on BAILII but a note has been published on the subscription-based Lawtel service.In these proceedings the defendants argued that EGA' case on infringement should be embodied in a pleading in order to achieve clarity and so that the defendants (i) could see whether there was a reasonably arguable case, (ii) could assess the shape of the action and formulate their own case, and (iii) could prepare their product and process description (PPD).
Mann J refused the application on the basis that, in the circumstances, the letters from EGA's solicitors were likely to be treated in the proceedings as though they were a pleading. In his view:
* since the defendants' operations took place on ships in the middle of the sea, EGA was relying on publicly available materials in order to give the best particulars it could of the allegedly infringing acts.
* what was requested by the defendant was something that was not usually done in patent cases and which appeared to be unnecessary in the circumstances. The letters from EGA's solicitors appeared to be full, frank and comprehensive. While the case set out in them was subject to variation after disclosure, that would be the normal position with a pleading.
* the defendants were sufficiently tied down to their case in the correspondence, and it would be a waste of time and costs to require a pleading -- there being nothing to be gained from the small degree of additional fixing which would result from a pleading.
* if the defendants wanted to apply to strike out the claim, they had the information on which to make a decision.
* if a pleading gave better particulars, the defendants might even be slightly worse off from that point of view and EGA would not be prevented from adducing further information to meet any strike-out application.
At this distance from the case and with no further facts, it's difficult to know whether the defendants were trying a crafty ploy or genuinely seeking clarification of EGA's claims. Either way, the court appears to have taken an approach which is both cautious and pragmatic.
Wednesday, 25 January 2012
When the Ravages of Time Contribute to Patent Infringement
In a recent decision published here, the Düsseldorf district court had to decide on a preliminary injunction requested based on a patent on a substance containing a mixture of two substances – Leflunomid and Teriflunomid. The claims required among other things that the amount of Teriflunomid should fall in a range between 0,3% and 50% of the amount of Leflunomid.
The allegedly infringing embodiments contained a priori only the first substance Leflunomid. However, the patentee argued that due to certain chemical processes occurring under regular storage conditions at a temperature of 25°C and a humidity of 60%, the amount of second substance Teriflunomid would increase and reach the threshold of 0,3% after 6 months for a first embodiment and after 9 months for a second embodiment.
The court found the product to infringe the patent directly and literally (no contributory infringement as one might have presumed). It referred to the earlier judgement with the title “Traktionshilfe” of the Düsseldorf Upper District Court (OLGDüsseldorf I-2 U 139/09) departing from the general rule according to which all the features of a claim must be realized in the moment of the use in cases where the circumstances will change in a reliable and predictable way such that the features of the claim will be realized soon (“demnächst”).
Interestingly, the Bundespatentgericht (BPatG) had rejected a novelty attack based on very similar arguments in the co-pending nullity procedure (which can be found here ). The BPatG found that even if the conversion of Leflunomid into Teriflunomid may actually have taken place under certain storage conditions prior to the priority date, this conversion was not the consequence of a methodical (planvoll) technical activity but rather uncontrolled such that the actual technical teaching to use product with a certain composition was not known. If achieving the claimed composition is the result of a certain choice of storage conditions and there is no evidence that this choice was actually made, the claimed composition is novel.
This highlights an important difference between the assessment of novelty and infringement which results from the focus on the technical idea rather than on the tangible features. According to the decision, a product may be infringing a patent accidentally, e.g. as the result of a natural decay/conversion process. If the same product (but not the decay process) was known before the priority date and the circumstances leading to the decay/conversion process are actually not inevitable, it would not be novelty destroying.
Tuesday, 20 December 2011
FRAND Defense Requires Accepting Damages for the Past
The Mannheim District Court had to decide on an infringement action brought against Apple by Motorola Mobility. Apple had raised an "Orange Book" defense, which is the german version a FRAND defense by unconditionally offering a license agreement under FRAND terms for the future but without accepting damages beyond the FRAND license fees for the past.
The hitherto unresolved question whether the FRAND offer has to be made before using the patent has been answered in the negative. However, the court decided that in the case where the patent had been used in the past, the FRAND offer has to include an unconditional acknowledgement of the damages for this past use.
First of all, the patentee behaves contradictory when he admits, in the license agreement, that certain acts in the future use the patent while disputing that the same acts in the past have been infringing the patent. According to the Mannheim Court, antitrust regulations do not require any limitation of damages for the past to amounts corresponding to FRAND-License fees.
Further reasons for the latter decision depend on details of the German jurisdiction and the specific circumstances of the case. It might therefore be difficult to use the arguments in other jurisdictions.
In the case at issue, a nullity suit was co-pending at the Federal German Patent Court and the Mannheim court had decided not to stay the infringement procedure until the decision in the nullity suit was made. In the event that the patent will be revoked, any damages paid would be refunded, whereas this is not the case for license fees. The party raising the FRAND defense would therefore achieve an economical advantage over the diligent licensee asking for a license before using the patent.
Further, the case-law unambiguously accepts the right of the patentee to terminate a license agreement for good cause if the licensee files a nullity action against the patent. In the case of a co-pending nullity suit, the principle of dolo agit, qui petit, quod statim redditurus est therefore forecloses the licensee from forcing the patentee into an agreement which could then be terminated by the patentee immediately.
As a consequence, the unconditional withdrawal of the nullity action as well as the acceptance of damages for the past appear to be mandatory prerequisites for raising a FRAND-defense in Germany.
Further points discussed but not decided include the question whether the unconditional FRAND offer by the licensee has to encompass all of the standard-essential patents of the patentee (or of the patentee’s parent company?) in all countries so as to terminate all the ongoing litigations between the party and the question whether the FRAND defense may not be raised at all in cases of willful infringement.
The full text of the decision has been published under "Justiz in Baden-Württemberg" here.
The hitherto unresolved question whether the FRAND offer has to be made before using the patent has been answered in the negative. However, the court decided that in the case where the patent had been used in the past, the FRAND offer has to include an unconditional acknowledgement of the damages for this past use.
First of all, the patentee behaves contradictory when he admits, in the license agreement, that certain acts in the future use the patent while disputing that the same acts in the past have been infringing the patent. According to the Mannheim Court, antitrust regulations do not require any limitation of damages for the past to amounts corresponding to FRAND-License fees.
Further reasons for the latter decision depend on details of the German jurisdiction and the specific circumstances of the case. It might therefore be difficult to use the arguments in other jurisdictions.
In the case at issue, a nullity suit was co-pending at the Federal German Patent Court and the Mannheim court had decided not to stay the infringement procedure until the decision in the nullity suit was made. In the event that the patent will be revoked, any damages paid would be refunded, whereas this is not the case for license fees. The party raising the FRAND defense would therefore achieve an economical advantage over the diligent licensee asking for a license before using the patent.
Further, the case-law unambiguously accepts the right of the patentee to terminate a license agreement for good cause if the licensee files a nullity action against the patent. In the case of a co-pending nullity suit, the principle of dolo agit, qui petit, quod statim redditurus est therefore forecloses the licensee from forcing the patentee into an agreement which could then be terminated by the patentee immediately.
As a consequence, the unconditional withdrawal of the nullity action as well as the acceptance of damages for the past appear to be mandatory prerequisites for raising a FRAND-defense in Germany.
Further points discussed but not decided include the question whether the unconditional FRAND offer by the licensee has to encompass all of the standard-essential patents of the patentee (or of the patentee’s parent company?) in all countries so as to terminate all the ongoing litigations between the party and the question whether the FRAND defense may not be raised at all in cases of willful infringement.
The full text of the decision has been published under "Justiz in Baden-Württemberg" here.
Monday, 14 November 2011
Should I Stay Or Should I Go?
One of the particularities of the German double-tracked patent litigation system with infringement judged at a court differing from the court judging validity (Bundespatentgericht) is the synchronization of these procedures.
In theory, both tracks are joined at the BGH acting as the 2nd instance in nullity and – if leave to revision is granted – as the 3rd instance in the infringement procedure. The decision not to grant leave to revision may be appealed in a so-called “Nichtzulassungsbeschwerde” (NZB), which may be based on the grounds of severe violations of the right to be heard or other grounds of fundamental importance. Isolated wrong decisions do usually not count among these grounds.
In a fairly surprising decision with the name “Crimpwerkzeug III” (see english summary here), the BGH found that differences in claim construction between the 2nd instance infringement proceedings and the 2nd instance nullity proceedings may in fact constitute such a ground of fundamental importance provided that the correct interpretation (i.e. that of the BGH) would have led to a differing result in the infringement proceedings. The main argument was that the persistence of and lack of legal remedies against such irreconcilable discrepancies between the two tracks would derogate the public reliance in the German double-tracked legal system as a whole.
As a matter of course, the assessment of the differing claim construction requires that both procedures are completed. As a consequence, it appeared to be mandatory for the BGH to stay the NZB procedure (and delay the legal force of the infringement sentence) until completion of the nullity/invalidity procedure after Crimpwerkzeug III. I presume that this has led to an increasing number of NZB procedures pending at the BGH.
The recent decision “Klimaschrank” has now put limits to the rule that the BGH has to stay the NZB procedure in these cases. The BGH emphasized that the decision to stay the procedure is not mandatory but still at the discretion of the senate and that a late filing of the nullity action (as compared to the filing of the infringement action) may be of some weight in favour of the patentee. In the case at issue, the nullity suit had been lodged eight months after the pronouncement of the infringement decision and only briefly after the publication of Crimpwerkzeug III.
In cases where the nullity action is filed more or less directly in response to the infringement action, it remains the standing advice to lodge a NZB if the district court decides on the infringement without waiting for the nullity procedure as a precautionary measure.
In theory, both tracks are joined at the BGH acting as the 2nd instance in nullity and – if leave to revision is granted – as the 3rd instance in the infringement procedure. The decision not to grant leave to revision may be appealed in a so-called “Nichtzulassungsbeschwerde” (NZB), which may be based on the grounds of severe violations of the right to be heard or other grounds of fundamental importance. Isolated wrong decisions do usually not count among these grounds.
In a fairly surprising decision with the name “Crimpwerkzeug III” (see english summary here), the BGH found that differences in claim construction between the 2nd instance infringement proceedings and the 2nd instance nullity proceedings may in fact constitute such a ground of fundamental importance provided that the correct interpretation (i.e. that of the BGH) would have led to a differing result in the infringement proceedings. The main argument was that the persistence of and lack of legal remedies against such irreconcilable discrepancies between the two tracks would derogate the public reliance in the German double-tracked legal system as a whole.
As a matter of course, the assessment of the differing claim construction requires that both procedures are completed. As a consequence, it appeared to be mandatory for the BGH to stay the NZB procedure (and delay the legal force of the infringement sentence) until completion of the nullity/invalidity procedure after Crimpwerkzeug III. I presume that this has led to an increasing number of NZB procedures pending at the BGH.
The recent decision “Klimaschrank” has now put limits to the rule that the BGH has to stay the NZB procedure in these cases. The BGH emphasized that the decision to stay the procedure is not mandatory but still at the discretion of the senate and that a late filing of the nullity action (as compared to the filing of the infringement action) may be of some weight in favour of the patentee. In the case at issue, the nullity suit had been lodged eight months after the pronouncement of the infringement decision and only briefly after the publication of Crimpwerkzeug III.
In cases where the nullity action is filed more or less directly in response to the infringement action, it remains the standing advice to lodge a NZB if the district court decides on the infringement without waiting for the nullity procedure as a precautionary measure.
Wednesday, 23 June 2010
STOP PRESS: KCI negative pressure patent valid and infringed
Here's a treat for PatLit readers: today's ruling by Mr Justice Arnold in KCI Licensing Inc and others v Smith & Nephew plc and others [2010] EWHC 1487 (Pat) -- all 214 paragraphs of it --is not yet available on the BAILII website, but you can read the full text here. In short, the judge found KCI's negative pressure wound drainage patents valid and infringed. I've not yet had a chance to read this decision -- but will revert to it once I've done so.
More on negative pressure wound therapy here.
More on negative pressure wound therapy here.
Subscribe to:
Posts (Atom)







