The Bilski case can be summed up in one simple question: "Shall a method for calculations be patentable?"
This has an obvious impact on the software patent debate: If the answer is "yes", software will be patentable, too, if the answer is "no", it won't. But such an unconditional answer is unlikely - we'll get something like "yes, but only if..." So if the part after "only if" applies to Software, the court might very well *reaffirm* the patentability of software, even if many/most/all so-called "business method patents" are excluded.
In re Bilski, the Federal Circuit didn't answer the above question directly, but by introducing the so-called machine-or-transformation test (which already had been proposed by the US Government's ACB during Gottschalk v. Benson in 1972):
“A claimed process is surely patent-eligible under § 101 if: (1) it is tied to a particular machine or apparatus, or (2) it transforms a particular article into a different state or thing.”
What remains unclear is if a (=any) computer is a particular enough machine in the eyes of the courts and the patent offices to exclude software. Second, we don't know if the transformation of a signal is enough transformation to pass the test. Keep in mind that the two conditions are connected by "or" not by "and". These are mentioned explicitly in case law: For the "particular machine" question, see Ansonia Brass & Copper Co. v. Elec. Supply Co., 144 U.S. 11, 18 (1892), "if an old device or process be put to a new use, which is not analogous to the old one, and the adaptation of such process to the new use is of such a character as to require the exercise of inventive skill to produce it, such new use will not be denied the merit of patentability"; for the signal transformation e.g. in re Schrader, 22 F.3d 290 (Fed. Cir. 1994), also Judge Rader's dissent in the Bilski case.