Terminal lawsuit

Standard-essential patent (SEP) litigation is something of a venture, as already illustrated in a previous post. The “terminal lawsuit” reported in today’s post provides yet further illustration. In fact, not only was this a lawsuit on a terminal, namely an electronic payment terminal, but it also turned out to be terminal for the patent proprietor’s business strategy.

The story begins, as always in such a case, with licence negotiations between a patent proprietor, Koninklijke KPN NV, which I will be happy to simply refer to as KPN, and another big player, France-based Ingenico Group. The standard at stake is the EMV standard, which stands for Europay, Mastercard, Visa. Self-understandably, this is a standard relating to payment cards, payment terminals, and the protocols according to which they interact together.

KPN owns European patent EP 0873554. Ingenico manufactures and markets EMV-compliant payment terminals and KPN argued that the patent is implemented by the EMV standard. KPN sued Ingenico for infringement in front of the Paris Tribunal de grande instance after the licence negotiations failed.

In short, KPN lost. For the long story, we need to refer to the very detailed judgment which is dated April 14, 2016.

A key part of the discussion relates to the determination of the scope of protection of the patent. The judgment contains a number of interesting (albeit sometimes controversial) statements in this respect.

Here is claim 1 of the patent. Not exactly an easy read, but we have seen worse, haven’t we:

Method of protectedly performing a transaction using an electronic payment means and a payment station, the method comprising:

– an initial step, in which:

– the payment station transfers a first random value to the payment means,

– the payment means, in response to said first random value, transfers a first authentication code to the payment station, which authentication code is determined on the basis of at least a first start value, a first random value and first transaction data of the payment means using a predetermined process, the process further producing a first end value,

– a further step, in which:

– the payment station transfers a second random value to the payment means,

– the payment means transfers a second authentication code to the payment station, which authentication code is determined on the basis of at least a second start value, the second random value and second transaction data of the payment means using said process, the second start value being based on the first end value.

The scope of claim 1 was in dispute. Claim 1 generically mentions an “electronic payment means”. But the defendant argued that the term should be construed as only covering a card containing units to be debited, and not other electronic cards such as bank cards.

This claim construction was of course vigorously fought by the plaintiff, since the allegedly infringing payment terminals were used with bank cards. Said KPN:

[…] The embodiment mentioned in the description concerning prepayment electronic cards does not restrict the scope of the invention, which relates to any payment mode using a chip card such as a bank card. The transfer of the balance can be omitted or replaced by an acknowledgment receipt or by a card authentication. […] In the claims as well as the description and drawings, in particular Figure 1 which shows a chip card used to make a payment in a shop and connected to the card carrier’s bank and the shopkeeper’s bank, the expression “electronic payment means” encompasses any type of chip card and not only a prepaid, rechargeable card. This expression, […] just like the terms “chip card” or “smart card”, covers any type of payment card, including bank cards. 

The court started with making a few general remarks on claim interpretation. The following passage in particular, on claim drafting, caught my attention and might leave some patent attorneys flabbergasted:

[…] The structure of the claim is essential for determining the […] scope of protection afforded by the right. The claim preamble sets forth the state of the art, whereas the characterizing part, introduced by the words “characterized in that”, sets out the elements which make up the invention, namely the novel and inventive means taken in their form and function, which apply to the object comprised in the state of the art and benefit from exclusive protection. The characterizing part is only considered “in relation” with the preamble because the latter is the support for the former, and because the means for which protection is claimed apply to the product described in the preamble. This relation, just like the interpretation which is made in the light of the description and drawings, does not result in an extension of protection to elements which cannot benefit from a monopoly for they are comprised in the state of the art; but it identifies the concrete subject-matter of the means which make up the invention. The violation of the exclusive right is in fact characterized not by the reproduction of elements in the preamble but by that of the means claimed in the characterizing part

Many professionals (including myself) are uncomfortable with the idea that the interpretation of a claim should depend on where in the claim the “characterized in that” phrase is placed. The division of the claim into two parts may not necessarily accurately reflect the contribution of the invention to the art: only one piece of prior art is usually taken into account when drafting a two-part claim; the applicant may not agree with the characterizing portion demanded by the examiner but may not have a lot of leeway to challenge his/her position; etc.

Although it makes perfect sense for the court to focus on what the invention is really about, relying on the two-part form to perform this analysis could lead to unfair outcomes.

The court regretted that, in the present case, “the structure of claim 1 does not comply with the implementing regulations since it does not comprise a clearly stated characterizing part“. The court seems to have overlooked that R. 43 EPC explicitly provides that the two-part form should be used “wherever appropriate“, as opposed to in all circumstances. The Guidelines for examination provide a number of examples in which it is not appropriate to use the two-part form.

Actually, after reading this judgment, I think it is certainly advisable to avoid casting a claim into a two-part form whenever possible since this may unduly restrict claim interpretation in subsequent litigation.

The court then turned to a detailed analysis of the contents of the description of the patent.

In passing, the court noted that the file wrapper should not be taken into account in the claim construction process:

[…] Examination proceedings and potential variations between the application and the granted patent are irrelevant in the context of analyzing the scope of the patent, which is solely made considering the latter, because on the one hand it defines the protection from the publication under article 64 of the Convention (since the protection conferred by article 67 to the patent application is only temporary) and on the other hand legal certainty explicitly mentioned in the interpretative protocol of article 69 demands that third parties can determine the extent of protection solely based on reading the patent without having to embark on uncertain interpretation of the applicant’s intent in view of modifications made since filing. 

Makes sense. Although, shouldn’t a distinction be made between statements made by patentees during prosecution that third parties may want to rely on vs. self-serving statements?

Anyway, back to the (complex) analysis of the description and drawings.

The court came to the conclusion that the invention is about a closed dialogue between the card and the terminal, as opposed to one which would include a bank server – as would be the case if the electronic payment means were a bank card.

One of the main arguments raised by the patent proprietor was probably that Fig.1 of the patent does schematically show the interaction of the card and terminal with respective banks. Here is what the court had to say about the drawing:

[Fig. 1] is not explained an the dotted line between the card or terminal and the bank is not defined. The association between the card and the bank and the one between the terminal and the bank are not understandable. The only exchanges described are those between the card and the terminal and then between the banks, which take place respectively during and after the transaction phase […]. And, the sentence “During a transaction there basically does not take place communication between the payment station and the payment institution in question (so-called offline system)” means that there is absolutely no exchange between the card or terminal and the bank, not only because it is defined by the expression between brackets, but also because the sentence is a reminder of the risks of abuse attributable to this lack of communication […].  

And so here is the overall conclusion:

Thus the sole payment means contemplated in claim 1 […] is a card comprising a balance of units to be debited or credited and not a bank card. The fact that the latter is also a payment means and is designated under this generic name in the EMV specifications is irrelevant. The transaction data transferred to the terminal is restricted (with the exception of authentication codes) to the balance on the card. Consequently, the protection of the process does not extend to securing bank card transactions, which necessitates an open and more complex dialogue.  

So all of this did not bode well for the patentee’s infringement claim, since Ingenico’s terminals are meant to be used for payments via bank cards.

They did preserve the patent though, as Ingenico’s counterclaim for invalidity was unsuccessful.

Among others, an auxiliary objection of extension of subject-matter was raised in case the court were to retain a broad interpretation of claim 1. That is a “squeeze“-type, either-I-am-not-infringing-the-patent-or-there-was-some-extension-of-subject-matter-but-you-cannot-have-it-both-ways-type of argument. The court bought the narrow claim interpretation and accordingly rejected the auxiliary objection.

Credit card: the nightmare of teleshopping addicts.
Suggestion for a crosswords definition of electronic payment means: the nightmare of teleshopping addicts.

A few more words on the infringement part of the judgment. First, the court noted a technical difference between the subject-matter of claim 1 and the EMV standard – which I am not going to comment on.

Second, and more interestingly, the court held that there was no contributory infringement anyway, irrespective of whether the patent was implemented in the standard.

The court stated that the statutory provisions on contributory infringement need to be appraised in a very strict manner. In particular, there can only be contributory infringement if “means for implementing the invention […], relating to an essential element thereof” are supplied by the alleged infringer.

Ingenico is accused of selling payment terminals compliant to standards which implement the patented process. […] The essential element [of the invention] lies in the dependence between the second authentication code and the first one. It is the sole novel and inventive element which provides the desired security. And it is exclusively performed by the undefined and unclaimed function of the card. […] The role of the terminal is limited to sending the card a random value to initiate the authentication step. This phase was known from the prior art and is not novel. 

It is true that the first final value is calculated partly based on this random value. But the essential element does not lie in this use but in the conservation of the cryptographic residue from the processing as a first final value, which is the basis for the second code. The terminal is not involved in this process. The fact that it is mentioned in claim 1 (in the part which should actually have been the preamble) is irrelevant. 

The invention covered by claim 1 is implemented by the function of the card, which is the subject-matter of product claims. It is not implemented by the terminal […]. 

Therefore, no matter whether the claims are reproduced (literally or by equivalence) by the EMV standards, the means supplied by Ingenico, i.e. the terminal, cannot generate the result of the invention and does not relate to a constitutive element thereof. 

The way I understand French case law, very schematically, contributory infringement applies if an invention is made of two interacting parts, for instance the wipers discussed here. But if the novel and inventive features of the invention lie in one element only, and if someone then supplies another element to be used with the former, there is no infringement.

I think the case law is clear in this respect, which is certainly a very good thing.


CASE REFERENCE: Tribunal de grande instance de Paris, 3ème chambre 1ère section, April 14, 2016, Koninklijke KPN NV v. SA Ingenico Group, RG No. 14/11998.

Hardcore litigation

A few years back, it was a common thought that standard-essential patents (SEPs) are the jewel in the crown of a telecom IP portfolio. SEPs are patents that are mandatory to work for the implementation of an industry standard, such as for instance the GSM (2G), UMTS (3G) or LTE (4G) standards in the telecom industry. Because it is not possible to make and market devices which do not implement the relevant standards, SEPs may seem like the perfect tool for putting pressure on potential licensees because any non-infringement defense is supposed to be made more difficult.

But nowadays, SEPs probably look much less powerful than they used to.

First of all, patent owners have to abide by the rules of standards-setting organizations, which require them to offer licenses on FRAND (Fair, Reasonable And Non Discriminatory) terms. Defendants have questioned whether it is allowable at all to enforce FRAND-encumbered IP rights against third parties, especially when it comes to the mother of all remedies, namely injunctive relief. This FRAND defense has given rise to complex legal discussions (revolving in particular around the issue of what is a willing or unwilling licensee) which have culminated in the CJEU.

But more fundamentally, the overall outcome of the many smartphone patent wars which have been waged in the past few years, both based on SEPs and on non-SEPs, is far from being positive for patent owners – despite a few resounding wins. Therefore, it seems that SEPs did not deliver on their promise of being easier to win on by practically excluding any non-infringement challenge.

Although SEP litigation has been much more abundant in Germany and the U.S. than in France, one case decided on last year confirms this global trend in a spectacular fashion.

Core Wireless Licensing is a monetizing entity holding a portfolio of more than 2,000 patents acquired from prolific patent applicant Nokia. The majority of these patents were declared as essential for the 2G, 3G and/or 4G standards at the ETSI (European Telecomunications Standards Institute). Core negotiated a license on this portfolio with the Korean giant LG Electronics between 2012 and 2014, without success.

In 2014, Core started legal action in France against LG. Core did not request an injunction but requested that the court should award damages for the past as well as force LG into a license and set the level of royalties for the future.

It is not possible in practice to ask a court to look at an entire portfolio of hundreds or thousands of patents. On the other hand, Core did need a finding of infringement in order to get the court to issue the requested order.

Thus, Core selected a group of five SEPs from the portfolio and claimed that these five patents were infringed by LG.

Why five patents and not just one? Well, the rationale may have been that every single patent is at a risk of being found invalid, or not infringed. But the more patents you pile up, the more unlikely it becomes, on a purely statistical standpoint, that they will all be found invalid or not infringed. The defendant needs to fend off all patents in order to win, while the plaintiff only needs one to get through. So, was this indeed a rock-solid strategy?

The answer is no.

The Tribunal de Grande Instance (TGI), defying the odds, held that there was no evidence of infringement of any of the five SEPs, and rejected all of Core’s requests.

I know that it is always much easier to explain why a strategy failed after the facts, but in my view there are two main circumstances which may account for this prima facie surprising outcome.

Contrary to what you could think, telecom litigation is not always kawaii.
Contrary to what you could think, telecom litigation is not always kawaii.

The first circumstance is that Core requested accelerated proceedings. Note that these were proceedings on the merits, and not proceedings for obtaining a provisional order. But the claimant did request and obtain an extremely tight schedule based on the alleged urgency of the case. Practically speaking, the judge immediately set a date for the trial at the onset of the action: the complaint was served on LG on September 30, 2014, and the pleadings hearing took place on January 16, 2015, the latest written submissions of the parties having been filed respectively four days and one day before that date…

This is as much of a rocket docket as you can possibly think of.

The second circumstance is that the case was extremely complex. Both on a legal standpoint, with five patents at stake and thus five separate validity and infringement discussions, plus a FRAND defense, plus an exhaustion defense based on alleged pass-through rights from the electronic chip manufacturer Qualcomm; and on a technical standpoint, because the five patents were rather abstract and difficult to grasp for a non-specialist.

We can easily imagine that the French judges, with no technical education and no technical assistance whatsoever, confronted with probably hundreds of pages of written submissions to digest containing highly sophisticated developments on signal processing, having heard oral explanations for a couple of hours at best, and having dozens of other cases on their desks to review in the same period of time, were faced with an insurmountable task.

This is reflected in the decision itself. If we take the first patent in the list, one issue with respect to infringement related to two steps in a process which appeared to be in a different order in the patent claims and in the technical specification for the UMTS standard which was relied upon by the plaintiff as evidence of infringement. The court held:

However, as rightly put by the defendants, contending that the operations of data propagation and power modification can be performed in any order in the 3GPP TS 25.213 V.11.4.0 specification is a statement that is not confirmed or justified by any evidence. 

More generally, any claimant has the burden of proving its case. It is indisputable that Core have limited themselves, in order to show that the 786 patent is essential, to filing an analysis drafted by an expert that they have hired and selected, without proceeding in a meticulous, precise and understandable way with demonstrating that the reduction in power described by the patent is necessarily reproduced by all cell phones implementing the 3GPP TS 25.213 V.11.4.0 and 3GPP TS 25.302 V10.0.0 specifications. It should be added that infringement by equivalence requires another demonstration, namely that other means can be used to implement the same patentable result, which is not present either.   

The court’s findings on the other four patents were similar: each time, the court stated that the defendants’ non-infringement defense had not been clearly proven wrong by the claimant.

The bottom line is that demonstrating that an SEP is infringed is not a piece of cake – since a declaration at the relevant standards-setting organization that the patent is or may be essential cannot be taken for granted, as the organization does not check whether this is indeed the case. So, any declaration of essentiality notwithstanding, the burden of proof remains on the patent proprietor to demonstrate infringement, and the TGI’s decision reminds us that it is a heavy one.

Indeed, the wording and technical explanations used in a patent are usually very different from the wording and technical explanations used in a standard specification. Again, unless you are a specialist in signal processing, it can prove difficult to get your bearings.

So, is there any solution at all for telecom patent owners – except waiting for the advent of the UPC, which will likely have more resources and be better equipped for handling complex litigation of this kind, notably owing to the presence of technical judges?

Well, there may be one. Article L. 615-20 of the Code de la propriété intellectuelle provides that the court may

either ex officio or upon request of a party, appoint any consultant for following the proceedings from the case management, and be present at the hearing. The consultant can be authorized to ask questions to their parties or their representatives. 

Judges without any technical background can thus get assistance from a technically skilled consultant in order to make a better-informed decision on a case.

This option has seldom been used by courts and litigants. Does it add a layer of procedural complexity to a case? Yes. Is the outcome of the case highly dependent on the consultant’s skills, not only on the technical standpoint, but also in terms of understanding the patent law issues which are really at stake for the court? Certainly. So, isn’t this a solution which is far from ideal? I would agree, but it could also be the only acceptable one in this particular context.

To finish on the Core case, the court rejected all claims and counterclaims. The invalidity counterclaims were not examined as they were submitted in an auxiliary manner. The court also held that there was no need for setting a FRAND royalty rate in the absence of demonstration of the essential character of the patents; and that there was no need to force Core to file the Nokia / Qualcomm license agreement, again in view of the essentiality finding. Furthermore, the court rejected LG’s counterclaim for damages based on Core’s alleged bad faith in the negotiations, as well as LG’s counterclaims for abuse of dominant position and for abuse of procedure.

The court’s position on the negotiations between the parties is actually noteworthy:

[…] the negotiations between the parties lasted for more than two years, which tends to show that neither of them behaved with enough bad faith to prompt the other one to stop this phase. 

Besides, the parties blame each other so that the court, which does not know everything about these negotiations, is not in a position to state that bad faith is more on one side than the other. 

Again, the message is clear: each party should prove its case, and allegations will not suffice.


CASE REFERENCE: Tribunal de Grande Instance de Paris, 3ème chambre, 2ème section, April 17, 2015, Core Wireless Licensing Sàrl v. LG Electronics France & LG Electronics Inc., RG No. 14/14124.