Open source, file systems and patents

Do you ever wonder how things work? Here at TechnoLlama we spend an obscene amount of time wondering precisely how applications and gadgets work. How does a media server allow me to watch downloaded TV shows in my PS3? How are long files negotiated between a Mac and a PC? Which file protocols are required to make Linux and Windows systems talk to one another? The answer to some of these essential yet often neglected questions is one of standards. If every device in the world operated its own file system, then things would not be able to communicate with one another. Mac, Linux and PC formats would each exist in their own little worlds and the technology world would be a little bit more difficult than what it is. Thankfully, file systems can talk to one another, allowing us to execute files across platforms, so the ext* system in Linux can access information in Windows FAT drive.

This seems straightforward enough, but the issue is that some file systems are protected by patents, specifically, Windows File Allocation Table (for example, this 1989 U.S. patent). Traditionally, interaction with devices has been permitted through cross-licensing agreements between the major file system providers, which has allowed some of these patents to go unchallenged. This has been for everyone's benefit, as a world where different file systems could not talk to one another would be problematic to say the least. However, the delicate balance has finally been broken as Microsoft has sued car-navigation maker TomTom for infringement of their file system patents.

Some background information is required. Gadgets and electronic devices that have some form of graphical user interface require an operating system to run on. Media players, digital cameras, mobile phones, GPS systems, PDAs, all of them have built-in and often bespoke operating systems. The Linux kernel is very popular with gadget manufacturers because it offers a stable, scalable, modifiable and cheap option for manufacturers. In order for these devices to talk to Windows, they have to implement some form of compatibility with Microsoft's FAT, which would require a licence. These are commonly handled in the industry through cross-licensing, Microsoft allows the manufacturer to use FAT if they in turn licence them with some of their own claims. Cross-licensing is widely practised, and allows the industry to operate without having to go to the courts at the drop of a hat. However, the Linux kernel is licensed through the GPL v2, which imposes an obligation on developers using modified code to allow the user to "copy, distribute or modify the Program subject to these terms and conditions." Furthermore, the GPL makes it clear that even if a developer is subject to patent infringement restrictions through litigation or an agreement, then they must stop distribution of the modified code. Section 7 reads:

"If, as a consequence of a court judgement or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program."

This is clear as day. If anyone develops a program using the Linux kernel code and embeds it to a gadget, then they must make that code available without restrictions. What has come to light is that Microsoft has entered into secret cross-licensing agreements with various manufacturers who use the Linux kernel that amount to breach of the condition set in section 7 of the GPL. TomTom refused to sign such an agreement, so they got sued.

This could very well be the opening shot in the much awaited software patent war between Microsoft and FOSS developers. If the allegations are true, then the Free Software Foundation may not have any other option but to start suing people for breach of the GPL. The question is, who will be sued first?

*opens popcorn and sits watching avidly the ensuing Armageddon*

iPhone infringes patent

Not to belabour the case against software patents, but as a newly converted follower of the Cult of Mac I found this one quite interesting. Apple has been sued by small web developer EMG Technology for infringement of its patent protecting an "apparatus and method of manipulating a region on a wireless device screen for viewing, zooming and scrolling internet content" (US Patent 7,441,196). No prizes will be awarded for guessing where the suit was filed.

The abstract reads:

"A method and apparatus of simplified navigation. A web page is provided having a link to a sister site. The sister site facilitates simplified navigation. Pages from the sister site are served responsive to actuation of the sister site link. In one embodiment, the sister site includes matrix pages to permit matrix navigation."

Let me see, so this patent covers any sort of navigation simplified for mobile devices, and was filed in March 2000. I guess the examiners missed that the Wireless Application Protocol (WAP) was first established in 1997. Or perhaps they also missed that in that same year Openwave had a functional HDML browser for AT&T handsets. Or they also missed that Microsoft has had its Microsoft Mobile Explorer since 1999, from technology developed also in 1997.

Let me get my patent troll detection kit:

• Exceptionally broad and obvious claim (check).
• Small company, (check)
• with suspiciously empty website (check).
  
• Patent application filed around 1999-2000 (check).
  
• Plenty of prior art (check)
  

Seriously, who examines these applications anyway? Don't they have Google?

NCSoft sued for avatar patent infringement

(Via Colin Miller) NCSoft, the producers of virtual worlds such as Lineage, City of Heroes and Guild Wars has been sued by virtual platform developer Worlds.com, which holds several patents on 3D environments. Back in December, Terra Nova had reported that Worlds.com had announced that it would be enforcing its patents, and as everyone in the comments section agreed, this was a completely spurious claim. As far as I can tell, Worlds.com is not involved in the MMO market, as it makes bespoke virtual enviroments for corporations and the enterainment industry, yet it has warned that it will enforce its MMO-related patents.

The patent in question is U.S. 7,181,690, which protects a system and method for enabling users to interact in a virtual space. The abstract reads:

"The present invention provides a highly scalable architecture for a three-dimensional graphical, multi-user, interactive virtual world system. In a preferred embodiment a plurality of users interact in the three-dimensional, computer-generated graphical space where each user executes a client process to view a virtual world from the perspective of that user. The virtual world shows avatars representing the other users who are neighbors of the user viewing the virtual word. In order that the view can be updated to reflect the motion of the remote user's avatars, motion, information is transmitted to a central server process which provides positions updates to client processes for neighbors of the user at that client process. The client process also uses an environment database to determine which background objects to render as well as to limit the movement of the user's avatar."

In this line of work I have become used to ludicrous patents, but this one should be awarded a price. Given that the filing date is August 2000, I am sure that any examiner should have come up with examples of graphical avatar interaction in a 3D environment. Meridian 59 and The Realm Online were released in 1996, Ultima Online in 1997, Lineage in 1998, and Everquest in 1999. All of them embody exactly the patent claim, so how could it have been awarded with such extensive prior art?

This is why so many people are opposed to software patents. All you need in order to make money out of litigation is to make a vague claim for which there is a mature market, get it issued because the examiners do not know anything about the subject, and then start suing market leaders in order to extort licence fees from those who are actually innovating and making popular products.

By the way, the drawing with the penguins is part of the patent claim. I kid you not.

Bilski: an end to stupid patents?

I finally got around to reading In Re Bilski (via Groklaw), the latest landmark case in the United States with regards to patentability issues. While abstract ideas are not patentable, ever since the case State Street, the U.S. has allowed mere abstractions as patentable subject matter if they produce a "useful, concrete and tangible result". This means that one did not need to have anything tangible, technical or even plausible, in order to get a patent, but that the result would be tangible; which confirms my firm belief that patent law is 90% sophistry and 10% mumbo-jumbo. It is no coincidence that software patents and business method patent applications exploded after State Street was decided. The impact of allowing abstract "inventions" can be seen when one compares the patent landscape in Europe and the United States, as mere abstractions or business methods are not patentable subject matter in Europe. The result of such practice is a more reasonable and rational patent policy, with better quality patents and less worry for innovators.

First some background. In 2002 Mr Bernard Bilski filed and application (08/833,892) to protect a "Capped Bill System". The abstract reads:

"A method of providing one of a good or a service to at least one entity at one of a payment, rate, or price that is capped at a pre-determined amount. The method includes producing an offer for the entity, wherein the offer represents at least one of a capped maximum payment, a capped maximum rate, a capped maximum usage, a capped maximum consumption, or a capped maximum price amount. The method also includes providing the good or service to the entity at one of a payment, rate, or price that may fluctuate, wherein the payment, rate, or price cannot exceed the capped maximum payment, capped maximum rate, capped maximum usage, capped maximum consumption, or capped maximum price amount."

At first reading, one realises that this does not describe anything specific, it is purely the abstract description of a capped payment method. Where is the novelty? Where is the tangible result? The examiner agreed and rejected the application. The Board of Patent Appeals agreed with the examiner. Mr Bilski then appealed to the United States Court of Appeals for the Federal Circuit (CAFC), which affirmed the rejection of the Bilski application. The question at the heart of the majority decision is whether the method described in the Bilski application is patentable subject matter. One of the ways in which business method patents were allowed in the past was by the application of what is known as the machine-or-transformation test, namely, that if the abstract process described could be carried out by a machine, or allowed some physical transformation, then it would be patentable. The CAFC majority disagreed that the Bilski application met such test. They say:

"Even if it is true that Applicant's claim "can only be practiced by a series of physical acts" as they argue, see id. at 9, its clear failure to satisfy the machine-or-transformation test is fatal. Thus, while we agree with Applicants that the only limit to patent-eligibility imposed by Congress is that the invention fall within one of the four categories enumerated in § 101, we must apply the Supreme Court's test to determine whether a claim to a process is drawn to a statutory "process" within the meaning of § 101."

To me the most interesting part of the ruling is the dissenting opinion by Judge Mayer, which goes even further, and makes the point that the question is one of allowing abstract thoughts to be patented. He says clearly and astutely:

"The patent system is intended to protect and promote advances in science and technology, not ideas about how to structure commercial transactions. Claim 1 of the application of Bernard L. Bilski and Rand A. Warsaw (“Bilski”) is not eligible for patent protection because it is directed to a method of conducting business. Affording patent protection to business methods lacks constitutional and statutory support, serves to hinder rather than promote innovation and usurps that which rightfully belongs in the public domain. State Street and AT&T should be overruled."

Judge Rader seems more favourable to business and software method patents, but even he agrees that the majority decision is unnecessarily complex. He says that:

"In sum, this court today invents several circuitous and unnecessary tests. It should have merely noted that Bilski attempts to patent an abstract idea. Nothing more was needed."

It would be easy to miss the importance of Bilski amongst the patent law technicalities, but make no mistake about it, it is a massive shift in U.S. patent policy, something that draws USPTO practice closer to Europe. It seems clear that the United States patent system is finally waking up after a patent binge, and is realising that it has a very bad hangover. Business method patents are an aberration for several reasons, chiefly amongst them is the fact that they disproportionately reward applicants who have incurred in little or no investment to come up with their "innovations". They reward trolls and punish legitimate businesses, making the mere business competitive environment a more hazardous place to conduct private endeavours. One has to hope that Bilski will be the beginning of the end to the most outlandish patents out there, and may inject some rationality back to patent law.

So far, the only people I have heard complaining are patent attorneys. That says it all really.

*Bleeping* patents

Habitual readers may suspect already that I think software patents are a *bleeping* bad idea. Yeah, they are often filed by *bleeping* patent trolls, or are awarded to *bleeping* obvious "inventions" that do not advance the state of the art in any *bleeping* way.

Enter U.S. Patent 7,437,290 which describes a method for automatic censorship of audio data for broadcast. The abstract reads:

"An input audio data stream comprising speech is processed by an automatic censoring filter in either a real-time mode, or a batch mode, producing censored speech that has been altered so that undesired words or phrases are either unintelligible or inaudible. The automatic censoring filter employs a lattice comprising either phonemes and/or words derived from phonemes for comparison against corresponding phonemes or words included in undesired speech data. If the probability that a phoneme or word in the input audio data stream matches a corresponding phoneme or word in the undesired speech data is greater than a probability threshold, the input audio data stream is altered so that the undesired word or a phrase comprising a plurality of such words is unintelligible or inaudible. The censored speech can either be stored or made available to an audience in real-time."

What the *bleep*? What sort of stupid *bleep* is that? How can Microsoft claim such *bleeping* idiocy?

Seriously though, I actually think this is a useful invention. Otherwise the above text would be rather colourful. Now I'm off to swear in the shower, the last refuge of the foul mouthed.

Symbian software patent appeal rejected

I have been following with interest the software patent case of Symbian v Comptroller General for Patents. The Court of Appeals has rejected an appeal by the UKIPO, which seems to move English software patent rulings closer to te EPO Board of Appeals interpretation.

As background, Symbian is a UK company which produces a popular mobile phone operating system, and in 2004 made a PCT application for "Mapping of dynamic link libraries in computer devices". The patent was awarded by the European Patent Office (EP1678608), but it was rejected by the UKIPO (GB0325145.1). Symbian appealed the decision and won. The UKIPO found the ruling by Patten J to be faulty, and filed an appeal, which as I mentioned, has been rejected.

The decision has been drafted by Lord Neuberger, and I have to say that it does not go into too much detail of the law, and therefore is a less compelling read than Aerotel and other rulings, but it still contains some interesting discussion. The core issue of the Symbian appeal is rather straightforward. Was the UKIPO justified in excluding Symbian's patent application? Patten J opined that it was not, and the Court of Appeals agrees. Lord Neuberger is of the mind that excluding inventions in computers is arbitrary and unfair:

"It can also be said in favour of Symbian's case that it would be somewhat arbitrary and unfair to discriminate against people who invent programs which improve the performance of computers against those who invent programs which improve the performance of other machines. However, as against that, what goes on inside a computer can be said to be closer to a mathematical method (which is, of course, not patentable by virtue of art 52(2)(a)) than what goes on inside other machines."

Nueberger L makes a strong case against exclusion of software by the mere fact that it is software, while acknowledging the inherent difficulties in looking for prior art when the source code is absent. Nueberger L spends some time trying to determine whether or not there is a technical contribution in the Symbian "invention", and concludes that there is, and therefore the patent application cannot be excluded on those grounds. The case then is not about the novelty of the application, it is about unwarranted exclusion of a valid application on the grounds of the patentability of software.

What is clear is that the Symbian appeal is a concerted attempt to bring the EPO and English decisions closer together. While Aerotel placed a large wedge between the EPO Board of Appeals and English courts, recent cases have been bridging the gap, particularly in the area of obviousness in non-software cases (namely, Conor v Angiotech and Actavis v Merck). Lord Neuberger spends his last paragraphs building bridges and making a friendly gesture to the EPO. He says:

"It is, of course, inevitable that there will be cases where the EPO will grant patents in this field when UKIPO should not, at least so long as the view in Pension Benefit and Hitachi is applied by the Board and is not applied here. The fact that the two offices and their supervisory courts have their own responsibilities means that discrepancies, even in approach or principle, are occasionally inevitable. However, the fact that such discrepancies have been characterised as "absurd" by Nicholls LJ, and the reasoning in [3] of Conor emphasise the strong desirability of the approaches and principles in the two offices marching together as far as possible. This means that there is a need for a two-way dialogue between national tribunals and the EPO, coupled with a degree of mutual compromise. More directly relevant to the present appeal, it means that, where there may be a difference of approach or of principle, one must try to minimise the consequent differences in terms of the outcome in particular patent cases."

This seems reasonable, but to my mind the Symbian appeal is a wasted opportunity to clarify concepts. The ruling completely fails to address the fact that the Symbian patent application is woefully obvious, and should not have been granted on the grounds of novelt and inventive step. True, this case has never really been about the application itself, it has been about the narrowness of interpretation of what constitutes software as such, technical effects, and therefore whether software should be excluded from patentability on those grounds. However, most of the definitions of technical effect in the case law delimit it as a novel step that advances the state of the art in a comprehensive manner. The case might have analysed that the Symbian application is seriously lacking in those grounds, and therefore did not meet the technical contribution requirements. However, Neuberger L and Patten J before him do not go into this vital question in detail, and I believe they should.

It seem like the UKIPO will have to redraft their guidelines again.

Update: The Times Online has an article on the ruling, which I found quite misleading. They are making it seem as if software patents could not be obtained in the UK prior to the ruling, which is not true. As I mentioned, Symbian has brought EPO and UK practices closer together, which does indeed mean that more software will be subject to patent protection.

Aerotel patent repealed

While I was away presenting on software patents in Brazil an important decision took place last week. Aerotel v Telco is one of the most important cases in this area, and the patent that originated it has now been repealed because of obviousness. Aerotel is an Israeli patent troll that provides no services and does nothing but engage in patent infringement suits all over the world defending their patent. The patent protects a method of making pre-paid telephone calls.

Fysh QC delivered the reasoning, and it makes for interesting reading (for those inclined to be interested in software patent litigation that is). Fysh QC states that:

"Even if it is proper to pose the first question in this context, the answer must I think, be this: the general idea of either post or pre-payment charging for telephone calls would unhesitatingly have been present in the mind of a skilled worker at the priority date. These were the alternatives. A telephone company must of course be paid and both methods of payment were not just well known but were the norm. [...] Thus, assuming it to be a relevant consideration, the idea of requiring payment in advance for telephone calls is in my judgement, completely obvious."

Hurray! One of the points that I make when talking about software patents is that there may be scope for some protection of computer implemented inventions, but that the implementation has been atrocious. In my opinion, most software patents fail in obviousness, prior art and disclosure. The problem is that patent litigation is expensive, and defending yourself even against a bad patent may cost a fortune.

Microsoft wins OOXML standard

The geeks are up in arms because Microsoft has won approval of its much maligned Office Open XML (OOXML) format as an open standard. The International Organization for Standardization (ISO) has finally allowed OOXML to become an international standard after its initial rejection last year, but the decision still can be appealed.

Why is this controversial, and why should we care? You may already be familiar with OOXML, if you own Office 2007 (or 2008 for Mac), then you are using it; you know, the annoying .docx file format that is not fully backward compatible. However, the format is much better than previous XML schema used in the likes of Office 2003, and despite the many problems with backward compatibility, I must admit that it does produce nice PowerPoints. One problem highlighted with the OOXML is that it is protected by various patents. While holding a patent over a standard technology is not such a problem, Microsoft's history of abusing a dominant position has made some people nervous. Under current standardisation practice, patent holders must offer their patented technology subject to standard approval on a Reasonable and Non Discriminatory basis (RAND). Arguably, Microsoft has fulfilled this requirement by issuing the
Microsoft Open Specification Promise, a unilateral promise of non-enforcement of their claims on OOXML. Nevertheless, many people have been suspicious of the promise.

The other problem with the OOXML format is that it has been pitted against the Open Document Format (ODF), the one favoured by the open source community and developed for the Open Office project. The ODF has already received ISO standard approval, which has raised questions about the need for a competing standard. There have also been serious issues about OOXML's approval procedure, with accusations of bribery and worse. The accusations have been so strident that there is talk of further action by the European Commission.

I must admit that at the moment I have not made up my mind. While I am disturbed by the many reports of browbeating and corruption, I am somehow troubled by the strident opposition from the open source community against the format. ODF has already been approved, and I believe that the battle should take place with the consumer. Open Document proponents should fight to make Open Office better, so that it will get wider adoption. In my experience there is a good window of opportunity at the moment, as Office 2007 has failed to capture the market, much like Vista has failed to dominate. Open Office and the Open Document standard then should take the fight directly to Microsoft, not at the ISO, but at the PC.

Blackboard's patent on the way out

(via Ray Corrigan) I have been following the saga of U.S. Patent 6,988,138 protecting "Internet-based education support system and methods". After winning an injunction in a Texas court, Blackboard has seen 44 of its claims knocked out in a preliminary injunction by the USPTO. Ray has a comprehensive report, so I will not repeat it.

Blackboard's claim exemplifies everything that is wrong with software patents. Theirs is a patent that should not exist, and if there is any justice in the world, it will be repealed and <Hugo Weaving> cast back into the fiery chasm from whence it came </Hugo Weaving>.

Computer programs are not software, UK judge says

I mentioned yesterday the new case in the High Court of England and Wales regarding software patents, Symbian Ltd v Comptroller General Of Patents. Symbian Ltd is a UK company which produces a mobile operating system, and in 2004 made a PCT application for "Mapping of dynamic link libraries in computer devices". The title itself should give a hint as to its potential validity, as DLLs are at the heart of a computer's functionality. The patent has been awarded by the European Patent Office (EP1678608), but it was rejected by the UKIPO (GB0325145.1). The patent is currently in the application stages at the USPTO. The abstract reads:

"A dynamic link library (DLL) in a computing device is provided in the form of a first part and an extension part. The first part has selected entry point ordinals by which an application program may link to first functions. The application program may only link to further functions via the extension part of the DLL."

I have read the application, and I must be missing something, but this does not seem to meet patentability requirements. Symbian seems to be saying that they have found a new way to link DLLs so that it will improve performance and interoperability. How does this amount to an inventive step? Moreover, the practice both in the UK and at the UKIPO has been to award patents only to software that offers an advance in the state of the art, which this invention fails (at least in my uneducated and opinionated view).

Symbian appealed the application rejection by the UK Intellectual Property Office, and the ruling makes for some interesting reading. Firstly, we get a re-run of the case law in this area. Then, we get yet another convoluted reasoning to assume that Art 52 of the EPC does not say what it clearly says, namely, that computer programs as such are not inventions, and therefore not patentable. The ruling then goes on to apply Aerotel's test. Software patent geeks will know this by heart. The test says:

1. Construe the claim
2. Identify the actual contribution
3. Ask whether it falls solely within the excluded subject matter
4. Check whether the actual or alleged contribution is actually technical in nature

Patten J goes through the first three steps of the case, but forgets completely to answer the fourth! Patten J first construes the claim:

"This is not a problem in the present case. As explained earlier, the substance of the claim is the re-organisation of the DLL into two parts and the provision of a library interface for the extension DLL so as to improve the linking of any EXE program running on the computer with the available functions contained in the DLL files. No issues of construction arise."

Then he identifies the contribution:

"Symbian's case before the Hearing Officer was that the contribution made by the invention lay in the improved reliability of a computing device enabled by the provision of a novel interface. This enabled the EXE program to access available functionality regardless of additions or amendments made by the updates to the DLL and its ordinal numbers. "

Then he wonders if the contribution is excluded software matter (in other words, whether or not it is "software as such"):

"So is this invention no more than the running of the program? Having regard to the earlier authorities the answer has to be that it depends on what the program does and not merely how it does it. The mere fact that it involves the use of a computer program does not exclude it [...] The key elements in her reasoning appear to be that the use of the new interface to obtain better linking between the EXE program and the updated DLL does not involve a change in the role of the DLL but only in the way in which it is accessed in the new piece of software. This is nothing more than a computer program and is therefore, she decided, excluded from patentability. "

This is going well. He recognises that the application was rejected because it was software as such. He then goes on to agree that step 4 may not always be needed if the test falls at steps 2 or 3. In this particular case, the UKIPO has considered it to have failed at 3, as it is clear that DLLs are "software as such", and therefore should not be patentable software matter. I happen to agree with the UKIPO's interpretation. Unfortunately, Patten J has clearly hit his knowledge wall when it comes to software, and then rambles on about whether or not DLLs and operating systems are separate entities and should be considered as software (heavy theoretical Information Systems stuff coming from an English judge, ugh!). Patten J then delivers the single-most uninformed sentence with regards to software that I have ever read. He says:

"It is simply inaccurate to label all programs within the computer as software and on that basis to regard them as of equal importance in relation to its functionality."

What? When? How? Software is simply "a set of statements or instructions to be used directly or indirectly in a computer to bring about a certain result". Computer programs are... exactly the same thing!

Unsurprisingly, the conclusion is disappointing. Patten J disagrees with the narrowness of the UKIPO analysis, bypasses step 4, and concludes as follows:

"I think that the Hearing Officer took too narrow a view of the technical effect of the invention and was wrong to exclude it from patentability on the basis that it amounted to no more than a computer program. The appeal will therefore be allowed."

This is strange, because if he considered that Symbian's application was indeed patentable subject matter, then he should have gone to test if it had a technical contribution. Technical contribution has been defined by the EPO Board of Appeals as some technical advance on the prior art in the form of a new result. I cannot possibly see how making DLLs run faster amounts to such.

Now I understand why the ruling is being appealed by the UKIPO.

Symbian ruling will be appealed

The latest software patent UK High Court case, Symbian Ltd v Comptroller General Of Patents [2008] EWHC 518 (Pat), has continued following the line set by Astron Clinica. The case exemplifies the split between UK-IPO and EPO practice on software patents, as the EPO awarded the patent, while the UK-IPO did not. At the heart of the question is the issue of whether or not software as such can be subject to patentability.

Symbian's claim protects DLL indexing in a computer. I may be entirely wrong here, but where is the innovation in that? Anyway, the High Court ruled in Symbian in favour of the "inventor". The good news is that the UK-IPO has declared that it will appeal the ruling, as it is clear that the judge did not follow the Aerotel/Macrossan test. However, while I welcome the appeal, I feel like this may be too little too late, as the EPO keeps awarding preposterously bad software patents. While English courts have sometimes been standing in the way of EPO practice, the problem will not be solved until the EPO Board of Appeals starts interpreting art. 52 of the European Patent Convention differently.

The EPO is under a lot of commercial pressure from American software companies to harmonise practice across the Atlantic.

Blackboard wins e-learning patent suit

Since 2006 I have been following with interest the patent infringement case of Blackboard v Desire2Learn. Blackboard is a provider of educational software and virtual learning environments, which owns U.S. Patent 6,988,138 protecting "Internet-based education support system and methods". In 2006 they sued VLE provider Desire2Learn for infringement of aforementioned monopoly right. No points for guessing in which jurisdiction was the complaint filed, let's just say that it is a large state that borders Mexico, and it is not California. The case made it to trial, and the jury awarded Blackboard $2.5 million USD (about 44 million WoW gold) for lost profits and $630,000 USD in royalties.

At this point I should probably offer a rant about the evils of software patents, but I think that the facts speak for themselves. A company is using its patent in order to eliminate competitors from the market. Surely, this is not what the patent system is all about, is it?

UK IP Office issues statement on software patents

(via Arne Kolb) After the landmark ruling in Astron Clinica Ltd & Other, the UK IP Office has issued an amended notice with regards to patentable subject matter. The practice after Aerotel/Macrossan at the UKIPO was that "claims to computer programs or to programs on a carrier were not allowable." In other words, if you created some code and this was considered an invention, then you could not claim patent protection if the invention was implemented in software. You could claim protection for an invention, but not for the software as such. But now after the ruling, the practice has changed. The UKIPO says:

"In his judgment in Astron Clinica [...], Kitchin J has now clarified the law in this area. He has decided that where, as a result of applying the test formulated in Aerotel/Macrossan, claims to a method performed by running a suitably programmed computer or to a computer programmed to carry out the method are allowable then, in principle, a claim to the program itself should also be allowable. However, Kitchin J made it clear that the claim to the computer program must be drawn to reflect the features of the invention which would ensure the patentability of the method which the program is intended to carry out when it is run. Where, but only where, these conditions are met, examiners will no longer object to claims to a computer program or a program on a carrier."

This seems to me to be a grudging acceptance of the ruling. If the invention is implemented via software, then it can be subject to patent protection, but only if it meets with the requirements set out in Aerotel. These requirement is that weird concept of "technical contribution", which is generally considered to say that the software should make a considerable advance in the state of the art.

I have never had a problem with such an interpretation. If something is novel and inventive, then the medium of delivery is irrelevant to the patentability of said device. However, I still have a problem with the law. Apologies for repeating this rant again, but why not change the European Patent Convention and get this silliness over and done with?

Landmark software patent ruling in the UK

Back in November2007 I briefly commented on a case making its way to the High Court of England and Wales dealing with software patents. We now have a decision, and it may prove to be quite an important shift in the practice at the Intellectual Property Office.

The case is Astron Clinica Ltd & Others v The Comptroller General of Patents, Designs and Trade Marks [2008] EWHC 85 (Pat) . The ruling is particularly important as it was believed by many (myself included) that Aerotel was the final word with regards to software patentability. Those who follow the software patent debate, may recall that the IPO issued a set of guidelines arising from Aerotel which pretty much discourage outright software patent applications in the UK. This was not in line with practice at the EPO, or so it was believed by various parties which favour the patentability of software as such. This opinion was shared by a number of software developers, which then issued an appeal against the rejection of six separate patent applications made to the IPO. These applications had three things in common. They all represented claims for methods, devices and software. Examiners found that the claims on methods and devices were patentable, but that the software "as such" was not, based on the aforementioned guidelines and on the Aerotel ruling. What were the claims in those applications? I will quote from the ruling:

"Software 2000 has developed a method of generating bit masks for use with laser printers which results in higher quality images. It is implemented by programming a conventional computer, printer or copier to process images in a particular way. Software 2000 exploits its invention by selling the program to its commercial partners who then incorporate it in their printers and printer drivers, and distribute it to end users in the form of printers, computer discs and web downloads. The end users are located worldwide.

Astron Clinica was founded to commercialise skin imaging techniques developed at the University of Birmingham which enable images of the skin to be processed to identify the distribution and concentration of underlying skin chromophores. The invention described in its application provides a system and process for generating realistic images representing the results of planned cosmetic or surgical interventions which change the actual or apparent distribution of these chromophores. The invention is implemented by programming a computer to process images in a particular way. It is commercialised here and abroad by selling a disc which causes a computer to be configured so as to undertake the required processing.

Inrotis is a spin-off company established by the University of Newcastle upon Tyne to commercialise drug discovery and network analysis techniques. Broadly speaking, the inventions the subject of its two applications in issue concern methods of identifying groups of target proteins for drug therapy by processing proteome data defining proteins and protein interactions. The commercial product which Inrotis sells is a computer disc which causes a computer to be configured so as to carry out the necessary processing.

SurfKitchen is a mobile services company and has made an invention which improves the ability of mobile telephones to access services on the Internet. It is implemented by pre-storing a program on a mobile telephone memory or by downloading the program from the Internet. In either case the program is usually made available by one of SurfKitchen's commercial partners to whom it makes the program available on a computer disc.

Cyan Technology is a semi-conductor company which designs and builds microcontrollers. It has invented a method of generating data for configuring microcontrollers which greatly simplifies chip design and programming. The commercial products that implement the invention are computer discs and Internet downloads which cause a computer to be configured so as to undertake the required processing. Cyan Technology distributes these computer discs and Internet downloads worldwide."

As you can see, the claims are all very similar in the fact that there is a clear executable element to the invention. All of the applications relate to some form of software which runs on a device, be it printer, computer or mobile phone, which somehow improves the actual functionality of the recipient. The question facing the court was therefore whether the executable element is subject to patent protection. The answer is negative after Aerotel, but it also should be negative if one reads strictly Art 52(3) of the European Patent Convention, which clearly says that computer programs as such are not an invention. The problem has been that the practice at the EPO after various cases has been to allow patentability of executable computer programs, this has been repeatedly seen in various EPO Boards of Appeals cases (see T 1173/97, and most recently, T 0411/03).

The ruling has an extended section detailing existing legislation and case law, which I will ignore, as it is a common thread along these rulings. The meat of the matter however, is whether Kitchin J will agree with Aerotel and Jacob J in continuing the separate practice at IPO and EPO. I agree completely that the UK has been developing a completely different doctrine to that at the EPO with regards to software patentability, but I disagree that this is a bad thing. I have to say that I strongly believe that the various cases dealing with software patents in recent years had it right when it comes to analysing whether software as such is subject to patent protection. Jacob J clearly was not amused with the mental gymnastics exhibited by the EPO Board of Appeals to try to conveniently ignore the fact that the wording of Art 52(3) does not say what it clearly says. To be able to do this, the EPO invented the contrived and unworkable concept of "technical contribution", which does not exist in the legislation.

Unfortunately, Kitchin J decided that the split with the EPO should end. He claims to have read Aerotel closely, and he rightly points out that nothing in that decision prohibits the patentability of computer programs as such. He takes this as the permission needed to get back into EPO practice. He comments:

"... I do not detect anything in the reasoning of the Court of Appeal which suggests that all computer programs are necessarily excluded. I have identified the key aspects of the decision which relate to computer related inventions and they undoubtedly criticise the reasoning of the EPO Board of Appeal in each of the "trio" of cases. But the criticism is directed at the "any hardware will do" approach and the return to form over substance with the drawing of a distinction between a program as a set of instructions and a program on a carrier."

This is all the permission he needs, computer programs are back in the menu. He reasons:

"... it is highly undesirable that provisions of the EPC are construed differently in the EPO from the way they are construed in the national courts of a Contracting state. Moreover, decisions of the Board of Appeal are of great persuasive authority. In the light of Aetotel/Macrossan it is not open to this court to follow the decisions in the "trio". However the new approach can be interpreted to produce a result consistent with that obtained by applying the reasoning of the Boards of Appeal in IBM/Computer Program Product T1173/97 and IBM/Computer Program Product II T0935/97 - decisions which, I would add, are still followed in the EPO as shown, for example, by the decision of the Board of Appeal in Tao Group Limited (2007) T121/06."

While I can see the reasoning behind it, I have to say that this is an unfortunate decision. In my opinion, the wording of Art 52(3) of the EPC is more than clear, and ensuing cases have tried very hard to ignore the fact that it is very clear and unequivocal in order to allow for the patentability of computer programs. It is not up to the EPO Board of Appeals and the courts to change the wording of the treaty. Why not get it over and done with and actually push for amendment to the actual letter of the law?

You can see some thought on the case and an interesting series of comments at IPKat.

High Court to review software patents

(via press release) Four UK technology companies have appealed a ruling from the Intellectual Property Office denying them patentability over a computer implemented invention. The self-styled High Tech Four ( Astron Clinica, Software 2000, Surf Kitchen and Cyan Holdings) had an adverse ruling from the IPO where it applied its guidelines on patentable subject matter issued last year in the wake of Aerotel's decision.

I have not had time to read through their claims, but it seems that they object to the restriction of patentability of software as such, and they want to reverse current practice at the IPO which is unfavourable to the patenting of computer implemented inventions, and they claim is contrary to current practice at the EPO and other continental patent offices.

It seems like we will get another software patent ruling soon, which has to be welcome. However, being the shallow person that I am, I cannot help but comment on the name the High Tech Four. What are they? Patentabilty super heroes? Their motto: "We Fight for Justice, Freedom, and Fair Rulings in IPO Subject Matter Decisions!"

One-click patent claims rejected

(via Lilian Edwards and Boing Boing) Several claims belonging to the infamous Amazon One-Click patent (U.S. 5,960,411) have been rejected by the USPTO. For those who have never heard of it, this patent is for a "method and system for placing a purchase order via a communications network" a broad claim on the inventive use of cookies to make consumer transactions faster and easier.

The third party re-examination procedure was brought by one Matthew Graham in 2006, and has just been decided. The claims rejected are 1-5 and 11-26, which completely eviscerates the patent. The examiner has decided that many of the claims are already covered by previous patents, and therefore have prior-art.

Just in case we are feeling smug about the madness sweeping through the other side of the Atlantic in an effort to patent anything under the sun made by man, I would suggest that people have a look at European Patents EP1678658 and EP1825357 held by Amazon. Computer implemented inventions are alive and well in Europa.

Linux patent infringement suit

(via Groklaw) Red Hat and Novell, owners of the two largest Linux enterprise distributions, have been sued in Texas (where else?) by patent troll IP Innovation, part of the troll tribe Acacia. The subject of the litigation is U.S. Patent 5,072,412, which protects a user interface with multiple workspaces for sharing display system objects. The abstract reads:

"Workspaces provided by an object-based user interface appear to share windows and other display objects. Each workspace's data structure includes, for each window in that workspace, a linking data structure called a placement which links to the display system object which provides that window, which may be a display system object in a preexisting window system. The placement also contains display characteristics of the window when displayed in that workspace, such as position and size. Therefore, a display system object can be linked to several workspaces by a placement in each of the workspaces' data structures, and the window it provides to each of those workspaces can have unique display characteristics, yet appear to the user to be the same window or versions of the same window. As a result, the workspaces appear to be sharing a window."

I went through the complaint to see if it could elucidate what this steaming pile of gibberish means, but there was little clue. It took me a while going through the entire claim in order to get an idea of what this entails, as I could not believe that such a broad patent could exist. I thought initially that they were talking about previews, or even application linking. However, it seems that this is all about user interfaces, and it could easily apply to any sort of graphical display using nested windows linked through an operating system. It is important to note that IP Innovation sued Apple using the exactly same patent (a case that was settled out of court).

Pam at Groklaw has put forward the theory that Microsoft is behind this case, but I disagree, as Novell still seems to be in their good books. However, the fact that this preposterously broad patent has been used against Apple and now Linux would seem to indicate some involvement from a certain large software corporation based in Richmond, WA. Nevertheless, what seems evident is that patent trolls may start turning their heads towards open source, and large Linux providers are the obvious targets.

I was also wondering about the terribly adequate choice of name for this particular troll. IP Innovation, a company that does not make anything, it only tries to find innovative ways to detach true innovators from their cash. I say that all open source developers should leave the United States and move to Europe, it is a little bit saner over here. Our patent trolls are very small.

Software patent debate? No thank you, says EPO

There are many of us interested in the software patent debate, but the EPO seems not to be one of those. The EPO held a one-day conference attended by parliamentarians, industry, NGO and experts to discuss what has happened since the dismissal of the Computer Implemented Inventions Directive two years ago (2 years? How time flies!) The consensus in the conference was that the debate will not be re-opened, and that there is no need for more legislation, or to resurrect the failed Directive. The efforts should go towards better patent quality and patentability, and perhaps even to sort out the messy case law.

Indeed, case law has been rather messy in this area. After Jacob's controversial decision in Aerotel, there has been a case by the EPO Board of Appeals (T 0154/04) which refuses to answer the questions posed in the Aerotel ruling (more discussion about this in IPKat and IP::JUR). At the moment there are clearly two ways of dealing with software patents in European cases: firstly is the "technical effect scepticism" present in recent UK-based cases and exemplified by Aerotel. Then there is "technical effect purism" present in EPO Board of Appeals decisions, where they have directly criticised Jacob's interpretation of European Patent Convention (which is rich coming from them, as technical effect has no basis in the EPC whatsoever).

We are left then with a rather clouded picture. It seems like we are in for a continued court battle, the balls are in the air and it seems like if you want to litigate in this area, the decision will depend on the forum.

Nevertheless, I think that some industry players are beginning to understand that software patents were not such a great idea to begin with. IBM continues to be at the forefront of this practice. Yes, they are the ones who apply for and obtain more software patents, but they are also continuously making efforts to project an image of usefulness. IBM has recently announced that it will “perpetual and universal access” to more that 150 software standard patents. This move is incredibly important because standards are what allow interoperability between devices and software. If a company owns a patent over a standard, it will be able to stifle innovation in that area, or to charge licence fees for others wanting to interact with that technology. by opening these basic patents, IBM will allow further interoperability and innovation in those areas in which it holds patents, a wise commercial move in my opinion.

Peer to Patent live

The amazing Peer to Patent system is now live. The system is the creation of New York Law School Professor Beth Noveck, and it is an extremely interesting system of involving the online community on the review of patent claims in pending patent applications.

I believe this is an interesting exercise for many reasons, and it will hopefully help in helping the USPTO in getting its act together and cleaning up its woeful patent quality record.

Linux infringes Microsoft's patents

Fortune Magazine has published an interview with Microsoft's Steve Ballmer, where he has made the comment that Linux infringes 235 of its software patents, and it will be looking for licences from developers and corporate users. As your friendly neighbourhood Prophet of Doom, I have been announcing the advent of the Great Software Patent War for some time now. Yes, I know, being right all the time is tiring and risky business, Cassandra had a cruel fate after all.

The argument put forward by Microsoft seems straightforward. "We own some patents, Linux implements some of those inventions in their code, if you want to use Linux, then you must pay us". Nobody has been sued yet, but it is obvious that the interview has been designed in order to issue a blunt threat against Linux users and developers to enter into negotiations with Microsoft.

This threat is unlike the much debated and scorned SCO v IBM case, where SCO has been conducting a lengthy (and ultimately futile) suit arguing copyright infringement by IBM and Red Hat. Unfortunately for Free and Open Source developers and users, Microsoft's claims have more weight given the strength of software patent claims in the United States. 235 patents are not something to be shrugged-off, this could truly spell the demise of many small-scale FOSS projects.

I've been trying to figure out the strategy behind this, after all, there was a feeling in some sectors that Microsoft had been warming towards FOSS. There was the adoption of an open source strategy, and the heavily talked-about deal between Novell and Microsoft. Under that deal, Microsoft and Novell promised not to enforce each other's patents, while Novell promised to pay Microsoft a percentage of its revenue. This deal, of course, is not kosher in Free Software circles. Moglen made it clear that the GPL v3 would be re-drafted in order to make such a deal a breach of the GPL, which was eventually done with the latest draft. According to the new version:

"You may not convey a covered work if you are a party to an arrangement with a third party that is in the business of distributing software, under which you make payment to the third party based on the extent of your activity of conveying the work, and under which the third party grants, to any of the parties who would receive the covered work from you, a patent license (a) in connection with copies of the covered work conveyed by you, and/or copies made from those, or (b) primarily for and in connection with specific products or compilations that contain the covered work, which license does not cover, prohibits the exercise of, or is conditioned on the non-exercise of any of the rights that are specifically granted to recipients of the covered work under this License [...]"

This convoluted clause is designed specifically to attack the Novell and Microsoft deal, hence its nightmarish wording. Similarly, the new GPL contained a clause that could eventually be used to make the patent licence contained in the new GPL into a viral clause similar to existing copyleft clause in GPL v2. Obviously, these developments seemed to prompt action from Redmond, and today's announcement seems precisely to do that.

But why now? It seems obvious that the threat is designed to issue a clear threat against the GPL v3, and perhaps it attempts to influence the draft's discussion. This may be the reason why the FSS, Stallman and Moglen have answered forcefully, almost with a "bring it on" attitude. They know that the pendulum is swinging against unfettered software patents, and that this case could very well prove to be the silver bullet that fatally wounds the current system. Imagine a situation where large numbers of corporate Linux users are sued by Microsoft. The result could very well be a legislative push against broad patentability. Microsoft is also playing with fire by entering into IBM's turf. As one of open source's corporate patrons, IBM has an impressive software patent arsenal that it could deploy if things get to an all out litigation battle. It is an open secret in the industry that as things stand, everyone is infringing someone else's patents, and what sustains the balance at the moment is a complex network of cross-licensing between industry giants. I have always believed this is one of the reasons why FOSS projects have managed to remain litigation-free so far. If Microsoft sues, will IBM retaliate? What would such a case look like?

Expect geekdom and the blogosphere to go nuclear over this one.