Obviously, things that are obvious shouldn’t be patentable. Obvious improvements aren’t worthy of patents, instead being reserved for inventors who possess extra-ordinary skill, not merely every Tom, Dick and Harry to come up with something new …
Unfortunately, figuring out how to actually prove something obvious has plagued the patent system for centuries. Engineering is the practice of applying scientific principles (whether physical, chemical or electrical) to overcome “problems” to building a certain machine, device or composition that someone else thinks “can’t be done.” Engineering is the “yes it can” to the “can’t be done” attitude, and the reasons why “yes it can” prevails are what becomes the subject of a patent application.
The difficulty is, after you’ve described the reasons why the “can’t be done” was wrong, your path to the finished product appears obvious. But if it was so obvious prior to your invention, then the “can’t be done” guy wouldn’t have said that in the first place.
In real life, we don’t necessarily have a conversation with “can’t be done” guy … but the Supreme Court created one the landmark case called Graham v. Deere. The Court basically told us that if we want to know whether something is obvious, all we have to do is create a ficticious person by:
- Figuring out what this person knows – (Scope and content of the prior art.)
- Telling him what the solution to the problem was – (Differences between prior art and the claimed invention.)
- Figuring out how clever this person is – (The level of “ordinary skill” in the pertinent art.)
Once we’ve done this, all we have to do is ask this made-up person whether he would have said “can’t be done” or not. If he would say “can’t be done”, then the invention can’t be obvious. Otherwise, it is, unless we decide that he’s just flat wrong (e.g. the so-called “secondary considerations” or objective indicia of non-obviousness).
Unsurprisingly, this test can lead to some inconsistent results. For one thing, by informing our made-up person about the problem’s solution, we invite a hind-sight based review which the test is supposed to avoid. It’s fundamentally unfair to tell an inventor that his invention is unpatentable simply because, after he’s finished explaining it to us, we now understand how and why it works.
The Court then made this problem worse by taking this very fact-intensive inquiry out of the hands of juries (the part of our judicial system that’s supposed to tell us what happened) and giving courts free reign to strike patents down as a matter of law. Here’s a recent example, summarized by PatentlyO:
It would have been obvious to one of ordinary skill and creativity to adapt the safety mechanisms of the prior art cigarette lighters, as disclosed in Floriot and/or Morris, to fit a utility lighter as disclosed by Shike, even if it required some variation in the selection or arrangement of particular components.
Sure sounds like the court is telling us what the facts are, instead of saying something like: if the jury decides that the variation in selection or arrangement of the components was within ordinary skill and creativity, then the invention is obvious and unpatentable. Judge Newman, in dissent, would have used the latter approach:
Nor is it reasonable to trivialize an improvement by its relative simplicity. To the contrary, the fact that this improvement eluded discovery, and that its advantages were immediately apparent to the marketplace and to the competition, weigh in favor of nonobviousness. At a minimum, summary judgment of invalidity was improperly granted. My colleagues err in sustaining that judgment.
Another commenter suggests that its time to do away with Graham’s confusing, mystery-man test altogether (a “person” that Patent Hawk refers to as Obzilla):
It is time to toss Graham out. And until the courts can devise a legal framework that produces both fair and predictable results, obviousness should be left to juries to decide.
- The Teeth of KSR: Obviousness on Summary Judgment (patentlyo.com)
- KSR Fears Realized: CAFC Off the Obviousness Deep End (ipwatchdog.com)