By reading the Amicus Brief of IBM to the CAFC, it is pretty clear that the machine tranformation test which allows software patents and ban business method patents was invented by IBM lawyers:
IBM respectfully submits that the gravamen of that precedent, as informed by the constitutional objective of "promot[ing] the Progress of Science and useful Arts," is that a patentable "process" within the meaning of § 101 is one that involves a technological contribution — namely, a process that either (i) is tied to a particular machine or apparatus, or (ii) causes transformation or reduction of an article to a different state or thing, and in either instance produces technologically beneficial results. This test sets forth a reasonable and balanced standard for subject matter eligibility.
This test is now being interpreted by several US courts and the Board of Patent Appeals and Interferences (BPAI) as allowing software patents (read the article "Bilski at the BPAI - What a Mess (Part 1)" on 271patent blog):
Claim: (Ex Parte Buhan) A method for storing content encrypted by control words in a receiver/decoder unit having a local storage unit and being connected to a security unit, said control words as well as a necessary right for the access to the content being transmitted in entitlement messages that can be decrypted by system keys, the method comprising …
BPAI: We note a receiver/decoder unit having a local storage unit is mentioned in the preamble, which storage unit is embodied in the first step of storing the encrypted content. We also note in the preamble a security unit, which is embodied in the second step of storing the system keys. Both the local storage unit and the security unit constitute tangible, solid, real-world machines, the former exemplified by a magnetic hard disk, and the latter by a smart card (See Fig. 1). We find these elements sufficient for satisfying the “particular machine” prong of the Bilski machine or transformation test, and thus find the Examiner erred in rejecting these method claims.
The Bilski test to validate software patents is an invention of IBM. Now let's guess what the new USPTO president David Kappos will defend at the US Supreme Court against Bilski. Conflict of interests, no one?
See also the IAM article:
IBM welcomes today's en banc Federal Circuit decision in the In re Bilski case, as it excludes from patent eligibility business method inventions that are not tied to a machine or transformative of an article. The Federal Circuit's "machine-or-transformation" test was a primary component of the test that IBM advocated in our amicus brief.