Enfish, LLC v. Microsoft Corp. (Fed. Cir. May 12, 2016) (Before Moore, Taranto, and Hughes, J.) (Opinion for the court, Hughes, J.). Click Here for a copy of the opinion.
Enfish sued Microsoft for infringing two “self-referential” database patents, Patents 6,151,604 and 6,163,775. The district court held that all of Enfish’s patent claims were invalid as directed to patent-ineligible subject matter under § 101, and some claims were anticipated under § 102. The district court also held that Microsoft’s software did not infringe Enfish’s patents. Enfish appealed. The Federal Circuit reversed the district court’s holding under §101, vacated the district court’s holding that certain claims were anticipated, but affirmed the summary judgment findings regarding Microsoft’s non-infringement.
Traditional relational database models maintain each category of data in separate tables that have relationships with each other. For example, suppose there are three tables for three different types of data: a Document Table, a Person Table, and a Company Table. The ID number of an entry in the Company Table can be referenced by the Person Table to provide employer information, and an individual in the Person Table can be referenced by the Document Table to provide author information, and so on. In contrast, Enfish’s self-referential database stores all such data in a single table, and can generate new columns from information in rows in the table. This allows the user to modify the self-referential table by inputting new field values into rows that in turn generate new columns in the same table.
The Federal Circuit held that Enfish’s claims were patent-eligible under the two-step patent-eligibility test set forth in Alice Corp. v. CLS Bank Int’l. In the first step, the court determines whether the claims are directed to a patent-ineligible concept. If the claims are so directed, the second step looks to see whether any element or elements in the claim “transform the nature of the claim” into something patent eligible.
The Federal Circuit does not read Alice as broadly holding that “all improvements in computer-related technology are inherently abstract.” Instead, where the claims are directed to an improvement to computer functionality, they are not abstract under the first step of Alice, and thus no step-two analysis is necessary. Here, the Federal Circuit found that Enfish’s self-referential table was directed to a specific improvement in computer capabilities, unlike Alice, where the claimed technology only added a computer to a traditional business practice. For this reason, the Court held that Enfish’s claims were not abstract under the first step of Alice, and therefore did not warrant the application of step two.
The district court construed the claims very generally, describing them as the “concept of organizing information using tabular formats.” However, the Federal Circuit cautioned against this approach, saying that “describing the claims at such a high level of abstraction and untethered from the language of the claims all but ensures that the exceptions to §101 swallow the rule.” The Federal Circuit instead construed the claims as specifically directed to self-referential tables only, when viewed in light of the specification’s assertions that the claimed tables function much more effectively than traditional database models. By oversimplifying the claims, the district court largely ignored the benefits of the invention. Further, patent claims do not have to be tied by reference to physical components to be patent eligible, nor does an invention’s ability to run on a “general purpose computer” doom its claims. Ultimately, because Enfish’s claims were directed to a “specific implementation of a solution to a problem in the software arts,” the claims were patent-eligible.
The district court held that Enfish’s claims were anticipated by the “pivot table” feature of Microsoft Excel 5.0 under §102(b). This feature takes a first table of raw data, and generates a second, summarized table called a “pivot table.” Pivot tables can generate new columns in the table based on input from rows in the raw data, which could seem to anticipate the self-referential tables. However, the Court emphasized that Enfish’s claims do not generate a secondary table, which differentiates them from the pivot table feature. The pivot table and the raw data table are still two separate tables, regardless of their presence on the same spreadsheet. Thus, the Court held that the pivot table feature of Excel 5.0 did not anticipate Enfish’s claims.
Finally, the Federal Circuit affirmed the district court’s grant of summary judgment of no infringement. The district court construed claim 17 of Enfish’s ‘604 patent , which required “means for indexing data stored in [a] logical table,” as a three-part algorithm, including storing pointers to corresponding index entries in the text cells of the logical tables, and including in the index, pointers to the text cells. Under this construction, the district court found that Microsoft did not infringe Enfish’s claims. Enfish appealed the construction of this algorithm, asserting that all three parts were never meant to be required by the claims.
In challenging the claim construction, Enfish alleged that the pointers mentioned in Step 3 of the claims did not have to be “bi-directional,” meaning they point from both the index to the text cells and the text cells to the index. The Court examined the figures in Enfish’s specification and found that the claims did, in fact, require bi-directional pointers. Therefore, the district court properly construed Enfish’s claim. Microsoft used uni-directional pointers, not bi-directional. Additionally, Microsoft’s product did not store any actual text values in an index; rather, Microsoft stored references to the cells containing the text values. Due to these differences, the Federal Circuit affirmed the holding of the district court that Microsoft’s product did not infringe any of Enfish’s claims.
Parker Hancock and Puja Dave contributed to this summary.