In reality, though, the categories of "things you can do with a PC and software" and "things you can do with a peripheral" are not mutually exclusive.

Take VoIP, for example. We have VoIP software for use with a computer, webcam, and mic. Yet we also have standalone VoIP devices that perform substantially the same functions. These devices are basically stripped down computers.

So what of that? VoIP on the computer does not deserve patent protection, but the same technology on a stripped down device does?

I certainly appreciate your interpretation, but I would counter with this point:

Without getting into too much detail, the physical transformation inquiry only arises (at least in this context) when an invention is an "abstract idea." Method claims often fit this bill.

However, computer hardware is physical, tangible, and concrete--it is not merely an abstract idea. Thus the physical transformation inquiry is not apposite.

People seldom consider the implications of abolishing software patents.

Sure, there are a lot of good arguments against granting artificial monopolies on computer software--and many of them ARE good arguments. However, categorically denying patent protection to software creates some logical difficulties.

Most computer aficionados are familiar with the idea that software and hardware are logically equivalent. We CAN build specialized hardware to do what we would have our software do.

It would seem a bit anomalous, therefore, to allow a patent on specialized hardware that embodies precisely the same inventive character as its patent ineligible software counterpart.

So, in my opinion, the real issue we are seeking to resolve here is more subtle and obscure than we are admitting. Certainly, affording computer software a unique status as patent ineligible subject matter is not the most complete solution.

I think that's what the court was getting at in Bilski. They were searching for some kind of logical test rather than an unexplainable, static, and inflexible prohibition on a certain class of invention.

I'm not saying the court was right--I'm trying to shed a little more light on the playing field.

From a completely subjective point of view, the thrust of your argument makes intuitive sense--why should a method that is not patent eligible be able to skirt this result by reciting a general purpose machine for its execution?

Unfortunately, in the legal world, intuition must sometimes give way to legal formalisms.

In this case the formalisms are the statutory categories of invention. Processes (methods) constitute one distinct statutory category, and machines (apparatuses) another.

Getting back to your point, you are correct to say that, after Bilski, a method (process) claim that is otherwise not patent eligible may not become eligible by simply providing that the disclosed method is executed on a general purpose machine (computer).

HOWEVER, an apparatus claim is separate and distinct from a method claim that is tied to an apparatus.

So while it is true that tying a method claim to a general purpose computer is not sufficient to overcome Bilski's machine requirement, it remains true that the holding of Bilski DOES NOT APPLY TO APPARATUS CLAIMS.

As an example, consider the following:

Method claim: "A method of ... comprising the steps of ...."

Method claim tied to a general purpose computer: "A method of ... executed by a computer processor, comprising the steps of ...."

APPARATUS claim: "A computer readable storage medium containing instructions, when executed by a computer processor, operable to ...."

Unfortunately, this patent contains apparatus claims as well as method claims.

The Bilski decision was limited to method claims, and so it won't apply without a fairly liberal extension of that holding.