I have seen research that has resulted in patents that become part of a companies defensive patent portfolio. At the time of research, the development was being done for innovative reasons, but for reasons of cost, strategy etc the innovation is put on the backburner over time.

Hence it is likely to be difficult to stipulate what is a ‘defensive’ patent and what is an ‘innovative’ patent, particularly at the time of filing, and it is difficult to see how this can be resolved without making the patent system even more unwieldy (e.g reviewing the use of a patent over time)

Tactical patenting is not necessarily an immoral thing to do. If you company has a large and effective patents department, then you should have a collection of bright people who can spot ‘crossover hits’ were an innovation in one field can be applied in a completely different field. Such ‘crossover hits’ are often very valuable – they can go directly to a useable product without the heavy R&D risk. And, if they don’t to it, someone else will.

What we need is some way of reducing tactical patenting. We can only do this with clear and fair rules. Disallowing software patents is a very good case of such a rule. I do not believe there is any absolute argument why software patents should not be allowed – but I do feel the arguments for allowing them are very poor, and preventing them will also do away with much unnecessary work. If I was still doing patents for Canon, I would regard a worldwide ban on software patents as a relief – relief that we no longer have to work against the threat of being tactically outmanoeuvered by other
companies. The other side of the coin would be we would also lose the slight possibility of downing a hated rival using a software patent, but I could happily forgo that and all the risks involved.

MEMS technology allows us to make much smaller versions of familiar designs. It is far from clear whether translating a maroscopic device into a microscopic, integrated device is invention or just good design. Patenting a second use for a patented drug is allowed – this encourages us to re-investigate already tested compounds instead of having to endlessly innovate new ones, but it is capable of being used tactically. However, there seems little case for anti-tactical patent rulings in these fields unless we first forbid software patents.