For a couple of years, patents have hit the headlines with companies struggling to buy out portfolios of bankrupted competitors, with more and more ridiculous obvious patents granted by patent offices, or with “trials of the century” going on and on. This inflation of concerns around patents has culminated on August 24th, 2012, with Samsung being found liable for infringing some of Apple's mobile patents by a Californian jury. This over one billion dollars fine has given concrete expression to Steve Jobs' testimony, as laid down in his posthumous biography [2]: “I’m going to destroy Android, because it’s a stolen product, I’m willing to go thermonuclear war on this.”
This article does not recount such events. Instead, its purpose is to show with a particular Apple's software patent, which was recently successfully enforced before a German court, how such a thermonuclear patent war is threaten to explode in Europe, would the current project about a unitary patent be adopted.
The weapon of massive destruction
When any war is to be studied, a main question is to know exactly which weapons are used to cause damages. Here, we are going to focus on a patent held by Apple, which was included in the US utility patents Samsung has been convicted – although not in final instance – to infringe. It seems also to be of a special importance for Apple, since this one patent is tried to be enforced all over the world against various Android phone makers. And on September 13th, a Munich court has found [3] it infringed by some Motorola Mobility – a Google subsidiary, bought for $12.5 billion in 2011– devices.
This patent, commonly known as “rubber band” or ”overscroll bounce” patent, is relatively easy to understand. When it comes to understand the scope of a patent – in order to know whether a competitor product does infringe, and if it does, what is the exact part of the infringing product which is guilty, and how –, the patent claims, as written down in the granted letters patent, are to be read carefully. Because patent claims are the exact written definitions of what is included in right to exclude conferred by the patent.
The main claim of the ”overscroll bounce” patent can be read [4] as followed (references to drawings omitted):
1. A computer-implemented method, comprising: at a device with a touch screen display, detecting a movement of an object on or near the touch screen display; in response to detecting the movement, translating an electronic document displayed on the touch screen display in a first direction; characterized by, in response to an edge of the electronic document being reached while translating the electronic document in the first direction while the object is still detected on or near the touch screen display, displaying an area beyond the edge of the document; and after the object is no longer detected on or near the touch screen display, translating the document in a second direction until the area beyond the edge of the document is no longer displayed.
Written like this, the patent may eventually not be so easy to understand. But this is the typical language used in a patent claim. And this particular claim is quite clear in comparison with many other software patent claims. If you read it slowly again and again, you are likely to understand it. While many software patents, as written in patent claims, are not understood even by the computer engineer who has written the related computer code.
The trick is to understand that the “object” as referred to in this claim is a finger – which is the subject of claim 3: “3. The computer-implemented method of claim 1, wherein the object is a finger.” – or any other means that can be used as input for a touch screen device, such as a stylus, since the claim is drafted as to include the broadest possible means. With this information in mind, the claim is easily understood to include the effect that happens on a smartphone or a digital tablet, when you scroll down on an electronic document, and when reaching the end of the document, you continue to scroll down, then a grey – or black, or white, or whatever – area appears after the end of the document, so you are aware that the end of the document as been reached, so you release your finger, which leads to the document been bounced in the opposite direction, until the end of the document is displayed on the bottom of the screen. Usually, when you reach the end of the document, the scrolling just stop, and you may wonder whether the touch screen is just failing to reply to your finger's input. Conversely, this patent claim discloses a method to make you aware that you've reached the end of the document, so you are encouraged to release your finger.
Here's an example of how this can be easily implemented in a program code:
while (user_input == 'down') {
if (current_pos <= end_doc)
scroll(doc, ++current_pos);
else
scroll(doc + grey_area, ++current_pos)
}
if (current_pos > end_doc)
bounce(doc, end_doc)
But how brilliant is this idea is not really the problem. The main issue is that the teaching of this patent claim is nothing more than processing data. Indeed any implementation of this claim amounts to detect a user input – scrolling down – on a document, and if a condition is met – the end of the document is reaching the bottom of the viewable area — to triggers a further data processing – display a uniform area behind the document – until another condition is fulfilled – no more user input —, triggering a final data processing – display the bottom of the document with an animated “bounce back” effect.
This is a software patent, in the sense that the teaching of this patent lies only in a data processing.
Is this a conventional weapon?
Actually, in Europe, the patent law, as laid down [5] in the European Patent Convention (EPC) – an international agreement that has been implemented as of today in 38 national laws –, explicitly forbids the patentability of software innovations (emphasis added):
Article 52 – Patentable inventions
(1) European patents shall be granted for any inventions, in all fields of technology, provided that they are new, involve an inventive step and are susceptible of industrial application.
(2)The following in particular shall not be regarded as inventions within the meaning of paragraph 1: (a) discoveries, scientific theories and mathematical methods; (b) aesthetic creations; (c) schemes, rules and methods for performing mental acts, playing games or doing business, and programs for computers; (d) presentations of information.
(3) Paragraph 2 shall exclude the patentability of the subject-matter or activities referred to therein only to the extent to which a European patent application or European patent relates to such subject‑matter or activities as such.
Clearly, the above claim relates to a program for computers, which is nothing more than a mathematical method – every data processing can be written as a mathematical equation. Moreover, it is intended to answer to a mental act – scrolling down a document, following a user's intent – and proposes a solution consisting in a presentation of information – displaying an area below the end of the document and bouncing back the document. Obviously such a software patent provides all reasons for not have been regarded as a patentable invention, according to the EPC.
Nevertheless, the patent was granted. The European patent office (EPO) – the patent granting administration set up by the EPC – has actually adopted for more than 20 years a practice based on the “as such” restriction on the exception to patentability, laid down in Article 52.3 EPC. For the EPO, the “as such“ wording means that the exclusions of patentability defined in previous Article 52.3 are effective, if and only if the claimed invention does not imply any “technical” effect. If the claimed invention solves a technical problem, is based on a technical solution or implies technical means, then the claimed object is not regarded as such, and can therefore become patentable.
The flaw in this rationale is obvious: everything rests on the definition of what is technical. But the EPO has never ever defined the notion of technical – even confessing in a decision [6], that “its comparatively broad interpretation of the term ‘invention’ in Article 52(1) EPC will include activities which are so familiar that their technical character tends to be overlooked, such as the act of writing using pen and paper.”
The “technical” contribution of the ”overscroll bounce” patent is not so obvious. The first examination at the EPO has wondered [7] what was the “technical” problem this patent was supposed to solve:
The subject matter of claim 2 [which has been subsequently renumbered to claim 1, as quoted above] differs from the disclosure of document D1 [US 2004/021676 A1] in that the document is allowed to be translated beyond its edge, which is then aligned with the edge of the view window, when the finger movement is no longer detected.
It is not clear what technical problem is solved by these features. The application identifies the problem to be solved as implementing the translation in such a way as to reflect the user's intent. However, it is clear from the wording of claim 2, that "the object is still detected on the touch screen display", even after "an edge of the electronic document being reached", that the user's intent is not to align the document's edge with that of the view window but to keep translating the electronic document beyond its edge; hence, the aforementioned features go against the user's intent.
It is to be noted that the EPO examiner does not even asks whether the claimed subject matter is patentable, the practice of the EPO is to ignore exclusions of Article 52.2 EPC, and to just check whether the claimed invention is novel, that is: whether there is any prior art disclosing what is claimed, and whether the “technical” features of the claimed invention involve an inventive step, that is: whether there is a non-obvious contribution to what was existing before the claimed invention — the state of the art. Since this first examination hasn't understood what could be the “technical” problem at stake, it has concluded:
Since, depending on the definition of the objective, the additional features defined in claim 2 either do not serve to solve a technical problem (i.e., they are pure design options) or correspond to a known user-interface design technique, the subject matter of claim 2 does not involve an inventive step in the sense of Article 33(3) PCT.
But the applicant has the opportunity to discuss with the EPO examiner. The justification [8] of the “technical“ objective of the claimed invention by Apple's patent attorney is indeed worth to read (original emphasis):
(A) in response to an edge of the electronic document being reached while translating the electronic document in the first direction while the object is still detected on or near the touch screen display, displaying an area beyond the edge of the document; and
(B) in response to detecting that the object is no longer on or near the touch screen display, translating document in a second direction until the area beyond the edge of the electronic document is no longer displayed.These combined differences lead to the technical effect that a very eye-catching yet simple visual indication that one or more edges of an electronic document are displayed is provided to a user, thus effectively signalling to the user to stop making scrolling gestures, at least in the given direction. Without such an eye-catching indication a user, probably assuming a sluggish alignment of the touch screen response with the object movement, would continue making scrolling input gestures although the edge has already been reached, as it would take a little while until the user would realize that he can stop his gestures because they will not lead to any further scrolling action. Since scrolling basically means to skim an optical representation of a data base for retrieving a data item, the invention must be seen in the overall context of data retrieval in a electronic device, which is generally considered to be technical subject-matter. The invention makes the retrieval process faster in that multiple superfluous input attempts are avoided when the end of the data base (document) has been reached. This is achieved by visually indicating the internal state of the device to the user, i.e. that the scrolling mode is no longer available (at least in the direction chosen) and that the device has switched to a steady state, thereby prompting appropriate further user interaction with the device.
The objective technical problem solved by the present invention can therefore be seen in providing a user a faster scrolling method that allows an easy distinction of internal scrolling and non-scrolling modes of the electronic device, thus aiming at a more efficient man machine interface for searching in data bases. This relates clearly to a technical problem as the operability of the device is to be improved.
So that is what the claimed invention is really about: a more efficient man machine interface for searching in data bases? But – wait a second – this is precisely what computer programs as such are made for: retrieving, processing, outputting data from and to a man machine interface! If this is not a software patent per se, nothing is. Nevertheless, Apple's patent attorney has just spoken the truth: “data retrieval is generally considered to be technical [i.e. patentable] subject matter” in the usual EPO practice.
This was enough to convince the EPO that the ”overscroll bounce” was indeed patentable. And the reply [9] from the EPO went even further:
the applicant submitted that the objective technical problem solved by the method of claim 1, in view of the closest prior art, is to inform the user about the internal state of the device, i.e., that the device is operating properly and is receiving the user input, and to avoid multiple superfluous input attempts (e.g., if the user perceives the lack of response to input gesture, as per D1, as a faulty input operation) when no more scrollable content is available.
Here we are: “informing the user” has become a “technical” problem. With this one and only software patent being granted, the EPO has thrown out no less than four of the exceptions to patentability as laid down in the EPC: mathematical methods – we've disclosed above an algorithm, and algorithms are a part of mathematics –, methods for performing mental acts – since the “overscroll bounce” patent is aimed to help the user to realise he has reached the end of a document –, programs for computers — we have shown this was a software patent — and, presentations of information – since the problem is said to be an information presented to the user. Definitely, this is not a conventional weapon!
The current theatre of operations
But, you gonna ask, isn't there some kind of United Nation Agency that can take action against the proliferation of unconventional weapons? Well, there is such a kind of forum, the problem is that it is fully included inside the EPO itself. When someone thinks that a patent has been mistakenly granted by the examining division of the EPO – or, conversely, when an applicant thinks the examiner should not have rejected her patent application –, she can file a complaint before the opposition division of the EPO, up to nine months after the decision to grant – or respectively to reject – the patent.
The ”overscroll bounce” patent is currently opposed before the EPO by three parties: Motorola Mobility, HTC and Samsung. But it should be noted that none of these opponents is opposing the ”overscroll bounce” patent because it's a software patent that should be excluded from patentability according to the EPC. Rather, they have made extended researches to find some documents in the prior art proving that the alleged invention is not new or does not involve an inventive step. The reasons are easy to understand. First, the opposition division of the EPO follows the same practice and interpretation of the EPC than its sister examining division, therefore there is no need to argue on the patentable subject-matter requirement which would be regarded – sorry, not even regarded — like it was – not – during the examination. Second, the opposition makes no sense for a field which is excluded by law from patentability. Economic players of this field are not supposed to look at patents. And those who look at patents are also patent holders who do not want a decision excluding their own patent altogether with the one they are opposing to. Indeed in this case, the three big firms who are opposing Apple's patent have also a big portfolio of software patents.
And once there is a decision from the opposition division of the EPO, if third parties or the applicant is still unsatisfied, they can make a recourse before a Technical Board of Appeal (TBA) of the EPO. TBAs are quite like courts, and in very exceptional case, their decisions can be reviewed by an Enlarged Board of Appeal (EBA). But all these Boards of Appeal (BoAs) are not at all independent courts, in compliance with the EU Charter of Fundamental Rights, nor with the European Convention on Human Rights. The judges sitting in BoAs are not only legally qualified judges as usual, but also some technically qualified engineers with some basic legal training. Moreover, BoAs are administratively and organisationally attached to the EPO. Given this background, there is no surprise that the practice of the EPO to grant software patents despite their exclusion according to the EPC, has followed some BoAs decisions which have been codified in EPO Guidelines for Examinations. In short, the “judicial” branch of the EPO is tied to its executive authority. And do not count on a legislative body to guarantee democratic checks and balances: the legislative body of the EPO is its Administrative Council, mostly composed by heads of national patent offices. The bottom line is that the EPO has a major role – it can even be said the first role – in making patent law.
Whatever flawed is the opposition procedure of the EPO, it should be underlined that it comes a posteriori. The patent has already been granted, the weapon is already ready to fire. And that is exactly what Apple is doing by suing various Android phone makers in various European courts – and also all over the world, but this article is focused on the old continent. For accuracy, once the patent has been granted by the EPO, there is some additional requirements, it should be validated by each national patent office designated by the patent holder. But this is merely a formal step, where the only requirement is usually the filing of the translation of the patent as a whole, or just of the patent claims, in a national language. This translation requirement is even relaxed for countries having signed the London Agreement [10], an optional complementary international agreement to the EPC that dispenses with the translation requirements.
Then, the current battlefield for patents is national courts. National courts have competence to decide whether a patent has been infringed and impose penalties in the case it has, including preliminary or permanent injunctions in order to refrain the alleged or established infringer from selling the alleged or established infringing product or service. The scope of their rulings is of course limited to the national territory. This means that, in theory, a patent holder who believes that her patent is infringed all over Europe, has to sue the alleged infringer before each and every national European court. Just theoretically, because practically, once there is one final ruling from one national court, in most cases the parties are likely to consider that courts of other countries would adopt the same reasoning. Instead of wasting time and money in more pending lawsuits, they try most of the time to find an out-of-court agreement.
In our case, the decisive ruling came from the Munich district court. Although we have not been able to read this judgement, but it can be safely inferred that it came at no surprise. German courts have actually adopted for more than a decade a case law which is fully in line with EPO practices. An overview of the case law in every Member state of the EU would go beyond the scope of this article. Suffice to say that the patent litigation activity in EU are only significant in four countries: Germany, France, United-Kingdom, and Netherlands. According to a study [11] ordered by the European Commission, these four countries “accounts for about 90% of all patent litigation activity in the EU“. France case law about software patents is the most reluctant [12] to confirm the validity of software patents, but there are no clear and constant jurisprudence [13]. UK courts, after having developed a test excluding [14] most software patents for being no more than mental acts [15], have finally given in [16] to pressure from EPO, after the EPO has declared [17] the UK practice irreconcilable with the EPC. We have already underlined a case [18] in The Hague, where an Apple's software patent has been enforced against Samsung, resulting in a preliminary injunction. What should be retained is that there is undoubtedly an influence by the EPO on national patent courts.
But the German patent court system has some features that make it specifically attractive for patent owners. We are going to detail these features right now, since the Unified Patent Court, as currently envisaged by the EU, would adopt such features.
The thermonuclear escalation
After having outlined in details the current state of play with regard to software patents in Europe, it is now time to look at the envisaged evolution. The European patent system is currently under reform by the European Union (EU). Instead of having a bundle of national patents granted by the EPO, and then validated in each desired national patent office, a new unitary patent is supposed to be created, without any additional requirement for validation, nor for translation. The EPO will still be the granting authority for the unitary patent, but its territorial scope would expand to 25 countries at once – Spain and Italy having refused to join this project. Additionally, the uniform effect of the unitary patent would be enforced before a Unified Patent Court (UPC). Litigations about usual EPO patents would also be ruled by the UPC, depriving national courts of their current competence.
Since the EPO would grant unitary patents, just as it currently grants bundles of national patents, one can wonder whether there would be any actual change with the unitary patent/UPC project. The answer is unfortunately and undoubtedly that there would be some changes. For several reasons, the situation with regard to the damages caused by software patents in Europe would be worse, with several degrees of magnitude.
First, the unitary patent would introduce, by definition, a change of scale in the conferred rights. Since a patent is a State granted right to exclude competitors from using an inventive technology, the unitary patent would extend such an exclusion power from a few national territories to the whole EU – minus Spain and Italy.
This can be easily understood with our example of the ”overscroll bounce” patent and its enforcement granted by the Munich court. Since this German court has found that Motorola Mobility has infringed Apple's patent, the latter is entitled to be granted an injunction against the former, forbidding the sell of infringing products in Germany, and even forcing the destruction of existing products. With the unitary patent and the Unified Patent Court, such a blocking or destroying would cover not only the German market but the whole European market – minus Spain and Italy.
Therefore, the first change introduced by the project would be an expansion of damages caused by patents, including software patents, granted by the EPO.
Second, since the main claimed objective [19] of the project is to make “access to the [European] patent system easier [and] less costly”, it can be inferred that this would lead to a significant increase in patent applications, grants and enforcements. Otherwise, it would be a failure of the project. But the side effect of such an inflation, as it has been observed [20] in the United-States patent system, is that much more dubious patents, and specially software patents, are granted, resulting in a patent litigation explosion [21].
The simple reason for this is that there is a pressure from patent applicants on the granting patent offices, leading to a lower quality of examinations. Additionally a similar increased pressure is put on patent judges to deliver fast rulings, also in detriment of their quality. Such a pressure is caused by the fear of an unworkable increase in backlogs both for patent applications and patent litigations. Moreover, as we have seen with the ”overscroll bounce” patent, software patents are more particularly exposed, since software innovations are easy to develop – just look at the simplicity of the above example of algorithm — and abundant – a single smartphone is said [22] to include hundreds, if not thousands, of such patented algorithms.
Therefore, a second change introduced by the project would be an explosion of litigations over bad quality patents, and especially software patents, granted by the EPO.
Third, as we have anticipated in the previous section, the UPC would borrowed some worrying features to the German patent litigations model. It is to be noted that the German government has lobbied [23] the Council to include these features in the envisaged unified patent jurisdiction, as soon as October, 2007.
The main feature is called “bifurcation”, or “separation of invalidity and infringement proceedings”. It means that an accused patent infringer cannot raise a defence based on the invalidity of the patents at stake. Instead, the defender should file another lawsuit to revoke them. The purported advantage of such a separation is to speed up infringement proceedings. Indeed, in the German patent litigation system, as well as in the envisaged UPC, the judge in the counterfeiting trial is not forced to stay the trial until a definitive ruling about validity – or invalidity – of the patents is pronounced. This allows to avoid examining in details technical considerations defining the scope of the patents, and to focus on the less difficult issue whether the a priori valid patents are counterfeited.
The obvious disadvantage of a bifurcation system is that it is unbalanced, in favour of patentees. These patent holders can indeed succeed in getting an injunction, even if the patents at stake are judged to be non-valid. There are of course some safeguards, for instance the patent holder has to post a bond for the injunction to be enforced, but the interdiction to sell a product or a service can be fatal to the defender, specially when it is a small or medium enterprise (SME).
It should be underlined that the will to introduce bifurcation in the unitary patent/UPC project is so strong that the regulation on the unitary patent has rightfully included provisions about direct and indirect counterfeiting and limitations of rights conferred by a patent, but substantive patent law defining the patentability has deliberately been excluded from this regulation – which in our opinion [24] is not compliant with the EU treaties. This implies that the Court of Justice of the European Union (CJEU) would have competence to review decisions with regards to counterfeiting, while decisions about invalidity would be out of its reach.
Another feature common to the German patent litigation system and the envisaged UPC is to have some technical judges sitting together with legal judges. These technical judges are engineers, with a minimal training in patent law. Their technical background is supposed to help the understanding of technical considerations defining the scope of patents. But, here again, the drawback is that technical judges are likely to neither question practices defining patentability, including against the letter and the spirit of the written patent law, nor to balance rights conferred by patent law which are in conflict with some other legal areas, such as competition law or fundamental rights and freedoms. Once again, this confines decisions to the patent microcosm in favour of patentees.
Therefore, other changes introduced by the project would result in the lack of questioning of the legality of the EPO practices to grant software patents.
Fourth, the drawbacks underlined above about technical judges are amplified by the specialised character of the Unified Patent court. There is a constant claim [25] by the patent microcosm that patent law is so technical, that only specialised courts can deliver quick and above all predictable rulings. This claim amounts to ask for a jurisdiction of exception, with no regular judge but only members of the patent microcosm.
Fortunately – or unfortunately – the United-States patent litigation system provides a disastrous example of impacts generated by having a specialised patent Court. For thirty years, appeals on patent litigations are centralised in the US Court of Appeal for the Federal Circuit (CAFC). Observers [26] and academic literature [27] are unanimous to assess that such a specialised patent court has led to a bias in favour of patentees, and an extension of patentable subject matter, including software. The US Supreme Court has been forced many time to correct the drifts of the CAFC. Unfortunately, with regard to software patents, the highest court has not taken yet the opportunity to clearly state the exclusion of software from patentability.
Moreover the permeability between the UPC and the patent microcosm is further endangered since there is no guarantees that members of the EPO or national patent offices cannot also sit as judges for the UPC. Such a minimal guarantee was included in previous versions [28] of the text until March 2008, stating that “Members of Boards of Appeal of a national patent office or of the European Patent Office shall not be eligible to serve as a Judge of the Court until expiry of a period of six months after the termination of their previous function”. And a comparable provision has even be voted [29] by the Committee on Industry (ITRE) of the European Parliament in November, 2011: “3a. Suggests that Members of boards of appeal of a national patent office or of the EPO shall not be eligible to serve as a judge of the court until the end of a 6-month period after the termination of their previous function so as to guarantee their neutrality”. But the rapporteurs for the European Parliament have decided to not retain this minimal safeguard.
Therefore, the change introduced by the specialised character of the Unified Patent Court underlines the need to have UPC decisions subject to a review from a superior generalist independent court, namely the CJEU.
The result of this is that the thermonuclear patent war has its epicentre in the United-States. For now. But, as we have shown, the unitary patent and the Unified Patent Court are more than likely to make it explode in Europe too.
To avoid such a disaster, we have crafted some amendments [30] which fix most of the features underlined above: clearly reaffirm the non-patentability of software, and put the European patent system under EU control, instead of waiving more power to an unaccountable and uncontrollable EPO. We have also set up a tool so that you can contact [31] – free of charge – Members of the European Parliament and ask them to support these amendments.