June 26, 2006
Becoming Less Obvious
This morning, the Supreme Court granted KSR's writ of certiorari in its appeal from infringing Teleflex's patents. At issue is whether prior art anticipation is obvious absent any suggestion or motivation to combine references. Many patented inventions are combinations of what had been previously been known, but the combination applied in a way hitherto unknown.
The current case law standard requires "suggestion, teaching, or motivation that would have led a person of ordinary skill in the art to combine the relevant prior art teachings in the manner claimed." Under the current standard, the suggestion may be implicit; the court hearing the argument for combination must simply find it compelling.
In agreeing to hear the case, the Supreme Court (SC) has signaled likelihood of lowering the bar for finding anticipatory prior art in combination; few observers doubt that. The real question is whether the SC will provide a usable compass that lower courts may consistently employ. The current SC is deeply divided, with common near-split decisions, multiple concurring opinions that vary widely, and its rulings this session often don't give lower courts workable guidelines. This observer considers the current Supreme Court ill-disciplined, at times appearing inadequate to even useful interpretation of the law; e.g. the recent wetlands case (Rapanos vs. United States).
The recent SC eBay decision only went so far as to signal displeasure with the previous semi-regular pattern of granting permanent injunctions once infringement was found, and taking pot shots at irrelevancies, such as the relative merit of business method patents. A subsequent district court ruling in z4 v. Microsoft saw that transformed into a case-by-case application of "the principles of equity," where equity took into account the infringer's well-being as an equal counterweight to an inventor's fundamental right to prevent others from practicing a patented invention. That trial court decision, though carrying little weight by comparison, was more eloquent and useful than the Supreme Court ruling.
Computer companies would very much like to weaken patent protection, while pharmaceutical companies rely upon patents for their profitability, and thus sit on opposites sides of this fence. But the root issue here is not pro- or anti-patent, it's for providing a coherent standard that can be readily applied by the patent office and the lower courts.
Regardless of the decision, what does seem a likely outcome of the SC taking the eBay and KSR cases is lessening the outcry for patent reform legislation, the computer industry being somewhat assuaged, feeling less threatened with injunctions and feeling more confident it can kill junk patents.
Here's solid coverage of this case from Dennis Crouch at Patently O.
Posted by Patent Hawk at June 26, 2006 7:44 PM | Prior Art
David French writes:
Here is a suggestion for new language on the topic of invention obviousness.
The Golden Rule of patent law is that a patent, once granted, should not interfere with anything that was previously available to the public. What is previously available to the public includes, not only what has been done or described previously, but also obvious variants on what has already been done or described.
In otherwords, the test for "obviousness" is whether what is proposed as a new invention is so similar to the prior art that it was, effectively, already "available to the public". In which case, this obvious variant is assimilated under the Golden Rule as being part of the prior art and should not be patentable.
I'm simply searching for a new perspective on the issues.
Posted by: David J. French at June 27, 2006 12:26 PM
David, with all due respect, your new language adds nothing to the issue for which cert was granted in KSR. It seems like your definition of obviousness will cause as many, if not more, problems as the current definition.
The main issue essentially boils down to what evidence is needed to establish prima facie obviousness. For example, assuming that the Examiner has shown references teaching each limitation of the claim, does the Examiner need to show more? If so, what and how much more?
Posted by: John at June 27, 2006 1:41 PM
I think there are two issues here: the first is what is the standard for determining when a combination of prior art references renders a patent claim obvious. Look at the Graham factors: I doubt there will be any change in the "scope and content of the prior art" prong: to get in the door an accused infringer needs to show that all of the limitations of an asserted patent claim are shown in a combination of prior art references. The 64 billion dollar question (literally) is what standard must be met to combine the prior art references.
There is the old synergism/spark of genius test from the Supreme Court. The problem there is that there is no underlying objective analysis to apply: it's simply "I know it when I see it."
There is the European inventive step analysis. While adopting thist test would foster harmonization, my experience so far has been that US practitioners, some examiners, and patentees are somewhat hostile to this test for reasons similar to the opposition to synergism.
One solution I like is burden shifting: once the accused infringer presents a prima facie case of obviousness via a combination of references containing all elements in the asserted claim, then the burden should shift to the patentee to show why, to the preponderence standard, the claimed combination is NOT obvious or why there is NOT a motivation to combine the references. The patentee is typically in a much better position to show NONobviousness than the asserted infringer is to show obviousness. Sometimes this would be simple: if the combination is from disparate art fields, or if there is no motiviation to combine, the patentee's job is easy. However, when a good combination is asserted, then the patentee should have to show (just like in obviousness prosecution practice) why the combination is patentable, what is new, what is an unexpected result, etc., in the new combination.
Posted by: doug at June 28, 2006 9:02 AM
Yes, I think burden-shifting is the way to go. For combinations that produce no new result, I think the simple answer is to say that there is a presumption of obviousness. Then the burden shifts to the applicant to prove, with evidence, that it is non-obvious or that the presumption should not apply. This is not that difficult or onerous.
Burden shifting is a simple tweak to the existing test and thus does not overturn decades of case law. I don't think the Court is likely to deviate too much from the established law of the Federal Circuit
Posted by: Anthony at June 28, 2006 11:18 AM