Side 1 Andrew Chin AndrewChin.com A “Means” is a Cause USPTO Roundtable on Quality of Software-Related Patents February 27, 2013

Side 2 Andrew Chin AndrewChin.com Category Mistake an ontological error in which things of one kind are presented as if they belonged to another See generally Andrew Chin, The Ontological Function of the Patent Document, 74 U. Pitt. L. Rev. __ (2013) (describing the patent system as a project to construct a coherent ontology of the “useful Arts”)

Side 3 Andrew Chin AndrewChin.com Category Mistake “Colorless green ideas sleep furiously.” — Noam Chomsky

Side 4 Andrew Chin AndrewChin.com In re Bilski (Rader, J., dissenting) “[A]n abstract claim would appear in a form that is not even susceptible to examination against prior art under the traditional tests for patentability. Thus this court would wish to ensure that the claim supplied some concrete, tangible technology for examination.”

Side 5 Andrew Chin AndrewChin.com Diamond v. Diehr “It is for the discovery or invention of some practical method or means of producing a beneficial result or effect, that a patent is granted….” A “means” is a cause. “Structure, material or acts” support a “means” only insofar as they participate in its causal processes.

Side 6 Andrew Chin AndrewChin.com These Are Not Causal Processes Mathematical theorems and equations Aristocrat and Dossel patents both lack structural support Geometric and kinematic properties Turing Machine algorithms

Side 7 Andrew Chin AndrewChin.com These Are Causal Processes Colliding billiard balls

Side 8 Andrew Chin AndrewChin.com These Are Causal Processes Colliding billiard balls The running instance of PowerPoint displaying this presentation

Side 9 Andrew Chin AndrewChin.com These Are Causal Processes “… schedulers, device drivers, file management systems, memory management systems, compilers, interpreters, interrupt handlers, caches, programmable firmware stores, error-correcting memory, wired and wireless network interfaces, network protocol handlers, systems, web browsers, and many more …” “… running instances of virtual machines …” See generally Aaron Sloman, What Cognitive Scientists Need to Know About Virtual Machines

Side 10 Andrew Chin AndrewChin.com In re Alappat Cited (even by Lemley) for proposition that the specification of a programmed computer can provide sufficient structure But the Federal Circuit’s § 112(6) construction did not take a programmed computer to be the supporting structure! See Mark Lemley, Software Patents and the Return of Functional Claiming

Side 11 Andrew Chin AndrewChin.com In re Alappat 15. A rasterizer … comprising: (a)[an arithmetic logic circuit …] for determining the vertical distance … (b)[an arithmetic logic circuit …] for determining the elevation … (c)[a pair of barrel shifters …] for normalizing … (d)[a read only memory …] for outputting …

Side 12 Andrew Chin AndrewChin.com Alappat’s Specification “[V]arious operations of rasterizer 40 … are timed by clock signals produced by a state machine in accordance with control data… One signal is a ‘pixel clock’ signal that is asserted to cause the rasterizer to receive each new vector list data element… This [ALU] value is stored in a register 76 on the next pixel clock cycle.” “[P]riority encoder 86 causes barrel shifter 84 to shift its input to the left by the number of bits required…” The 8-bit intensity data stored in register 90 addresses a read only memory (ROM) 92 and causes ROM 92 to read out a 4-bit intensity data value which is stored in a register 94 on the next pixel clock cycle.

Side 13 Andrew Chin AndrewChin.com The “Concrete Causation” Standard “The utility of a patentable invention must be amenable to explanation by a single causal account that specifies the resources brought into play by the invention’s use.” See Andrew Chin, Let’s Create a Concreteness Standard for Abstract Software Patents, Wired.com, 11/2012