Open Enterprise
Glyn Moody
Subscribe to this blog
About Author
Glyn Moody's look at all levels of the enterprise open source stack. The blog will look at the organisations that are embracing open source, old and new alike (start-ups welcome), and the communities of users and developers that have formed around them (or not, as the case may be).
Contact Author
Email : Glyn
Patent Differences: Canonical vs. Microsoft
Published 10:27, 11 May 09
I make no apologies for returning to the subject of the European Patent Office's referral of a “point of law” concerning software patents.
Dull as many might find the intricate theoretical arguments, the outcome will have very real consequences. If software patents become easier to obtain, it will have a hugely negative effect on free software, which will find itself subject to more attacks on the legal front.
Recently I commented on the submissions of Red Hat and the FSFE. The full list of “amicus curiae briefs” can be found here; I'd like to pick out those from two high-profile names for their contrasting positions: Canonical (the company behind Ubuntu) and Microsoft.
Canonical's offering is very similar in tone to that of Red Hat: it's very matter of fact, written in a highly-accessible language that makes its points simply but effectively. Here's its main argument:
In Europe, computer programs, i.e. software, are protected by copyright as a creative work. It is Canonical's clear and firm belief that software does not require further regulation from any other form of intellectual property and that attempts to provide further protection via patents are and have been counter-productive.
Copyright arises at law and does not require registration in Europe. It is a free protection offered to the creative work of software developers and their employers. It is an appropriate way to protect both closed and Open Source software.
Money is not a barrier to entry to the reward of intellectual property protection of copyright in software. It is therefore a genuine reward and protection for small companies, community projects and individuals developers as well as large corporations, offering everyone automatic protection on a level footing. This is very relevant in Europe as “more than three fifths of the worldwide FLOSS developer community live in the EU.”
Canonical also has some harsh words to say on the vexed issues of what “as such” and “technical character” mean (something I've also written about at length):
Article 52(2) [of the European Patent Convention] clearly states that software methods are not inventions and do not attract the coverage of patent protection, however, the EPO has become increasingly focused on the analysis of the use of the words “as such” and whether this in fact should imply that there is some inclusion of computer programs in the patent regime.
The phrase “as such” does not have any special meaning in law. It is our view that this is an attempt to analyse working with a view to achieving software patentability despite the clear intention of the legislation. Article 52 clearly states that computer programs “shall not be regarded as inventions”.
In attempting to follow through the logic of the “as such” debate the Board of the EPO has further engaged in a second somewhat artificial anahysis, and created a ground for the distinction (as to whether patentability exists) to be made, depending on whether or not the program has a “technical character”. It is hard to imagine how software can not have a technical character?
Canonical would respectfully suggest that the wording of Article 52 should be interpreted in a simple fashion as it was probably intended and that this artificial distinction between “software” and “software as such” be removed. In turn this would remove the need for the tenuous further consideration of “technical effect”.
This is essentially the same as Red Hat's position. Microsoft's submission, made jointly with General Electric (GE), not only takes completely the opposite position, it also takes a completely different approach from Canonical. Where the latter's paper is down-to-earth and couched in everyday language, Microsoft's is a rat's nest of legal points, quibbles and citations, and is written in stiff and formal legalese.
The main thrust seems to be pointing out the existing case law that supports the patenting of software. Canonical, by contrast, argues against software patents as a matter of principle and pragmatism.
Microsoft and GE conclude their attempt to buttress and laud the current system with the following:
From the perspective of GE and Microsoft as frequent applicants and significant patent holders, it does not appear that obtaining patents for computer-implemented inventions has become easier, but our perception is that the process of examination is more logical and orderly, and the ultimate outcome is more consistent and predictable.
This type of predictability and consistency is crucial if the patent system to have its intended effect of creating incentives to invest in research, development and commercialization of new technologies. The more that innovators, patent owners, and third parties are able to rely on predictable, consistent outcomes in examination, the more likely it is that they will rely on the patent system to incentivize their efforts to advance development of technology and its commercialization.
Irrespective of the legal arguments offered, this is where the fundamental misrepresentation at the heart of the joint Microsoft and GE submission is most blatant. Whether the statement “the more that innovators, patent owners, and third parties are able to rely on predictable, consistent outcomes in examination, the more likely it is that they will rely on the patent system to incentivize their efforts to advance development of technology and its commercialization” is true or not, is irrelevant.
The point is, they don't *need* to rely on the patent system: software is not only able to survive without the crutch of patents, it actually thrives. Indeed, the best single proof of this is Microsoft itself: for decades it developed and commercialised technology without taking out software patents so successfully that it came to dominate the software industry as a monopoly.
Software simply does not need to be patentable in order to encourage innovation in the world of computers: innovation lies at the heart of coding, since this is how old programs are improved, and new ones written. Instead, patenting software brings with it nothing but disadvantages in terms of making it harder for lone inventors or small companies to deal with the massive legal risks and costs that software patents impose.
It is no coincidence that two of the biggest corporations in the world, Microsoft and GE, should get together to present this love-letter to software patenting: both stand to benefit hugely from the additional burdens such patents impose on everyone, because they can support them better than smaller competitors. Allowing software patents simply distorts the market in favour of larger players.
Even though the logic is all on Canonical's side, I do worry that Microsoft's submission, couched as it is in precisely the graceless prose that software patent lawyers love, will go down better at the European Patent Office. In essence, Microsoft speaks the language of the EPO, and its main argument boils down to shrewd flattery of the EPO's previous decisions in this sphere.
Unfortunately, encouraging the EPO to affirm that its previous decisions were correct is probably a better strategy than trying to convince it to see the error of its ways, and I fear that it will be Microsoft's rather than Canonical's view of software patents that prevails.
Follow me @glynmoody on Twitter or identi.ca.
Print