
The Berkheimer decision may favorably shift the landscape for software patents, with potential increases in allowance rates from historically low levels. While the ruling may seem straightforward, it introduces a significant change by allowing factual determinations on whether certain limitations are “well known.” This adjustment makes it more challenging for examiners to reject claims based on ineligibility and for defendants to seek early dismissal for failure to state a claim under Rule 12(b)(6).
Current State of Software Patent Protection
Software-based inventions currently experience a much lower allowance rate compared to other invention types, such as mechanical inventions. The overall allowance rate for all inventions generally fluctuates between 45% and 60%, while for software, it falls below 10%. This discrepancy suggests that software inventions are singled out and treated differently, often because of determinations on what is “well known.”
Whether this low allowance rate is beneficial or detrimental depends largely on one’s perspective of the patent system. For those who view the patent system as a tool for fostering innovation, the low rate for software patents is concerning. Conversely, those wary of patents disrupting businesses via so-called patent trolls may see the low rate as favorable.
This trend began in 2014 with the Bilski, Alice, and Mayo decisions, which raised questions about patent eligibility as a primary issue, overshadowing prior analyses that focused on novelty and nonobviousness. Following these cases, the USPTO’s patent eligibility framework increasingly led to rejections of software inventions, both in examination and in court.
Overview of Berkheimer v. HP
In Berkheimer v. HP, the Federal Circuit ruled that the second step of the Mayo/Alice test may include factual questions regarding whether a limitation is “well known.” This resulted in vacating a summary judgment that had found the claims ineligible under § 101. Though seemingly minor, the Berkheimer ruling could alleviate some negative effects of the current eligibility framework, potentially boosting software allowance rates.
The defendant sought an en banc review of the decision, which was ultimately denied. However, a dissenting opinion highlighted potential impacts on both examination and litigation processes concerning “well-known” limitations.
Reasoning Behind the Denial of En Banc Review
The majority cited several factors in the denial:
- Burden of Proof: Under 35 USC 282(a), the challenger bears the burden of proving that a claim is not patent eligible.
- Summary Judgment Standard: Civil Procedure Rule 56 mandates that summary judgment be denied if a genuine dispute of material fact exists.
- Inventive Concept: Step two of the Alice/Mayo framework includes evaluating whether an inventive concept is “well known” within the industry. In Berkheimer, there was a factual dispute over whether a specific claim limitation was indeed well known.
The Federal Circuit also noted that patent specifications can indicate whether a limitation is well known, routine, or conventional, often based on explicit admissions in the specification. This highlights the need for patent drafters to avoid language that implies limitations are well known or conventional, such as “known in the art.”
USPTO Examination Procedure Changes Based on Berkheimer
On April 19, 2018, the USPTO released a memorandum modifying examination procedures, especially around determining if an element is well known. Under the new guidance, examiners need specific evidence to conclude that an element is well known, routine, or conventional. Acceptable evidence includes:
- A statement in the specification acknowledging the element as well known.
- A court decision or publication supporting its well-known status.
- Official notice by the examiner (as referenced in MPEP 2144.03).
This change makes it more difficult for examiners to reject applications on eligibility grounds based on “well-known” limitations and harder for courts to dismiss cases on similar grounds.
Moving Forward
If Berkheimer indeed tips the balance for software patents, the primary focus will likely shift to the standard for proving that an element is well known.
Although this decision is a positive development for software inventors, previous cases such as Enfish v. Microsoft, Bascom v. AT&T, and Internet Patents Corp. v. Active Networks were also favorable but did not substantially increase software patent allowance rates. Time will tell if Berkheimer has a lasting impact.
Please reach out with any patent questions at (949) 433-0900, and feel free to share this article with colleagues. As a patent attorney in Orange County, I serve clients throughout Irvine, Los Angeles, San Diego, and surrounding cities.