...Not even if the algorithms are run on some “general purpose computer” (whatever that means!)

Summary: Software patents and lawyers who depend on them are running out of time and running low on spin/cherry-picking/excuses

YESTERDAY we wrote about video streaming patents and a notorious podcasting patent getting invalidated, with admission from the USPTO that it granted in error (prior art, not just Alice). Some software patenting proponents caught up with it and posted a link to the full decision: "Dist. Of DE Kills 2 Video Streaming Patents with Alice/ 101: http://assets.law360news.com/0824000/824998/04313306922.pdf …"

The word "kills" is intended to give this a negative connotation, just like "attacks" and "death squad" (same person uses these words a lot).

This wasn't a CAFC decision and that's actually noteworthy because nowadays the Court of Appeals for the Federal Circuit (CAFC) is a lot more likely to invalidate/reject software patents based on Alice. Speaking of CAFC, a new article by Dennis Crouch speaks of scope limitation induced by “consisting of” inside patents. To quote:

Although our attention over the past few years has focused more on eligibility and definiteness issues, claim construction continues to befuddle the courts. In MultiLayer Stretch Cling Film v. Berry Plastics (Fed. Cir. 2016), the Federal Circuit has offered a split interpretation of the patent’s Markush group claim element.

The asserted cling-film claim requires “five identifiable inner layers with each layer being selected from the group consisting of LLDPE, VLDPE, ULDPE, and mLLDPE resins.”

So here we have another example of tightening patent scope, other than what's related to Alice.

Speaking of Alice and CAFC, Troutman Sanders LLP wrote in the software patenting proponents' site earlier today, noting that according to CAFC, "Software means plus function claims Indefinite for failure to disclose algorithm," or in other words software patents are not "function [if you] then disclose a “general purpose computer” as the structure performing that function."

This is noteworthy because efforts to pass/sneak in software patents in India, Europe and New Zealand have relied on attempts to ascribe physical properties to software by stating in the patent application that it runs on some machine (typically just some ordinary off-the-shelf PC or server). To quote the funnier part from the above: "The Court also affirmed that the this means-plus-function term was indefinite. In the case of computer-implemented functions, the specification must disclose an algorithm for performing the claimed function. The patents-in-suit did not disclose an operative algorithm for the claimed “symbol generator.” The function of generating symbols must be performed by some component of the patents, but the patents did not describe this component."

So the patent/s did not even describe an algorithm, let alone pseudo-code!

It definitely looks as though patent lawyers are growing nervous, seeing year after year how more and more abstract patents get thrown down the toilet/wastebasket, essentially devaluing the remainder (those not tested in the court or in PTAB, which we shall write about next). ⬆