OK, I still haven't read the transcript yet, but here goes.
I noticed the same 101/102/103 conflation problem as well. But, why should 101 be read so broadly anyway? I'm wary of letting the Federal Circuit determine what qualifies as for a government grant of monopoly considering their proclivity to lower the threshold for patent validity wherever possible. Obviousness provides a good example, as it's supposed to act as a limiting factor - but the CAFC defined "obvious" so narrowly that the Supreme Court got over it's distaste of patents just to slap the CAFC back a few steps.
And that's without getting into the fact that by narrowing 101, it allows the PTO to eliminate the patent on as it's written, rather then relying on the quality of the examiner's search for 102 to act as a bar, and waiting (possibly) until a finding of fact for 103 to kick in.
And of course, if the supreme court decides to roll back the scope of 101, the offended monied interests will know where they can reach Orin Hatch for yet another go at patent reform, this time with the urgency that only an irate special interest can bring to the table. That's the bottom line: Congress can always explicitly expand 101 patentability if the court gets it "wrong." The danger of reading it broadly is what you have now - 10 years of patents issued post-state street that maybe shouldnt have been.
Fun Fact: Sotomayor, being the only justice to have served as a district court judge, is therefore the only judge to have dealt with a patent suit up close and personally. I've found one instance of a Markman finding on her part: Intellectual Prop. Dev. v. UA-Columbia Cablevision, 1998 U.S. Dist. LEXIS 3901
(CAFC upholds her finding of non-infringement, overturns her finding of invalidity here: http://scholar.google.com/scholar_case?case=10580827877296089053