Federal Circuit Deals Another Harsh Blow to Software-Focused Patent Applications

The Federal Circuit struck down two of IBM’s patents – the ‘789 and ‘389 patents – on the ground of patent ineligibility under 35 U.S.C. § 101, dealing another blow to software-based patents and applications.[1]

The ‘789 patent is directed to methods for coordinated geospatial, list-based and filter-based selection, while the ‘389 patent is directed to methods of displaying layered data on a spatially oriented display. The ‘789 patent discloses taking a viewing area on a digital map (think GPS) and display elements such as restaurants, points of interest, etc. within that viewing area. The viewing area could be customizable by the end user to define the applicable area, and the list view could be customizable to present user-selected elements. For example, if a person wanted to look up all restaurants within a five-mile radius of a selected viewing area, then the method would generate a list view within those defined parameters. In a similar manner, the ‘389 patent is directed to methods of displaying this layered data. The end user could select the objects to be displayed in a layered view, where each layer was defined by distinguishable display attributes. So, if a person wanted to look up restaurants and hotels within a selected area, then the method of the ‘389 patent would parse this data to display all hotels within a selected area in one distinguishable manner and display all restaurants within a selected area in another distinguishable manner.

IBM filed a patent infringement lawsuit against Zillow in 2019, alleging that Zillow infringed seven of IBM’s patents (including the ‘789 and ‘389 patents). The District Court found both the ‘789 and ‘389 patents were directed to abstract ideas, concluding that they “contained no inventive concept, and failed to recite patent eligible subject matter.”[2] IBM appealed this ruling to the Federal Circuit.

Before we review the Federal Circuit’s decision, it would be worthwhile to refresh ourselves on the standards of patent-eligibility: “Laws of nature, natural phenomena, and abstract ideas are not patentable.”[3] The Supreme Court has provided a two-step framework to determine whether a particular claim is patent-eligible:

At step one, we determine whether a patent claim is directed to an unpatentable law of nature, natural phenomena, or abstract idea.[4]
– “In cases involving software innovations, [the step-one] inquiry often turns on whether the claims focus on specific asserted improvements in computer capabilities or instead on a process or system that qualifies [as] an abstract idea for which computers are invoked merely as a tool.”[5]

If so, we proceed to step two and determine whether the claim nonetheless includes an “inventive concept” sufficient to “‘transform the nature of the claim’ into a patent-eligible application.”[6]

The District Court found, and the Federal Circuit upheld, that the ‘789 patent was directed to the abstract idea “of responding to a user’s selection of a portion of a displayed map by simultaneously updating the map and a co-displayed list of items on the map.”[7] The District Court reasoned that claim 8’s method “could be performed by hand, using a printed map and related list of items on the map, a transparent overlay, a wet-erase marker, a blank sheet of opaque paper, and a knife or scissors.”[8] The District Court explained that one could put the transparent overlay on the map, draw on it with the marker, and then block off the “unselected area” of the map and corresponding list items with the opaque paper.[9] To choose a different “selection area,” the user would erase the previous marking, remove the paper, and start over. The District Court noted that “alterations to hardcopy materials were made or auditioned in this manner” long before the invention of the computer, and thus concluded that “[t]he ’789 patent merely contemplates automation using a computer.”[10]

The analysis then turned to the question of whether the claims presented additional features that constituted an inventive concept, and were more than well-understood, routine, conventional activities. The Federal Circuit found the cited “synchronizing” and “user determined shape” limitations used functional language, at a high level of generality and divorced from any computer technology, to recite the claimed functions. The limitations, as reasoned by the Court, recited the abstract method without providing more.[11]

In a similar manner, the Federal Circuit found the ‘389 patent to be directed to an abstract idea of categorizing and displaying information, as well as altering the manner of display upon user demand. It explained that humans have long used nonspatial characteristics like shapes, colors, line patterns, and other visual markers to distinguish between visually represented items, which were generated by hand before computers, “and the methods disclosed in Claims 1 and 2 could be similarly performed using colored pencils and translucent paper; each sheet of paper would display a ‘layer’ within the meaning of the ’389 Patent, and the sheets could be arranged, rearranged, and perhaps redrawn as desired to highlight particular objects or groups of objects.”[12] The Federal Circuit reasoned that such claims did not improve the functioning of a computer, make it operate more efficiently, or solve any technical problems. Instead, it was a mere rearrangement of generic information to assist end users in processing information more quickly.

When turning to the question of whether the ‘389 patent presented additional features constituting an inventive concept, the Federal Circuit’s conclusion was more of the same: Any of the patent’s improved efficiency “comes not from an improvement in the computer but from applying the claimed abstract idea to a computer display.”[13]

So, what can we take away from this ruling? Patent practitioners tasked with drafting heavily software-based applications must take care to be as transparent as possible when describing the supposed invention. Ask yourself: how does my software provide improvements over similar, previously known programs? Once you have this answer, you must make sure to clearly delineate these improvements presented by your client’s software over these prior programs in your application.

More particularly, when drafting, don’t just describe software-level operations with a broad stroke. Else, you may be doomed to the same conclusions made by the Federal Circuit in IBM; that is, a person could perform this software method by hand, and your only improvement is to implement it on a graphical display. Be sure to present the software in as much detail as possible: How is the operation accomplished? Why is it accomplished this way? Are there any specific hardware features being implemented in executing this operation? What are the improvements?

The Federal Circuit’s decision gave us hints on how to properly draft software-based applications in its decision. Essentially, be sure your claims are supported by a detailed description that describes the abstract method and provides more. Had the IBM applications specified how their interfaces greatly improved user interactivity and improved over the hypothetical “you could draw these interactive maps by hand” scenario, then perhaps they could have survived the oft-feared 101 patent-eligibility analysis.

[1] IBM Corp. v. Zillow Group, Inc., No. 21-2350 (Fed. Cir. October 17, 2022).
[2] IBM Corp. v. Zillow Group, Inc., 549 F. Supp. 3d 1247, 1264, 1268 (W.D. Wash. 2021).
[3] Ass’n for Molecular Pathology v. Myriad Genetics, Inc., 569 U.S. 576, 589 (2013) (quoting Mayo Collaborative Servs. V. Prometheus Labs., Inc., 566 U.S. 66, 70 (2012)).
[4] Alice Corp. Pty. Ltd. v. CLS Bank Int’l, 573 U.S. 208, 217 (2014).
[5] TecSec, Inc. v. Adobe Inc., 978 F.3d 1278, 1293 (Fed. Cir. 2020).
[6] Id. (quoting Mayo, 566 U.S. at 72, 78).
[7] 549 F. Supp. 3d at 1266–67.
[8] Id. at 1267.
[9] Id.
[10] Id.
[11] Case No. 21-2350 at 8.
[12] Id. at 12.
[13] Id. at 16.