KSR re-emphasized “obvious to try.”.  From KSR:

“When there is a design need or market pressure to solve a problem and there are a finite number of identified, predictable solutions, a person of ordinary skill has good reason to pursue the known options within his or her technical grasp.  If this leads to the anticipated success, it is likely the product is not of innovation, but of ordinary skill and common sense.  In that instance the fact that a combination was obvious to try might show that it was obvious under § 103”.

The Supreme Court has enabled the USPTO a free hand with “obvious to try” rejections due to the broad nature outlined in KSR. It can be difficult to convince an Examiner to withdraw obvious to try rejections for precisely this reason. Consider a situation where two or more actions are claimed in a method, and where the particular actions are claimed as being conducted at distinct times. For instance, consider a magnetic resonance imaging (MRI) scan, where an example claim states to conduct a first action during the scan, and conduct a second action after the scan. Now assume the first action and the second action are each taught in the prior art, but not at the specified times, and furthermore, no prior art is directed to “after the scan”. 

In such a case, an examiner may state that, for an MRI scan, there can only be “scan” or “non-scan” conditions (in other words a finite number of identified, predictable solutions as held by the KSR decision), and thus for the two claimed actions, it would be obvious to try them during the scan and non-scan times, even though the most relevant prior art did not specifically disclose the non-scan limitation.  What makes such a rejection so sinister is that in applying the “obvious to try” rationale, the examiner is using the applicant’s recognition of how to divide conditions against the inventor (e.g. dividing the MRI scan into scan and non-scan), even though it is that recognition that may be one of the novel and non-obvious aspects of the invention.  In other words, while it may be obvious to try two actions (e.g. the first action and the second action) if those actions are known, it is not necessarily true that it would be obvious to break the universe of times to carry out those actions into two specific and different operations at specific times (e.g. “during” the scan and “after” the scan).

Fortunately, a recent non-precedential opinion may prove useful for providing guidance in overcoming such rejections.  The case involved Google LLC (“Google”) vs. At Home Bondholders’ Liquidating Trust (“At Home”). Briefly, Google stated that a number of claims of U.S. Patents 6,286,045 and 6,014,698 to At Home were unpatentable as obvious.  The claims at stake related to “non-blockable requests” issued by a user’s terminal over a network to retrieve banners commonly used for advertising purposes.  For the obviousness claim, Google relied solely on U.S. Patent 5,933,811 (“Angles”) as teaching non-blockable requests by virtue of Angles disclosing an advertisement request that references a common gateway interference (CGI) script.  According to Google, a person of ordinary skill in the art would understand that disclosing a request to execute a CGI script would also disclose a non-blockable request, because there are only two options – either the request is blockable or non-blockable  Specifically, at oral argument, Google characterized a “request” as a genus consisting of two species: blockable requests and non-blockable requests.  This looks much like the reasoning with obvious to try rejections where the Examiner cannot find a teaching of the specific times (before and after a scan), but alleges they are obvious to try because there are only two. The Court rejected this reasoning, as it should, noting that it was a “good try”   but ultimately fails.  

So, remember that absence of disclosure coupled with the obvious to try rationale should not be used as bootstrap by Examiner to nullify an inventor’s recognition of how to break something down into a limited number of options.  KSR requires that there be “a finite number of identified, predictable solutions.” If the inventor is the one who recognized how to identify the limited number of options, then that those options would not be “identified” by the state of the art as in KSR. 

Looking to the example of above, if the prior art does not recognize how to properly divide a genus (e.g. MRI scan procedure) into two specific species (e.g. scan and non-scan), then it cannot be rendered obvious to try, due to lack of disclosure.  Stated another way, it is only the inventors who have recognized how to properly divide the genus, and such a division of genus cannot be obvious to try if prior art does not recognize such a potential division of genus.

Time will tell if such an argument will be persuasive in overcoming obvious to try rejections, but we think such an approach is worth considering in the right situation.