Software Patent

It is often said that you cannot patent software. This, however, is difficult to square with the vast number of patents granted by the European Patent Office (EPO), and many other patent offices around the World, that have claims beginning “a computer program comprising…”. As it happens, patents to software in Europe are widely sought after and commonly granted. So why is there such a misconception?

The European Patent Convention, which sets out the rules governing the grant of European patents, states that:

“European patents shall be granted for any inventions, in all fields of technology, provided that they are new, involve an inventive step and are susceptible of industrial application”.

A qualification is added to this stating that “rules and methods for performing mental acts” and “programs for computers” shall not be regarded as inventions – seemingly ruling out patents for software. But this qualification itself is then further qualified indicating that such exclusions apply only to the extent to which the patent relates to such subject-matter “as such” – perhaps leaving the door open.

It is therefore quite easy to see why the patentability, or otherwise, of software at the European Patent Office can appear rather opaque to those on the outside.

In reality, objections against computer software inventions based purely on the above prohibitions are rarely sustained. The inclusion, or use, of any hardware (however generic) means a claim is not towards a “computer program as such” and therefore outside of the above prohibitions. However, that is not the end of the story when it comes to patentability of software at the EPO.

As discussed in another article in this series – “Inventive Step at the EPO” – the EPO typically adopts what is known as the problem and solution approach when assessing inventive step. It is here that the EPO concludes not all software is created equal and it is here where the EPO discriminates between the various “types” of software it considers patentable or not.

In short, the software must produce some “technical” effect, solving a “technical” problem for the EPO to consider it patentable. Parts of the software that do not contribute to producing this technical effect and solving this technical problem are effectively ignored as irrelevant when the EPO apply their inventive step analysis.

For example, take computer software which uses a clever new mathematical algorithm to calculate annuities in a pension scheme to better manage financial risk. This would likely be rejected on the basis that the only effect of the new algorithm is the reduction of financial risk – something the EPO would not see as technical. The EPO would conclude such software lacked an inventive step as, although the algorithm is new, it does not give rise to anything technical happening.

Conversely, computer software which uses a new mathematical algorithm to enhance an image would be viewed rather differently by the EPO because enhancement of digital images is regarded as inherently technical. Therefore, the new mathematical algorithm in such a program would be seen as giving rise to something technical – image enhancement. Such software may very well be patentable, providing the use of the new algorithm is non-obvious (as discussed in “Inventive Step at the EPO”).

Unfortunately, the term “technical”, is barely used at all in the European Patent Convention, instead being developed as an idea in case law over several decades. The EPO’s Examination Guidelines give lists of examples of things that have been found technical in the past – such as controlling anti-lock braking systems, compressing video, encrypting electronic communication, load balancing and memory allocation, etc. Similarly, there are examples of activities which have been regarded by the EPO as inherently non-technical such as improved targeting of advertising based on user preferences, linguistic analysis of text, improved actuarial schemes (as mentioned above) and so on and so forth. However, a lot of software applications do not fall neatly into these categories and may indeed include a mix of activities, some of which are regarded as technical, some of which are regarded as non-technical and some of which are very much on the border line.

Looking at these issues early on in the process of seeking patent protection, ideally even before the stage of drafting the initial patent application, can dramatically increase the chances of later making it through the EPO’s analysis of inventive step. Indeed, good initial groundwork can often mean the difference between potentially patentable technical aspects of a software system being identified and properly put into the patent claims, leading to a successful outcome, and such aspects being overlooked and ultimately leading to a patent application that fails.