Protecting Innovation, Part 2

Last time, I looked at the foibles of the patent system and how it is increasingly being used as a litigious weapon against innovators – ultimately those with a lower bank balance; on the theory that if you can’t compete with your competition on your product’s grounds, prevent them from being in the market at all.

I said I’d look at what could be done with patents – I happen to think that patent reform as a whole is necessary to truly protect innovation.

The first problem is the USPTO itself. Given some of the recent filings, it seems suggestive to say that they do not really understand the material being submitted, especially with the more recent increase in computer science/engineering patents being filed in contrast to the conventional engineering/mathematical basis of the PTO.

Probably the greatest evidence for this speculation appears to be the number of patents filed where there are examples of prior art that are brought into play when the patent is held up in court; the examinations do not bring this into the light during review. It should be noted that this is not the only explanation, it is entirely possible that the infrastructure for searching for prior art is not adequate, however it would seem likely that this is related, rather than a completely separate factor.

The second problem is the fact that patents in their current form are a very double-edged weapon, and need to be cut back down to size.

I suggest the following two-pronged approach to dealing with it:

1. When bringing a case to court, the plantiff (person bringing the case) should be required to prove the patent is currently in use by them. I don’t really care how they do it but they should have to prove that they are currently using the patent in a non trivial way, such that companies cannot acquire a portfolio of patents and just start handing out lawsuits – and more importantly, they cannot file and then wait for years for the patent to inevitably be thought of again, only this time successfully.
2. Companies should be forced to renew the patent more frequently than they currently do – annually, or bi-annually, but not in such a way that would be onerous to perform. The cost, however, should be non trivial, rise each time but also be inversely linked to the cost of enforcement – so if a company does use a patent but it’s not triggered in court much, it’s probably not worth that much to them, so the renewal could be higher. The idea is that it would encourage companies to abandon them after they have acquired their costs back, and set them free.

That second point is quite important, and I want to draw on that some more. Patents do have valid uses, probably the best example is if a major drug were to be created – say, an instant cure for cancer. It would have taken vast amounts of R&D time, effort and money to develop, and such investment does not happen out of charity; there must be a monetary reward for a company engaging in this practice.

The idea then is that they file a patent on the drug, to protect their investment, and once their costs have been recouped, there is less incentive for them to hold onto the patent and so they will either pay more in reapplying for the patent (putting aside enforcement costs) or let it go such that others can make use of it.

Different industries have different product life-cycles; pharmaceutical companies tend to have a longer cycle (3-5 years) for new drugs due to testing and regulation while industries like IT have much shorter cycles, so invariably they could relinquish patents quicker.

The whole point of this is to put an end to many trivial patents that are simply in existence to be used as weapons, and to put an end to patents that lay dormant before being brushed off and used against those who would do innovative things on the same general approach.