Abramson on Software and Patent Reform


Tuesday, I was able to speak with Bruce Abramson, who has for years studied the copyright/patent/intellectual property rights conundrum that software presents. And he's not shy about saying that we're on course for a series of collisions.


First, Abramson says the copyright system in this country is broken because it is at odds with modern technology. Copyright law says copyright owners "get to control the flow of information they're creating." Technology, on the other hand, says that no one can control the distribution of bit strings. He says:

Once you get content or information that's translated into a bit string, the law says the designated owner can control its flow, but technology says no one can control it...Technological reality and legal enforcement are pushing in opposite directions.

Next, he says the patent system isn't broken yet, but it's dangerously close. Here's why:

Any system that says we will provide the same rules and offer the same rewards for inventors in different industries independent of the capital investment necessary, the likelihood of success, the relationship between a single innovation and a product, and the time to return, ends in a misallocation of resources. Further, everything you do to make life easier for one sector will make life harder for another.

He points to the tension within the patent system between the pharmaceutical sector and the software sector as a prime example. Software can be developed and taken to market relatively quickly, with not a lot of capital up front. And though software products are pretty complex (in that they consist of "a lot of little inventions"), developers often see returns for their efforts relatively quickly. Pharma, on the other hand, has very simple inventions that take a ton of time and capital up front. Further, because the regulatory process is so long and complicated, it could be a decade or more before the inventor sees any return. In fact, Abramson says if we changed the patent system so that it made sense for software, "we'd never see another drug. Period." Nobody would invest in pharma because there would be no incentive.


When open source is introduced into the mix, there's a completely new layer, Abramson says. Because open source is enforced via copyright law, the question then becomes "What do you do when the open source license your code is released under runs into rights protected and enforced via the patent system?"


In the bigger picture of the patent context, open source is not treated differently than propietary software. Simply put, if you write a piece of software and then later, someone else independently writes the same code, that person has not violated your copyright because independent creation is a defense to copyright infringement. However, if you write a piece of software and are granted patent protection for it, if someone else then independently writes the same software, you are entitled to sue that person to enforce your rights.


As Abramson explained:

Say you have patented software...Later, I independently write a piece of code that does the same thing, completely unaware that you have a patent on something in there, and completely in good faith. There's no trick anywhere. Several years later you're looking at this new open source software that suddenly came across your radar, and you say, "Wait a minute, these guys are infringing my patent." You can then sue anyone who's using it for patent infringement.

So what's the best way to correct the balance of equities? Abramson notes that what should be done and what will happen are totally different ideas. Ideally, he says, we start from a clean slate and come up with a set of intellectual property rights that make sense for software, separate and apart from the traditional copyright and patent systems. But it's Congress' job to do that, and Congress has more pressing problems right now.


For the time being then, the courts are left to decide the issues one lawsuit at a time.