Shortly before the term of a patent (if granted) would expire, in an ex-parte order issued by the IPAB authored by Justice Manmohan Singh, the IPAB finally granted a patent to Ferid Allani for an application filed almost two-decades ago for an invention relating to ‘A method and device for accessing information sources and services of the web.’
Brief prosecution history of the application – from filing to grant
Ferid Allani had an idea and filed a provisional patent application for it on 30.12.1999 in France and it was given the number 99/16704. Thereafter, a PCT application was filed on 29.12.2000 (PCT/FR2000/003759) and it entered India as IN/PCT/2002/00705/DEL on 17.07.2002. As such, the term of the patent expires on 29.12.2020 (twenty years is the life of a patent). The IPAB granted the patent on 20.07.2020 (i.e. about 5 months before expiry).
Critique: Even though the judgement takes pain to explain what the technical effect of the invention is, and the differences in the prior art cited, what is missed is whether the claims incorporate the features having the technical effect. Hence the claims run afoul of section 10(1)(c) of the Act – invention is limited in scope to the claims.
In my view, the patent claims are directed towards a structed search syntax generation and displaying corresponding results to user. A structured search syntax is user generated / based on user input logically created. For example, sports à teams à merchandise if the search is for a merchandise only.
The judgement (Para 23.e) provides,
e. Through the preliminary selection steps of the present invention, the query of the user is refined and narrowed down at each step so that the user is interactively made to target a particular and well-defined information and a website address is particularly generated locally (without using the internet bandwidth). Thereby, the user is taken directly to the said website, limiting the use of the bandwidth resource only once per search.
About the claim: The way I read the claim (the key feature is that it stores certain pages in a tree structure (like a directory and files within it), sends out a query based on user input and displays to the user results based on query and populates the display based on information so obtained.
The claim here does not mention how the internal directory structure is created. It just says…beforehand locally stored within said communication device…This means that the user provides the necessary feedback to structure the query.
Compare this with the US and European claims: both are narrower and include the invention.
| US Claim (Granted) | | 1. A method for accessing information- or service-providing web sites, from a mobile communication device connected through a communication network and provided with a display, comprising: a preliminary search for information or service, locally performed within said communication device in different themes and sub-themes within a plurality of successively displayed selection pages, said selection pages being organized in a structure of tree menu and beforehand locally stored within said communication device, each said selection page comprising a set of icons including one or more direct access icons for directly accessing an information- or service-providing web site and one or more selection icons for locally accessing and displaying another selection page within said structure of tree menu, wherein at least one of said selection pages comprises two or more direct access icons, and a step for emitting on the communication network, in response to a selection of a direct access icon in a selection page being currently displayed, an address of an information- or service-providing web site corresponding to said selected icon, said address being locally generated during the display and selection steps and directly accessing said information- or service-providing web site.
A method for accessing web-based information sources and services from mobile equipment connected via a communication network to the Internet, said mobile equipment comprising means for locally storing a set of organized selection pages at the within a menu tree structure and each containing a set of icons, characterized in that it comprises: a set of preliminary selection steps, including selection icon activations provided to access locally from a selection page to another selection page within said tree structure, until reaching in one of said selection pages a so-called direct access icon corresponding to a desired remote information source, a step for finally transmitting on the communication network, in response to an activation of said direct access icon thus reached, a complete address of access to said remote information source sought.
What is the invention: From the specification:
This part (about user choosing the tree) is simply stated as organized within the structure of a tree menu… and in my view should have been expanded to include this particular user input.
Second, I see this as a GUI (graphical user interface) to conduct a search which is itself optimized. Of course, search optimization is not new, nor was new, in 1999 (before the priority date). We had super computers which used time sharing concepts and had highly structured query systems for users to be able to make use (because time was a constraint, the problem had to be entered efficiently / optimized).
Consider an article of 1991, by I.V.Pottosin discussing ‘Analysis of Program optimization and further development’. In the article the author notes that, ‘[F]or contemporary supercomputers, the requirement of the proper use of hardware peculiarities is added’ ...in addition to… ‘saving computer resources such as time and memory’. As such query optimization is well known in the art given that the cost of creating a software to run the supercomputer was almost equal to the hardware cost in the mid 1980s.
For more information, one may also see the interesting book by Russell C. McGee titled “My Adventures with Dwarfs: A personal history in mainframe computers”, available here, and on time-sharing of mainframes – The Origin of the VM/370 Time Sharing System by R.J. Creasy – available here.
If one sees the history of Graphical User Interface (GUI) (see Arstechnica article here), a search system like that of the operating system can be built using the GUI. How does one search in the Windows / Linux / X-Windows / Apple GUI environments ? This was available in the 1990s to a person skilled in the art and whether it is obvious, is debatable.
But what is not debatable is that the key part of structing the query is missing from the claim language. This is where the invention resides and its absence in the claim is glaring. In 1999 when the invention was made, the basic ingredients to reach the invention were there – search optimization to a query and GUI were both available – search optimization from decades of mainframe time sharing and GUI from the Windows system.
I have no doubt that the combination of search (for something on the web) with GUI was not there in 1999 but this aspect is missing in the claims.
Second, the patent office should have been clearer in its rejections – a two-page rejection for a matter that has travelled to both the IPAB and the High Court shows that it is somewhat different from other ordinary matters. Repeating a citation with the same rejection is a recipe for inviting trouble (for the patent office) – the least the patent office could have done at any point in the history of this application – is to at least have its cited references properly with specificity. This should have not been left ambiguous.
Third, the patent office was not represented in during the hearing at the IPAB. Given that an expert body had already provided its views to the applicant multiple times – both before and after it went to the High Court and IPAB – the IPAB should have asked for written submissions from the patent office. This is even more important because the IPAB itself had rejected the patent application once before.
Further, given the IPAB had a different view – it should have heard the patent office – otherwise what stops any applicant going to the same forum and re-agitating the issue just because the bench composition is different.
There is no reasoning given by the IPAB on this aspect – what changed in the intervening period – that the application became patentable.
Other issues are highlighted as they appear in the post.
Journey of the application:
Patent Office ‹----› High Court ‹----› IPAB ‹----› High Court ‹----› Patent Office
The arrows above are only indicative of the process followed by the application. A request for examination for the patent application was filed on 19.11.2004 and the application was rejected by the patent office on 21.02.2005 on the grounds that the subject matter lacked an inventive step (Section 2(1)(j)) under EP0847019 (hereafter D1).
The Applicant then on 17.09.2005 replied to the First Examination Report (FER) arguing that the application was patentable. Four days later, on 21.09.2005, the patent office again rejected the claims as being unpatentable under Section 3(k) of the Act. No response was filed to the same and the application was deemed to be abandoned under Section 21(1) of the Act.
The Applicant then filed Writ Petition 6836 of 2006 challenging the order of abandonment on 2.05.2006.
The High Court of Delhi on 25.02.2008 directed the patent office to review the application by providing necessary opportunity of oral / written hearing.
The Applicant then filed a further response to the objection / rejection on 20.03.2008. The agents appeared for the hearing and submitted written submissions on 12.08.2008.
A couple of months later on 18.11.2008, the patent office again refused the application. An appeal against this refusal was filed before the IPAB on 20.02.2009 and the IPAB too dismissed this appeal.
Then an appeal against this IPAB rejection was filed before the Delhi High Court on 19.12.2013 and the DHC on 12.12.2019 disposed off the petition and directed the patent office to re-examine the patent application.
Then a hearing was held before the patent office on 27.01.2020 and on 07.02.2020 it once again refused the application under the same grounds that it lacks novelty, etc. under Section 3(k).
As already mentioned, the patent office should have been more clearer in its 2020 rejection and should have properly made out a case for rejection rather than repeat the content of the previous rejections.
That said, if one sees the delay – most of it is in courts – rather than before the patent office – so one really cannot point out the cause of delay as our patent office.
Analysis of the patent by the IPAB (Para 20 onwards)
The IPAB first discusses the state of art and the problem that was intended to be solved. In para 23, it provides:
Then in para 24, FIG. 1 of the application is reproduced with annotations (presumably provided by Applicant / agents).
Till here, the IPAB also structures the decision and identifies the key feature of the invention as essentially – forming a well-constructed query…emitted to the internet..
Given this background, the technical effect of reduction of bandwidth by a structured query is guaranteed.
Paragraph 27 of the decision discusses an example disclosed in the invention disclosure – but because it is not present in the claims it is kind of pointless in discussing it. This is what should have been in the claim : i.e. this is the specific tree structure of icons as thought of by the inventor – and should have been claimed. Because the current claims do not discuss this, they are overly broad.
Paragraph 29 of the decision provides more details as regards the invention:
Paras 30 – 41 discuss that the patent office had not applied Indian but EU law as regards computer related inventions. Here, both Indian law (even if one considers the CRI guidelines issued by the patent office) or European Law, or for that matter even US or Japanese patent law –the basic principles are the same: one has to first construe the claims, then identify the contribution made, then determine whether it falls under any exclusion and finally to determine if there is any technical contribution. Here the patent office did exactly that – it found that claims as provided by the Applicant were not inventive – as they did not contain the key inventive step – of the creation of the tree structure and the query stopped. The patent office cannot be blamed if the applicant despite several opportunities does not bring the claim in line with the requirements. Para 43 identifies that 19 ½ years have passed – but as identified earlier - the delay is not at the patent office end – but at the court’s end either at the High Court or the IPAB. In all about 12 years were passed just in courts – patent office took about 6 years including the time for filing of request for examination and about 1 ½ year in deciding multiple petitions filed by the Applicant.
In conclusion, this case will change the patent law landscape as it relates to software patents in India unless appealed. This is because courts now favour grant of patent applications for software.
Recently a German court, in a SEP / FRAND matter involving Nokia and Mercedes maker Daimler Motors, issued an injunction against Daimler for its infringement of Nokia’s SEP. See press release issued by the court here in German (you may choose to automatically translate the press release in English).
Basically, Daimler had infringed Nokia’s patent EP2981103. This patent is related to Zadoff-Chu sequences. You might ask what is so great about Zadoff-Sequences: well they are the basic access signal in LTE. In particular, the patent uses ZC sequences in the preamble for random access (i.e. to connect the mobile unit to the base station) / PRACH signal. Those that are technically oriented may see TS 36.211 to see the use of the signal.
Florian Mueller, who broke out the news in his blog Foss Patents notes that Nokia may have to give a bond of € 7 Billion (yes – you-read-that-right) Billion – with the capital B.
Florian also points out the Court failed to follow an injunction roadmap as laid down in Huawei v. ZTE – with the patentee first making a FRAND offer, and the would be licensee would then have to give a counter offer (with reasoning as to why its offer was FRAND and not the licensors).
He also points out that the judgement was given despite the fact that Daimler’s component manufacturers approached Nokia to get a FRAND license – but were not licensed.
Whether or not Nokia is fined for obtaining an injunction against a willing licensee (suppliers included) as per Motorola Samsung cases remains to be seen. I would say that this is a shocking development. In the rest of the post, I will consider the implications for India.
For starters, I am thankful that this patent is not registered in India or else everyone from mobile phone manufacturers to car / truck makers would be under its domain.
And because it is not registered in India, this is evidence that my theory regarding illegality of portfolio licenses. This is a prime example of why each patent should be held to be consistent review and not all patents be licensed under portfolio terms. Just imagine, this is a key patent for Nokia – as it pertains to the very basic structure of the LTE uplink signal. All devices that use LTE, infringe the patent, and it is immaterial whether a person is in Europe, India, or the United States (as all use LTE, but see caveat below). But surprisingly it is not even filed here in India. And when it is licensed to anybody under a portfolio here in India, it is brazen violation of the law. What is the value of something that is not even registered here?
This also proves my theory about perversity of confidentiality clubs – this is the kind of information that is designed to be protected by the licensor. The inclusion of this patent here in a portfolio license for an India specific license is illegal.
Caveat – China is not covered because it was smart enough to develop its own LTE standard – called TDD LTE – and is somewhat different than the one is used elsewhere. And Chinese companies do not have to pay such absurd royalties to companies that are not from China. I wish we too had such a system! But anyhow, back to the post.
One can only wonder why this patent was not filed in India – perhaps it is clear that it is hit by Sections 3(m) and 3(n) (because it is presentation of structure of a signal) and also Section 3(k) and we may speculate that the person responsible took the call that it was not worth it to spend money here. Whether or not my speculation is correct about the person responsible, I believe that our law has moved to the opposite end of the spectrum given the recent Ferid Allani judgement.
Recently, there was a post on spicyip.com regarding merits of confidentiality clubs to standard essential patent (SEP) litigation. When I read it, my view was that the confidentiality clubs being applied to a defendant in a SEP litigation was highly problematic and the same was not suited for the defendant: the very basis for using confidentiality clubs to share information in SEP litigation was highly questionable and should be challenged by a defendant.
This post concludes that it is not in a defendant’s best interests to agree to a confidentiality club arrangement in SEP litigation and provides several reasons for the same.
Having observed and being part of multiple patent litigations in India and elsewhere, I view a patent infringement trial as an ordeal. Patent litigation anywhere across the globe is the domain of the right holder, and if I might add, of one with deep pockets.
It is easily observable in any patent litigation that there is a basic minimum economic qualification. Right at the outset of the lis between parties, one element that has nothing to do with competing claims, is introduced. This element is the financial strength: ability to endure trial. This element is extremely important. A generalization can be made that a conclusive decision after going through a full trial, including evidence, expert witnesses, etc. is possible only in the case where both parties are evenly matched or are economic equals. This might be seen currently in Europe where several big-ticket SEP litigations involving vehicles are underway and both parties are evenly placed (Readers may be interested to go through the summary of the ongoing litigation between patent owners and vehicle manufacturers such as Diamler etc.).
One can also observe from past cases, that if a smaller party decides to defend, he may win at law, but still lose business. If he capitulates and settles, he sacrifices his independence, his competitive position, and in all probability some part of his patent position as well. It is immaterial whether the bigger party has dubious patents/ vice versa. The results to the competitive system and the independent judicial system are the same.
It is essential to recognize that when a patent suit is decided, something more than the respective rights of the litigants is involved. A patent infringement suit is predominantly a problem in public interest – it is in the public interest that the patent system finds its justification – and a problem in the relation between economic law and technology. The rights of the litigants are embraced within the public interest at issue and dependent upon it for the exceptional character of the rights contested.
For a SEP litigation, it is even more lop-sided. This is because in SEP litigation, it is not general competition in the market. Rather, it is competition for the market. The more entrenched a standard is, the more power a SEP owner has. Those that give examples of multiple businesses operating in the same market (where SEPs are implicated) mistake the presence of multiple players as a healthy sign. However, they are not comparing the right market and make multiple logical fallacies: mistake cause for effect, confuse between apples and oranges, etc. Just because there are 10 handset brands does not mean it is a healthy market. If one looks closely, all 10 brands are based on the same standard and all 10 have to approach the SEP owner for a license. There is no competition here as the market is already locked in into a particular standard. Just think, what other cellular standard are any of your phonebook contacts using today? Hint: They all use some form of LTE (Long Term Evolution)/ 4G technology as it is popularly known. You don’t even buy a phone because it is LTE compliant: you buy it because of other reasons, such as price etc.
The way I see it, confidentiality clubs are used as a sword by the right holder. And the defendant can’t even claim use of the same information as shield. Let me give some background to put things in perspective. In any confidentiality club, confidential information is sought to be confined to a limited set of people – some from the plaintiff/ some from the defendant. And, ostensibly the purpose is so that the defendant can make an informed commercial choice. That is almost never the case.
This is because of asymmetric information skew in favour of the plaintiff in SEP matters. What the plaintiff seeks to put under the cloak of confidentiality are the rates offered to third parties and the fact that those third parties adopted those rates. What is never disclosed are the circumstances under which that decision was taken by those third parties. Sometimes, even technical information such as claim charts mapping the patent to the relevant standard are also disclosed under this arrangement. This is extremely problematic and Prashant has written about it here.
The circumstances under which the supposedly similarly placed parties signed those agreements are not disclosed. It may very well be the case that these were signed under duress – the threat of an injunction to stop goods at customs. It is not possible now but surely it was possible up to three years ago. Or for that matter imposed in court – whether as a stop gap arrangement or having no choice, the defendant taking the license, having suffered not one but two reversals in court.
There are other questions such as: what if the agreement was signed after the injunction was issued? What agreements were shown to these parties? How many patents were there in the license at the time the agreement was signed? How many were sold off to a third party post licensing them to a party?
Suffice it to say that answers to these questions won’t be volunteered by the patent owner. These are obtained when the plaintiff is put under scrutiny, and that too in an environment that is not to its liking. One can see previous blog posts on Ericsson’s litigation with Micromax, Intex, Lava, iBall and others to see these issues raised.
Putting claim charts as confidential information in the confidentiality club is another aspect that is simply overreach. Mind you, the entire standard such as LTE is public and even the configuration is public information. Each party who is desirous of having its IP included in the standard agrees to its disclosure and there are several disclosures made regarding this particular IP. Once a standard is adopted, then the patent is automatically covered by a standard compliant device. Hiding claim charts and putting them in a confidentiality club is overkill because this is a public interest issue and can’t be brushed under the carpet simply by putting it in a confidentiality club.
Other manufacturers should be able to see what is happening and if possible be able to take out revocations/ or at least become part of the litigation once the process is initiated by the right holder by suing one party.
Now coming to confidentiality of the defendant’s information that is supposedly protected in a confidentiality club. This information by nature is public. It is easy to count the unit sales of the defendant and there are many ways and multiple databases from where the volume of sales can be deduced. First is IMEI (International Mobile Equipment Identification) series for a mobile, which is a 16 digit number and each series has a million units. This information is with customs and all mobile operators and is easy to determine which series are being used by each device provider. Trade journals, etc. are a resource that is also available to those inclined towards more specificity in terms of absolute numbers and margins.
The point here is that the defendant really has no information that can be protected under a confidentiality club.
So far, this discussion has focused on a single aspect of information that is sought to be protected. Now let us view the issue from another perspective.
This is the space where the SEP owner refuses to license component manufacturers and approaches the brand owner. This is obviously to extract more money from brand owners as the royalty base is much higher. The argument that it is easier to approach a single brand owner rather than multiple component manufacturers is plain bunk: there is only one component that adds communication capabilities to any device. And to say that it is easier to approach brand owners, to me, is going for the bigger target that can be squeezed ever harder. Obviously, this information is never forthcoming in a cozy confidentiality club where the rules are set by the right holder.
This strategy of reaching out to the end brand results is a great skew and distorts the entire market. This approach to the brand owner results in numerous inefficiencies as costs are passed on to the end-consumer. My view is that these inefficiencies will increase substantially in the context of 5G SEP licensing, given the plethora of standard compliant connected devices in 5G technology.
Coming back to the harm caused by a confidentiality club in SEP litigation: It is public interest that suffers when one has to pay extortionary costs to a SEP owner. These costs are not just higher royalties, but payment for expired patents that are bundled into a patent license, irrelevant patents, and non-essential patents.