There has been a lot of press coverage of technology patents over the past year or so, what with the Apple v Samsung litigation occurring across the globe, and the latest news comes in the form of yet another patent granted to Apple – a super-broad ‘slide-to-unlock’ patent. You may recall that slide-to-unlock was one of the features that Apple sued Samsung over, and that to avoid such lawsuits, the lock screen in Android 3.0 was radically overhauled. Apple’s original two patents covered unlocking a touchscreen device if a predefined gesture was performed, that is, moving the unlock image to a predefined location, or moving the unlock image along a predefined path, such as the unlock gestures in iOS and Android 2.x. With Android 3.0 (and 4.x), the unlock screen has been less restrictive, removing the predefined path, allowing you to move your hand over the screen however you like, as long as it ends up in the right place.
The new patent does not include the requirement that the unlock image is moved along a pre-defined path, or to a pre-defined location, but rather covers an implementation where an unlock image is moved to an unlock region on the display. This is considerably broader than the previous two patents, and definitely covers the unlock behaviour in Honeycomb and beyond. Another of the claims in the patent is the transitioning of the device from a locked state to an application if the unlock image is moved to an unlock region, which would include lock-screen shortcuts that have been a part of some custom builds of Android for a very long time.
So, considering that Apple’s new patent seemingly covers something that was invented a long time ago, why would it have been granted? I’m glad you asked, dear reader. The reasons are twofold – patent applications aren’t really examined in any great details before they’re granted; and America’s patent system is crazy. But to understand this, let’s get into a bit more detail.
The basics of patent law in Australia
Patents are a huge industry all across the world. They are intended to protect invention, and to reward inventors for investing time and money into developing new and exciting technology for the betterment of humanity. In return for sharing the technology with the public, inventors are given 20 years during which they control who can use the invention described in the patent. But there are requirements on what inventions can be patented: the claimed invention must form ‘patentable subject matter’; it must be a new invention, and must demonstrate some inventiveness. I will discuss these requirements in more detail below. The most important thing to note though, before we continue further, is that the granting of a patent is not proof of its validity. For administrative reasons, the Patents Office only conducts a preliminary examination before it will grant a patent, and that the validity of a granted patent must ultimately be decided by a court.
Patentable Subject Matter
Not all inventions are patentable, based on the nature of the invention. For example, it is not possible under any circumstances to patent a human being, or the biological process for the generation of human beings. Computer software is not expressly excluded from patentability, but its patentability is highly contentious. In many countries, computer software is deemed sufficiently well protected under copyright law, that it is not possible to patent a purely software solution. This is a very complex area of the law that is far from settled. For example, in the European Union, programs for computers are expressly excluded from patentability, and yet courts have held that patents for software that provides a non-obvious technical contribution can be valid. In Australia, the Patent Office has this to say:
‘where an invention is directed to the improved operation of a computer, or advantage achieved by the implementation of a method by computer (as in CCOM), then the requirement for a manner of manufacture is likely to be met. However, where an alleged invention refers to computer implementation but the advantage of the invention does not arise in the operation of the computer, for example if the advantage of a method predominately lies in the assemblage of intellectual information, then this may not be the case’ (from the Australian Patent Office Manual of Practice and Procedure, section 22.214.171.124)
Yep! Super confusing. Examples of software patents held to be valid were methods for drawing improved curves on a screen, or for rendering Chinese characters. Basically, as long as the claimed software invention actually does something, it will form patentable subject matter. Whether the subject of the application forms patentable subject matter or not is merely a threshold test. The application will still have to pass the other tests described below.
In Australia, it is not possible to patent a design, such as the ‘look and feel’ of a device. You might have heard about design patents in the context of the Apple v Samsung suit in the United States, but unlike America, Australia has a separate registration system for designs, which are afforded a much lesser protection, only being granted an exclusivity period of five years.
For a patent to be valid, the claimed invention must be new. That is, at the ‘priority date’ (in Australia and most other countries, this is the date at which the application is made, but in America this is the date at which the invention is invented), there must not be any ‘prior art’ that discloses the claimed invention. What this means is that there must not have been any publication of the claimed invention or any invention that is substantially the same as the claimed invention prior to the priority date. The terms ‘prior art’ and ‘publication’ can be misleading – there is no requirement that the prior art must be any drawings or documents, but the invention itself can serve as prior art. For example, if a patent application is lodged for a method of unlocking a mobile device by moving an unlock image to an unlock region, and that method has been used prior to the lodgement of the patent application, then the previously-implemented method will be prior art for the patent application, and the application will not be valid.
For an invention to be patentable, it must involve an inventive step, which is determined by whether the invention would have been obvious to a person skilled in the relevant field. This is a very complicated enquiry, which depends on a number of factors ranging from what forms the prior art base against which the inventiveness of the claimed invention is to be compared, as well as the level of skill and imagination of the skilled person. It is enough for our purposes to know that an invention cannot be patented if it is an obvious improvement over existing inventions, such as making a spoon from metal rather than wood. But a patent will be valid if a person takes an existing invention and improves upon it in a non-obvious way that contributes to its usefulness.
How is the US system different?
I’ve mentioned above that the US patent system is different to most of the world. There is a lot of litigation over patents in America, and there are companies who generate profits entirely from patent litigation, which are known colloquially as ‘patent trolls’. There are a number of key differences between the US system and the rest of the world, which I will now briefly discuss.
The most obvious difference between the US and every other country in the world is the point at which patent rights become enforceable, which is called the priority date. The entire world except for the US works off a ‘first-to-file’ system, where the first person to file a valid patent application will get the patent, whose rights are enforceable from the date of the application. The United States, at least until the America Invents Act comes into effect in March next year, uses a ‘first-to-invent’ system, where the priority date is the date of invention, regardless of whether the first inventor files a patent application or not. It seems like a great idea that the person who invents something first will be guaranteed rights in respect of their invention, but the system creates practical difficulties. People working in laboratories or research centres must keep highly detailed notebooks of their work, and if two people come up with the same invention at roughly the same time, at what point should the invention be deemed to have been invented for the purposes of the patent grant? Furthermore, if two people come up with substantially similar inventions, the inventors will have to spend a lot of time and money arguing about who invented it first, and where a person has been granted a patent, they could still be held liable for infringing the very same patent if at a later date, a second person shows up with documents proving that they actually invented the invention first. The law is generally designed to prevent uncertainty about legal rights, and the first-to-invent system creates uncertainty in the rights it is designed to protect.
Secondly, and relevant to this latest patent, is that the United States grants what are known as ‘continuation patents’. Put simply, continuation patents are attached to other patents to broaden the scope of the original patent, and the claims in the continuation patent will have the same priority date as the original patent. The requirement for a continuation patent is that the additional claims must be able to be supported by the original patent, and since Apple’s original slide-to-unlock patent was sufficiently broad, the new claims are able to fall within its scope. Because the priority date of the new patent is the same as the original patent, Android’s more modern unlock implementations would not form part of the prior art base for the new claims, and so they would not fail for want of novelty. It is quite worrying that software features that were designed to avoid patent liability can now infringe on an existing patent because a patent holder can expand the scope of their prior patents retroactively, but thankfully continuation patents are not available under Australian law.
The third, and perhaps most important difference between the US, and particularly the Australian patent system is that in the US, patent infringement cases are decided by jury, whereas in Australia and many other countries, such cases are decided exclusively by a judge. Patent law is highly complex, and to understand it requires expert knowledge of both the law and the subject area of the patent, which a random selection of people are unlikely to have. Experts in the field of the invention will present evidence in court, and the judge uses the expert testimony, plus his or her intimate knowledge of the law, to come to a conclusion about the validity of the patent. In the Apple v Samsung case in the US, we saw that juries can be fairly easily swayed, and are not always able to understand the many complexities of patent law, nor the technical aspects of electronic or software engineering.
Does a patent being granted in one country automatically mean it’s automatically granted in others?
No. The Paris Convention for the Protection of Industrial Property allows an applicant for a patent in one member country to easily submit his or her application in all convention countries, and that the priority date for the application in all member countries will be the date of filing in the original country. While the Paris Convention simplifies the process for applying for protection in multiple countries, the patent application is individually assessed in each country, according to their own patent laws and generally without reference to the application’s progress in any other country. Accordingly, the grant of the most recent ‘slide-to-unlock’ patent in America does not mean that the patent is valid in Australia, and an application under Australian law for the same patent would likely fail because of the prior art that is Android’s newer unlock implementation.
How are fights over US patents likely to affect Australia?
Well basically, because of the significant differences between patent law in the US and the rest of the world, decisions in patent infringement cases in America have little bearing anywhere else. While Apple won over $1 billion against Samsung in America, a similar law suit was completely dismissed in the United Kingdom. Australian judges tend to do a rigorous examination of prior art in patent infringement cases, whereas juries in America tend to dismiss prior art claims without much regard to their substance.
Realistically though, because Apple and Google are both American companies, attacks on Android in the United States are likely to affect the way Google designs it in the future. Hopefully Google re-invents Android for the better when necessary, rather than removing features and crippling our beloved OS to avoid law suits, but it is obvious that we have only seen the beginning of the patent wars.