Ex parte Sobecks
Appeal 2009013122, Application 10/123,457, Tech. Center 3600
Decided March 3, 2010
The claims on appeal relate to a retail product display which can be updated by a product representative in physical proximity to the display. The limitation at issue, for a representative method claim, was:
via a representative of the product source:
physically accessing the at least one memory interface to couple at least one physical memory to the at least one memory interface,
which at least one physical memory contains fresh product-source-to-consumer relationship-building displayable information suitable to display on the dynamic display.
The Examiner found that a single reference taught all the limitations except for "physically accessing" and "via a representative of the product source." In the final rejection, the Examiner asserted that each of these two limitations had a finite number of choices, and that the claimed choice was predictable:
However, there are a finite number of means for coupling a physical memory to a unit through a memory interface (i.e., physically with cables, inserting a disk, wirelessly, etc.) and there are a finite number of entities which could perform the task (representative, customer, etc.). Regardless of the coupling means selected or the entity performing the task, the same predictable result is achieved: the physical memory is coupled through a memory interface. ...
It would have been obvious to one of ordinary skill in the art at the time the invention was made to select an appropriate coupling means and an appropriate entity to perform the task, considering the predictable nature of the outcome, to enable the system to upload new product information and function as required.
The Applicant took issue with the Examiner's focus on "finite number" as an indicator of obviousness, and made the following argument in the Appeal Brief:
While it is undoubtedly true that there are only a finite number of means to make the described coupling, and only a finite number of entities who could perform the task, such an observation is distinctly unhelpful when assessing obviousness as the exact same thing can be said for essentially any invention. Furthermore, if it is the Examiner's intent to suggest that, by using the word "finite," only a small pool of candidates exist, the Examiner is in error. For example, the claim requires that the coupling be performed by a "representative of the product source." As only a very small fraction of the 6+ billion people presently on the Earth qualify in these regards in a given application setting, this literally leaves billions of entities who might be considered as being potentially available to perform the task who do not meet the requirements of the claim. Although candidates numbering in the "billions" are no doubt "finite," the number is still categorically large.
A prima facie showing of obviousness of course requires more than that the claimed approach be somehow more than a mere candidate amongst a pool of finite possibilities. ...
The Board was unimpressed by this argument because the Applicant failed to appreciate what finite meant in the context of the reference:
Treyz places its promotional system within a retail store context. In such a context there are only two sets of parties who have the product data that is required for product promotional campaigns, viz. the retailer and the product source. Thus, the finite number the Examiner found is two sets of parties rather than six billion individual people.
As to the finite number of means of access, Treyz relies on flash memory, which is a physical form of memory media. FF 04. Thus some form of physical access to the flash memory is required to update its contents. So although there may theoretically be a large number of ways to alter the contents of the memory in Treyz, the reliance on a physical memory card narrows the field considerably.
The Board went on to agree with the Examiner that the claimed features were predictable choices among the narrowed-down finite number, and thus affirmed the rejection.
My two cents: KSR says that obviousness may be likely when there are "a finite number of identified, predictable solutions to the recognized need or problem." The thinking here seems to be that it's obvious to try them all.
Although the solutions must also be "identified" and "predictable," I've always been bothered by the use of the word "finite." I've been waiting for an Applicant or patentee to attack the "finite" rationale, because after all, finite is a really large number. But this wasn't the right case for that sort of attack.
In some arts, the number of choices is so large that it's probably not obvious to try them all. In re Kubin, a biotech case, described this as "throw[ing] metaphorical darts at a board filled with combinatorial prior art possibilities," and said that obvious-to-try is not appropriate for such cases.
Sobecks was a software application with business method-y aspects. According to the Applicant in Sobecks, the universe of choices for which party would physically access the memory was really large: any person on earth. But "which party" isn't the same as "which individual," and there were only a few parties involved in the context of promotional displays in a retail store. Maybe the retailer who's using the display, the seller of the product being promoted, and the customer. So that's three, not six billion. The Board further narrowed it down to two, because only the retailer and the seller have the information which will be updated through the interface. But increasing the number of choices to three doesn't change the outcome: choosing among three is way more likely to be obvious than choosing among six billion.
Related posts: I think the Applicant was smart to realize that framing the number of choices affects obviousness. In the Ex parte Sato case that I blogged about last year (here), the Board reversed an obviousness rejection after agreeing with the Applicant that a line has an infinite number of directions that are orthogonal to it.