Patents and Juries

By Francisco Tolmasky on

One of the things that has surprised me a bit about the recent discourse relating to patents is the increasing number of people commenting about how they feel that perhaps juries are not the best final check on this system. I find this reaction curious, if not a bit frightening. I’ve seen people unhappy with verdicts before, but rarely does that lead to questioning the very use of juries. I’m hoping to understand why people who I believe would otherwise trust juries feel this way about this particular facet of law.

Let’s start with what I hope are uncontroversial statements about patents. I believe these to be uncontroversial because they don’t necessarily help either side of the argument: they could be used to both support or oppose the current patent system:

  • A patent effectively creates a “mini law” in a sense. This “mini law” simply states that the government has granted the patent holder a temporary monopoly on a certain invention. If you can prove someone is encroaching on this monopoly, the legal system will defend you against them. It’s kind of like a new temporary regulation in a market.
  • Patents can have huge effects on the economy. Again, not judging whether these effects are good or bad. Patent defenders will say that it has a huge positive effect by protecting inventors, while patent detractors will say they have a huge negative effect by harming smaller companies and stifling innovation.

The point of these two statements is that hopefully we can all at the very least agree that the questions relating to patents are important, and thus worthy of our scrutiny. However, when we scrutinize things, it’s usually popular to get bogged down in philosophical positions (“can a business method really be owned?”) or incredibly difficult to prove economic theories (“while usually detrimental to the point of requiring anti-trust facilities in the government, a small temporary monopoly can actually help the market in the long run”). We often take for granted the actual system that’s implemented, and consider only it’s idealized abstraction. For example, we talk about what “congress” wants, and forget that it’s just 535 human beings with sometimes completely opposing points of view. Similarly, we talk about the “patent system” as some sort of machine that would execute the goals we want, when in reality it too is ultimately just an organization of individuals.

So let’s not look at the ideal our system tries to achieve, but instead the stark reality of what we currently have in place. Luckily for us, there are a number of statistics published about this organization (the USPTO). So let’s take a look at them:

  • In 2011, there were 535,188 patent applications. 30,467 of these were Design Patent applications.
  • In 2011, there were 6780 patent examiners (according to wikipedia only 99 of these were design examiners!). Taking into account that the USPTO doesn’t work weekends, this means that to get through all the applications in one year would require an average of 3 days of scrutiny per patent. In reality this is probably to blame for a backlog of over 700,000 patents.
  • There is little to no public accountability for patent examiners at the USPTO. By this I mean that the USPTO is obviously not staffed by elected officials. In fact, there is a tenuous connection between anyone you may vote for and USPTO policy.
  • According to Wikipedia, patent examiners “are generally newly graduated scientists and engineers, recruited from various universities around the nation.” Unfortunately no citation is listed for this, however the USPTO careers page does seem to have an emphasis on recent graduates, and looking at job postings on the USPTO site demonstrates that the requirements do indeed seem to be just having a 4-year degree. Additionally, no law degree is required.
  • The salary range for a patent examiner seems to be $51,988 to $78,881, well below what an “expert” in the field would make in the industry.
  • Very questionable patents have been granted in the (recent) past. Everything from the one-click Amazon patent to faster than light travel.

I think by definition the job of a patent examiner is an extremely difficult one: they are meant to analyze, understand, and draw conclusions regarding the latest in bleeding edge technologies. I think most people would want a proven physics expert analyzing a newly invented nuclear engine, not a recent physics graduate. A recent graduate might consider more things novel, or be more easily fooled, than someone experienced in the field, wouldn’t you agree? I consider myself pretty proficient in computer science, having attained a degree and worked on a number of products and companies, but I only consider myself an expert in a very small part of my field, and only if I’ve just recently focused on it. In other words, I don’t know if I’d even trust myself with this job – even if I was meant to only look at the things that I directly had experience with. I think all it takes is reading a few actual computer science papers to quickly start doubting your own expertise in this field. This doesn’t even get into the question of whether such experts, if they existed, would feel compelled to enter the USPTO instead of actually inventing new things. What’s particularly frightening about this to me is that I only know about these issues precisely because it’s in a field I understand: not being in fields such as biology, it scares me even more what must be going on with those patents. Perhaps 6780 people could be enough to cover all the subfields of computer science, but are they enough for all subfields of all the sciences?

The response to this is that the USPTO is currently attempting to increase its hires by 1000, with “more emphasis on recruiting candidates with significant IP experience”, in contrast with “previous hiring which focused on scientific background and experience”. One of the benefits listed is that “this will result in reduced training time as well as an increased ability to examine applications much sooner than a new hire with little or no IP experience, thereby increasing overall production output.” (source) This shows the Catch-22 and cross-displiclnary nature of the situation: I’m not sure more people with a law background to get patents through the door quicker is the appropriate response to these problems.

The conclusion I reach looking at these statistics is that the USPTO is obviously understaffed, and more importantly necessarily under-qualified, even in the ideal scenario, which their current staffing certainly does not represent.

The reason this is ultimately important is to establish some context to our original question of juries. In other facets of law, we generally have an interesting balance: publicly elected and (ostensibly) professional people making rules which are only then tested in court against the sobering “common sense” of society, which offers a way out of possible group-think of people in that field. If the rules being created are repeatedly bad, we can ultimately choose someone else through elections. In the world of patents we instead seem to be missing this balance: on the one hand you have an overwhelmed and under-qualified entity establishing rules for commerce. It is difficult to know who to blame or how to do anything about it if they repeatedly make decisions we disagree with. And then we of course have these rules tested by 12 (effectively random) individuals. So what we are left with is incredibly complex questions being passed to one group of people with questionable expertise called the USPTO (which we as a voting population can’t realistically affect), and then finally to a second group of individuals who are by design also rarely experts. Regardless of where you fall in the patent dispute, I’m not sure how you can trust the decisions being made today by this arrangement.

Imagine momentarily if other sectors of law worked this way. What if you could apply for new traffic codes to the DMV, which would be approved or denied by DMV staff, but would then be enforced by actual state police. Then, your only option was to hope that a jury found this traffic code unreasonable. I think then you would start to see more people unhappy with the way courts functioned.

If this example seems silly or far fetched, consider that patents today can determine whether entire product lines can exist, whether a health company could potentially make a life saving product (or a generic and affordable drug), or increasingly complex questions such as whether we can patent bacteria or genes. Again, the point is not what side you fall on with these questions, but rather that these questions are important and largely beyond accountability to both actual experts and the voting population. When you take this into account it quickly becomes obvious that our current real-life patent scenario is actually much scarier than the fictional traffic code one.

These have the potential to be some of the most important decisions of our lifetime, and they are for the most part completely out of our control. This is why people are so frustrated by these patent trials - it is a feeling of helplessness. The granting of a patent can be as influential on our careers or our lives as the passing of a law, and yet it’s hard to feel like anything other than a mere spectator in the process. It’s inevitable that there will always be disatisfaction with verdicts, but juries here don’t feel like some last line of defense in an already accountable and trusted system - they instead seem like the only real chance for input the public had on these individual patents. And thus we more so feel they had a duty to represent our particular personal opinions.

This feeling is however unfair. Clearly the problem is not with these juries, but instead with the way in which patents are granted in the first place. So how can you create a system that practically makes appropriate decisions about patents? How many people would it take? How would you attract the qualified people, and would it be worth it to take them away from the industry? How could you make them accountable? And how could we realistically impose such a change and maintain such a system in the near future? I don’t think that these are simple matters of staffing or policy changes. Regardless of whether you believe in patents or not, its pretty clear to me that the current system is broken and needs some fundamental changes.

Discuss On Hacker News
Follow me on Twitter