Thursday, April 30, 2015

PTAB reverses indefiniteness rejection of claim using trademark Java


Takeaway: The Examiner rejected as indefinite the phrase "parsing the at least one Java annotation" on the grounds that Java was a trademark. The Examiner referred to MPEP 2173.05(u) , which directed the Examiner to reject a claim as indefinite when a trademark was used to identify a product. The Applicant appealed and argued that the relevant inquiry was whether a Person of Ordinary Skill In The Art, considering the term Java in context, could "discern the metes and bounds of the claimed subject matter." According to the Applicant, the term was not indefinite because Java was understood by a POSITA to refer to "a popular programming language that was originally developed by Sun Microsystems in the early 1990s and released in 1995.” The Board agreed with the Applicant and reversed the indefiniteness rejection.  (Ex parte Nikolov, PTAB 2015.)

Details:

Ex parte Nikolov
Appeal 20130000274; Appl. No. 12/314,706;  Tech. Center 2100
Decided:  March 25, 2015

The application on appeal was directed to web software that manipulated objects in a Service-Oriented Architecture (SoA) registry. The Specification acknowledged the existence of various application programming interfaces (APIs) that allowed programmatic access to an SoA registry. The same section mentioned one of these APIs: "the Java API for XML registries (JAXR). "

The Examiner rejected independent claim 1 as indefinite because the "trademark / trade name 'Java' [was used] to identify a programming language." The term "Java" appeared in the preamble ("accessible through a Java API for XML registries (JAXR)" and in the body ("parsing the at least one Java annotation").

The Examiner cited the (non-precedential) Board decision Ex parte Simpson, 218 USPQ 1020 (Bd. App. 1982). According to the Examiner, Simpson meant that "when a trademark or trade name is used in a claim as a limitation to identify or describe a particular material or product, the claim does not comply with the requirements of 35 U.S.C. 112, second paragraph.”

 The Applicant traversed the indefiniteness rejection, and made five separate arguments:
  • Java identifies the product (the programming language) rather than the source of the product
  • the goods associated with the Java mark do not include "programming language"
  • the mark in Ex parte Simpson was indefinite since it did not adequately convey properties of the material, where "here there is no such uncertainty regarding what 'Java' conveys"
  • it's "illogical" to treat Java and C++ differently in a claim simply because the former is trademarked and the latter is not
  • thousands of issued patents include Java in the claims
In a Final Office Action, the Examiner noted that MPEP 2173.05(u) governed the use of trademarks and trade names in a claim: "if a trademark or trade name is used in a claim as a limitation to identify or describe a particular material or product, the claim does not comply with the requirements of 35 U.S.C. 112, second paragraph.” Responding to the Applicant's mention of C++, the Examiner indicated that an amendment to C or C++ would result in "similar 112 2nd rejections of indefiniteness regarding the use of trademarks or trade names. Responding to another Applicant points, the Examiner considered the "thousands of issued patents with Java in the claims" to be an "interesting statistic," but noted that applications are examined on a case by case basis. The Examiner therefore invited the Applicant to analyze some of these patents and to "present any common facts that trigger an exception to the rules of trademarks in claims," then explain "why this case similarly triggers an exception."

In the same Final Office Action, the Examiner discussed Ex parte Simpson in further detail:
     [Simpson] specifically provides that questions regarding a well-known meaning / satisfactory definition in the literature are questions that resolve around a 35 USC 112 1st rejection for sufficiency of written description for a "name used in trade" ... "We have considered MPEP(v) and note that all address the issue of enablement of a written description under 35 USC, first paragraph."
     Similar to Ex parte Simpson, no enablement rejection has been made. Similar to the Ex parte Simpson court, the Examiner declines to extend the enablement prong of the test of proper use of a trade name to cover indefiniteness of a registered trademark.
On Appeal, the Applicant argued that the use of Java in the claim did not render the claims indefinite per se. The inquiry for indefiniteness was whether a person of ordinary skill in the art, considering the term Java in context, could "discern the metes and bounds of the claimed subject matter." The Applicant concluded that “a person of ordinary skill in the art would understand what is meant by ‘Java annotation’ and . . . ‘Java API for XML registries’”  because Java was "a popular programming language that was originally developed by Sun Microsystems in the early 1990s and released in 1995.”

The Board agreed with the Applicant and reversed the indefiniteness rejection. The Board found this scenario "closely analogous" to a different non-precedential PTAB decision, Ex Parte O’Farrell (PTAB 2014). The Board explained as follows:
The O’Farrell panel held that the recitation of the trademark “Java” was not indefinite because “ 'Java' is a well-known and widely-used term identifying a particular programming language.” Id. Similarly, we find one of ordinary skill in the relevant art would have understood “Java” in the context of the “Java annotation” and “Java API for XML registries” to refer to the programming language known by that name at the time of Appellants’ invention.
My two cents: There's no doubt in my mind that Java is understood by a POSITA. And since that's the ultimate question when resolving indefiniteness, I say the Board got it right. The name of a programming language is fundamentally different than the name of a chemical composition. Using the trademark of a chemical composition is a problem because the nature of the composition could changes, so you don't know if you infringe or not. On the other hand, as long as a newer version of Java includes the annotation feature, a POSITA still understands "Java annotation" and is thus put on notice of infringement. And if the annotation feature is removed in a future version of Java, yet a patentee asserted against an infringer using an annotation-less Java, well, the claim could be found indefinite at that later time. 

Some of the Applicant arguments made during prosecution dug into trademark specifics (product vs source of product, associated goods), and I give some points to the Applicant for getting creative here. On the other hand, the "other patents use this language" argument is weak, and I didn't find the Java vs. C++ consistency argument to be compelling either. I think both the Applicant and the Examiner got derailed with the lengthy discussion of the MPEP and it's use of Ex parte Simpson. Responding to every single point made by the Examiner is not always the best strategy. The Appeal Brief would have been stronger if the Applicant had demonstrated that a POSITA understood Java, briefly distinguished Simpson, and stopped there.

The Board's finding that the meaning of Java was tied to the time of the invention highlights an interesting issue, because programming languages do change over time. With literal claim scope limited to Java at the time of filing, it would seem Doctrine of Equivalents is needed to expand to future versions of the language. But of course all claim terms are interpreted as of the time of invention, which makes the Board's finding just a reminder of something we should already know.

The PTAB decision mentioned in the MPEP and relied on by the Examiner, Ex parte Simpson, doesn't appear to be freely available online. The 1983 decision date is far too old for the PTAB's own decision database, and even too old for the BPAI decisions archive maintained by Franklin Pierce School of Law