27 February 2007

Softly, Softly on Software Patents

This response to an e-petition "to make software patents clearly unenforcible" is interesting:

The Government remains committed to its policy that no patents should exist for inventions which make advances lying solely in the field of software. Although certain jurisdictions, such as the US, allow more liberal patenting of software-based inventions, these patents cannot be enforced in the UK.

The test used to discern between patentable and non-patentable subject matter in the UK has recently been clarified by the courts, and is applied rigorously by the Patent Office. Under this test, the true nature of the advance being claimed in a patent application must be determined, and if this advance lies solely in the field of software, or another non-technical field such as methods of doing business, the patent will not be granted. If the advance being made by an invention does lie in a technical field, it must also be non-obvious and sufficiently clearly described for the invention to be reproduced before a patent will be granted by the Patent Office.

The recently published Gowers Review of Intellectual Property, an independent review commissioned by the Government, recommended that patent rights should not be extended to cover pure software, business methods and genes. The Government will implement those recommendations for which it is responsible, and will therefore continue to exclude patents from areas where they may hinder innovation: including patents which are too broad, speculative, or obvious, or where the advance they make lies in an excluded area such as software.

Although it is little more than a statement that the current position will be maintained, it does contain the important confirmation that there are no plans to follow the US down the primrose path to the fiery furnace, and that US software patents are not enforceable in the UK. I suppose we should be grateful for small mercies.

Credit, too, to the 10 Downing Street site for responding in this fashion to a petition that garnered relatively few signatures (I didn't see it and so didn't sign).


Step Back said...

Hi Moody.
I like your blog.
It is utterly hysterical.
Higher up (later date) you want to get rid of all patents (and I suppose the Constitution also?) because it allows a patent owner to do crazy things, like suing in a case where the patent owners will get their ass kicked in.

The poetry is nice too: "follow the US down the primrose path to the fiery furnace". Good alliteration.

But how exactly will I know a "software patent" when I see one? Is it just like pornography? One need only stare at it and whammo bammo, one knows that he (or she) is looking at something totally lacking in social decency?

The other day I was in a museum. My heavens. There were even paintings of naked animals. Can you imagine? The horror, the horror.

Glyn Moody said...

Software patents tend to be recognisable because they involve patent claims about software. One of the advantages of declaring all software patents to be invalid is that you don't need to spend any time deciding whether a given application involving software is admissible or not because of Jesuitical casuistry over whether it is "really" a patent on software, or whether it simply uses software (as in the example of car braking systems that is usually invoked at this point).

If it's digital, it's information, and hence unpatentable; if it's analogue, it might be patentable if it's also non-obvious, novel etc etc. So, in a rational world, the brake bits of the braking system are patentable; the software that controls it is not.