Although I discuss patent issues generally, I hope to give special attention to the broad policy and philosophical issues arising under § 101 patent eligibility. Last Friday’s Federal Circuit decision relating to patent U.S. 7,702,682 provides a great example of how cases and the policy discussion generally often ignore the actual patents at issue. Evolutionary Intelligence LLC v. Sprint Nextel Corp. (Fed. Cir., Feb. 17, 2017).
The first thing to remember is that the scope of the invention is defined by the claims. Thus, the question to be asked when determining patent eligibility in the Mayo-Alice test is”what are the claims directed to?”. Unfortunately, we are often treated to a discussion of the invention or patent that makes no reference to the claims, and instead, relies principally on generic disclosure elsewhere in the patent.
The second thing to remember, although it is not important to this case, is that what “directed to” means is still an open question. Hence, the need for proposals that reaffirm the importance of claims being treated as a whole.
In this case, the only description of the invention was that it is “directed to systems and methods for allowing computers to process data that are dynamically modified based upon external-to-the-device information, such as location and time. See, e.g., ’536 patent Abstract.” As the quotation makes clear, only the abstract is used to characterize the invention. A hypothetical claim based on that abstract, “a computer implemented method comprising processing dynamically modified data based upon external-to-the-device information, such as location and time” would quickly and justifiably be invalidated.
The claims in the ‘682 patent are more detailed and specific than the abstract. For example, claim 1 reads:
1. A computer-implemented method comprising:
receiving a search query;
searching, using the computer, first container registers encapsulated and logically defined in a plurality of containers to identify identified containers responsive to the search query, the container registers having defined therein data comprising historical data associated with interactions of the identified containers with other containers from the plurality of containers, wherein searching the first container registers comprises searching the historical data; encapsulating the identified containers in a new container; updating second container registers of the identified containers with data associated with interactions of the identified containers with the new container; and
providing a list characterizing the identified containers.
I was a neuroscientist in my technical life, not a software engineer, so I won’t hazard an independent opinion of what was routine or conventional for the proper application of the Mayo-Alice test. It suffices to recognize that the claim is directed particular steps of running a software program, and it is those steps, as a whole, that should have been analyzed for abstractness. You may have a different opinion than what I have expressed about whether software should be patent eligible in the first place, but we should all be able to agree that it is the claims we should be analyzing, and not isolated statements in the abstract or specification.