The team is joined by Guest Kats Rosie Burbidge, Stephen Jones, Mathilde Parvis, and Eibhlin Vardy, and by InternKats Verónica Rodríguez Arguijo, Hayleigh Bosher, Tian Lu and Cecilia Sbrolli.

Wednesday, 10 January 2007

Super Mario has kart trouble

While browsing through the list of recent decisions at the UK Patent Office, the IPKat noticed this one, relating to that famous Italian-American plumber, Mario (aka Super Mario). In a particular series of his many Nintendo games, Mario and his friends take to driving karts round racing tracks, one of these being Super Mario Kart, "widely regarded as one of the finest videogames of the 16-bit era", according to Wikipedia. A problem with this is that the karts often collide and spin off the track (part of the fun, of course), and it can take a while to get back on, spoiling the gameplay. This is where Nintendo's alleged invention came in. Their invention apparently solved this problem with a new way of putting the kart back in place on the track and facing in the right direction, without any discontinuities or tiresome blank screens.

The decision related to refusal of Nintendo's application under section 1(2) of the UK Patents Act 1977. The examiner had maintained an objection regarding s1(2)(c), i.e. that the contribution lay solely in the field of computer programs, and consequently refused the application. Referring to Aerotel/Macrossan (previously reported in the IPKat here), the Hearing Officer applied the now approved four step test, noting that the decision must be treated as a definitive statement of how the law on patentable subject matter is now to be applied in the UK, and that it should not be necessary to refer back to previous UK or EPO case law regarding the issue.

Not having opposable thumbs, the IPKat has great difficulty getting to grips with those awkward controllers (from the marvellous

After construing the claims, the Hearing Officer considered the contribution to be the process of setting a kart upright and facing in the same direction as it was prior to crashing. This was seen to be clearly wholly within the area of a computer program, but not itself a scheme, method or rule for playing a game. Since the third step question was answered in the affirmative, it was not seen to be necessary to consider the fourth step, i.e. whether the contribution was "technical" in nature. The application was therefore refused.

The IPKat sees much sense in this decision, but wonders how it can be squared with the previous decisions of Sun Microsystems and ARM, both of which also related to ways of getting computer programs to do clever things purely implemented in software, but which were related to much more serious-sounding issues of bytecodes and compilers rather than silly racing karts. Merpel thinks that the Patent Office just don't like patenting things to do with having fun.


Gerontius said...

I am having difficulty understanding how the IPKat sees any sense in this decision. While I think we can all probably agree on the outcome, the reasoning fails to make sense.

Look at the PO argument:
1. The contribution is a particular sequence of steps
2. This sequence of steps is performed by a computer therefore the contribution is a computer program.

What?! Having decided what the contribution was, the PO then completely ignored that contribution in saying it was a computer program. Surely the fact that a contribution is implemented by a computer is irrelevant to the question of whether that contribution is itself patentable? This was made clear in Vicom, and affirmed by the UK courts so many times I have lost count, when they said: "Generally speaking, an invention which would be patentable in accordance with conventional patentability criteria should not be excluded from protection by the mere fact that, for its implementation, modern technical means in the form of a computer are used".

Therefore, once the PO has decided that the contribution is a way of inverting a kart and putting it back on the track facing the same direction, the question to ask is whether this, in itself, is patentable, irrespective of the fact that it is carried out by a computer. I am not sure that any sensible person would say that it is patentable, but it is important for the reliability of the patent granting process that the REASONS why are clear.

Personally, I think this application would best be refused with reference to Scaelectric sets. If the car falls off the track, the player has to use their hands to put the car back on the track, in the same position (based on the player's memory of where it was), the right way up and facing the same direction as it was before coming off the track. The only contribution provided by the claimed invention in this case is therefore the computer program which does this automatically. That is one good reason why the invention is a computer program as such, not this rubbish spouted by the PO.

This is in keeping with the Fujitsu reasoning. Notice that, in Fujitsu, the claimed invention was the computer implementation of a known method. This contrasts with Macrossan where it related to a computer implementation of a (presumable unknown) business method. Macrossan was therefore refused first as being a business method and second as being a computer program program. The two cases cover two very different examples of ways in which applications may relate to excluded subject matter, both of which are necessary to be able to examine applications that may relate to one or more different categories of excluded subject matter. The PO's statement that no reference to previous case law is therefore shocking in its short-sightedness.

Another way to reject this application is the EPO way. What technical problem is solved by turning a kart the right way up? None. It arguably makes the game more enjoyable to play but, unlike in the EPO Konami decision, there is no improvement in the capability of the player to play the game.

Anonymous said...

The reasoning inevitably follows the Aerotel judgment because the Patent Office does not want to get into deep and irrelevant discussions about case law; it merely wants to make decisions based on the decided law. After all, no Patent Office decision is binding on any other, and at best will give practitioners a rough guide as to what to expect from the Office. The most important part of the decision is therefore paragraph 15:

"In reaching its judgment, the Court fully considered all the precedent UK case law in this area. Following the principles discussed in Colchester Estates (Cardiff) v Carlton Industries [1986] 1 Ch 80, [1984] 2 All ER 601 and [1984] 3 WLR 693, Aerotel/Macrossan must be treated as a definitive statement of how the law on patentable subject matter is now to be applied in the UK. It should not therefore be necessary to refer back to previous UK or EPO case law regarding this issue".

Given this, the reasoning appears pretty sound. There is therefore much sense in the decision, but I can see why some will still want to argue back to Vicom and others. Points to note include: i) Vicom is not binding on anyone; ii) the Court of Appeal is binding on itself and lower courts; iii) even the EPO do not follow Vicom reasoning any more.

Whether the UKPO refused the application on strictly following the Aerotel steps or on some hodge-podge of previous decided law (which they, of course, were bound not to do), the same result would be arrived at. The invention was not patentable because the only contribution was that it was carried out on a computer. This easily distinguishes it from the previous decisions of Sun and ARM, where there were at least some (albeit tentative) arguments about increases in processing speed and efficiency. In Nintendo there were no such arguments to be had.

zoobab said...

Dear Gerontius,

Any method implemented in a computer program is not patentable. Otherwise software programmers should read patent claims before writing computer programs, which is not practical, and not feasible according to the US experience of american software programmers.

If a computer program does not implement methods, what does it do then?

David said...

Not strictly correct, I'm afraid, Zoozab. Only computer programs as such are not to be regarded as inventions (section 1(2)). This is not at all the same thing as saying that any method implemented in a computer program is not patentable. Many thousands of patents have been granted for novel and inventive methods that are implemented in computer programs. The Sun and ARM cases are just two of them. What the Patent Office want before they will grant an application to a method is (among other things) some kind of contribution that is novel, inventive, and not simply an implementation of a known method on a computer.

zoobab said...

It is well known that the "as such" trick renders the exclusion of computer programs *meaningless*.

David said...

Don't confuse your own point of view with factual reality. That way madness lies.

The "as such" bit is not a trick; it is in the actual statute. How would you interpret it otherwise?

Anonymous said...


Have you ever seen an explaination of "as such", I've never seen a book, surgical method, or other idea patented, but lots of computer programs, yet they are all unpatentable "as such".

I, and I believe any reasonable person, would interpret it as passing the duck test - if it looks, acts and sounds like software, it's software and unpatentable.

"As such" most definately does not means "not simply the implementation of a known method on a computer", even if pro-software patent people like yourself wish it to be.

I think it would be a whole lot clearer if the "as such" was removed, it has muddied the waters and created all sorts of confusion where it wasn't necessary - like the CII debacle.

I think everybody would benefit immensely if it was just made clear that surgical methods, etc cannot be patented at all, full stop - no loop holes or weasel words.


Aaron Trevena, BSc (Hons)
Software Engineer

David said...

Interesting point, Aaron, and one that hits the nail right on the head. The reason for the 'as such' exceptions is, I am afraid, shrouded in mystery, and the collection of areas that are not to be viewed as inventions are not really connected by any overarching definition or meaning. There is some discussion of this in the CFPH judgment. The EPO tries to connect them with its very unhelpful 'technical' test, but that merely shifts the problem elsewhere, i.e. how do you define 'technical'?

It is very difficult, and can be misleading, if one tries to come up with a 'simple' solution to the problem, i.e. as you suggest saying "computer programs are not patentable" ("end of", as they say). How do you define that? Do you mean that a claim to a method that could be implemented purely (or perhaps partly) on a computer is not allowable? Or perhaps it is not enforceable? I think it would help if the so-called anti-software patent lobby put some effort into developing some concrete ideas about how to change the law in practice that would result in a workable solution. It is simply not enough to say that things at the moment are unfair and something should be done. What exactly?

By the way:
i) surgical methods are not excluded as such, they are regarded as not having industrial application (see section 4(2). The 'as such' exception does not apply.
ii) I don't know where you get the idea that I am pro-software patents. I am merely pro-law, and am happy to implement it whatever it says for the benefit of my clients.

plh said...

"The EPO tries to connect them with its very unhelpful 'technical' test, but that merely shifts the problem elsewhere, i.e. how do you define 'technical'?...
I think it would help if the so-called anti-software patent lobby put some effort into developing some concrete ideas..."

I don't know why you haven't heard of it - perhaps you are too young to remember events such as the CII Directive and the UKPO's tech. contr. workshops etc. - but there is an organisation (the FFII) which has put a great deal of effort into doing exactly that in recent years. It more or less /is/ the anti-software patent lobby in Europe:


David said...

Of course I have heard of the CII and the workshops, as well as the FFII and other groups, none of which appear to have got anyone anywhere. I have yet to see a single feasible suggestion for a change in the law that would make sense and actually work. The "Ten Core Clarifications" sound a bit vague and useless to me, except perhaps number 5.

Gerontius said...

A comment on the "anti-sfotware patent lobby": There is no such single group that can be called by that name, and this causes a lot of confusion amongst programmers as to what a "non-patentable software patent" is. On the one hand, you have people like Richard Stallman who clearly do not understand patents, but whose opinions carry a lot of weight in the open source community because of their knowledge there. Then you get groups like the FFII, who do actually have some suggestions for redefining the law, although it is bloody difficult to get to the bottom of them - presumably because they want to appeal to as wide a group of discontented programmers as possible, including the hard-core opposers.

As far as I can tell, the FFII position is that software that only affects the internal operation of a computer should not be patentable. This covers computer-implemented business methods, as well as encryption/compression techniques such as the very unpopular MP3 or LZW patents and the EPO Vicom decision. However, they are happy when the software controls or affects something outside the computer, so the EPO Koch and Sterzel decision would be OK, as would patents on computer controlled ABS.

The FFII therefore have quite a moderate position, it seems to me, and what they want is not too far from the UKPO's current position, or even that far from what the EPO are doing - they just seem to make a big deal of these differences to ensure the issue isn't forgotten.

Personally, and this is just a gut feeling as I am not an economist, I would probably like to see encryption/compression methods being patentable, since they strike me as being patents on good and useful inventions that just happen to be implemented on a computer. But I would probably have stopped short of the EPO's position in Vicom since I don't see that as providing a technical effect.

As for the UKPO decision. Several commentors have responded to me by saying that "software isn't patentable". I'm not saying that it is. All I'm saying is that the UKPO arguments as to WHY software is not patentable (in some circumstances) are not supported by the case law in my view. If someone can explain how the reasoning in the UKPO decision is consistent with Macrossan (based on the problems I mentioned in my first comment) I would very much like to hear it since I'm banging my head against a wall.

Anonymous said...

Hi David,

You said :
"Of course I have heard of the CII and the workshops, as well as the FFII and other groups, none of which appear to have got anyone anywhere. I have yet to see a single feasible suggestion for a change in the law that would make sense and actually work. The "Ten Core Clarifications" sound a bit vague and useless to me, except perhaps number 5."

The FFII are under-funded and almost entirely volunteers, software patenting doesn't interest almost all voters (despite affecting them by creating monopolies, making software/hardware harder to use, and impeding innovation), and the pro-software-patent lobby have very deep pockets and includes several MEPs, who are representing their clients rather than their constituents. That is why they don't seem to be making much progress, although I believe that recent changes in UKPO and the UK Courts are slowing swinging back to sensible after years of bad decisions. (There are plenty of pure software patents that are clearly software, but slipped through while the courts and UKPO were asleep at the wheel).

I think the cc10 are very clear - much much clearer than anything the UKPO, EPO, EU Parliament/Commision or UK Parliament have ever provided. I think the problem is that Patent Lawyers don't understand software anywhere near as well as software engineers do Law.

The definition of a computer program is perfectly clear to anybody with a grasp of the subject, if Patent Lawyers have to ask Software Engineers for help, then I see that as a very postive rather than negative step, as left to their own devices Patent Lawyers/Clerks and Politicians are responsable for the current mess.

Geronimo - the problem is that the case law is an abysmal mess, even relatively recent cases are not consistent with each other, because the "as such" nonsense, makes the law less clear.

It would help immeasurably if the law itself was updated and a clear line drawn in the sand, because the UK and European Patent Offices have been too lax and given in on the patenting of what are clearly computer programs - "as such" or otherwise. When you have such a trainwreck of unclear language and poor decisions by both judges and patent offices, further tinkering won't help and you have to revise the legislation itself.


Aaron Trevena, BSc Hons
Software Engineer.

David said...

"I think the problem is that Patent Lawyers don't understand software anywhere near as well as software engineers do Law."

Excuse me while I fall off my chair in amusement.

Anonymous said...

I'd help you up, David, but I'm laughing too hard.

I'm a patent attorney with a software background, but I reckon I understand software better than I understand the legal and policy issues surrounding the patent protection of computer-implemented inventions.

The real problem is that the area is complicated and fraught. A sub-problem is that naïve, one-sided lobby groups fail to recognise that the issues are complex (nothing against the anti-software lobby - lobby groups are typically naïve, and invariably one-sided).

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':