Steve,
Streaming data is streaming data, whether it's audio files or instrumentation data. It's either bit perfect and time correct or it's not. Yes, jitter is problematic but, as I'm sure that you would agree, not nearly as much with USB2 as it was with USB1. I'm sure that my next comment will further endear me to the peanut gallery, but frankly, I will say that I believe that most audiophiles can't discerne jitter (w/ USB2), and wouldn't know what it sounded like even if if they could. What's the worst jitter you have measured for USB2 gear that was in spec? Probably not more than 125uS or so, and I don't know anyone who can detect that. I believe that most of the sound effects that are erroneously identified as jitter are in fact the result of elasticity buffer stack overflows.
Re: my comment for Gordon Rankin - I was not making a wholesale endorsement of Gordon or his products. I have never met the man, but I'm sure that he (like you) is a smart guy. I was simply recommending his whitepapers to another poster as a good primer on USB, since there seems to so much misinformation bandied about on this forum.