If you agree that audio files on your computer do not degrade at all when transferred around your computer digitally and over time, then we have no disagreement.
I don't recall ever saying that. 
Files and data don't 'degrade' however they can become corrupted for various reasons. My concern isn't transfer within the computer (not so much) but transfer to the computer via something like an optical drive. http://thewelltemperedcomputer.com/KB/Ripping.htm I'm not concerned about the general rules but the exceptions. When you have engineers using figures like 95% and saying 'Ripping has nothing to do with real time, some rippers use this to read a damaged part up to 80 times to establish the most likely value' it's certainly more complex than saying it's just '1's and '0's no need to worry. The actual origin of my concern has more to do w/ transfer to and from devices via interfaces like USB, Toslink and Coax. When you started talking about the perfect binary world with satellites crashing I began thinking of digital devices that do permit operation despite jitter, errors and imperfect data transmission. Digital errors don't guarantee catastrophic failure in all cases. Do you disagree? My other comment had to do w/ using lossy as an archive.
To the topical point at hand. I have ripped audio files in the past that seemed to make it past built-in checks and verification on EAC. You could play the first few seconds of the song and then it would skip to the next everytime. It was actually 'Hell Freezes Over' by the Eagles. I actually had to rerip the disc a 3-4 times before the problem was fixed. I have no idea why since this was supposed to be 'impossible'. You'd think it would have aborted and/or ejected the disk. Why a bad file would get transferred and be able to play (to a point) is beyond me.
Speaking of degradation. How about the CD as media? Do you believe that does not degrade? I've certainly had DVDs and audio discs go bad after working w/ no apparent damage or scratches. It obviously has to do w/ the substrate on top of the CD getting altered. People worry about scratches to the acrylic w/o concern for the substrate material actually containing the pits. If you have unseen damage to just one single pit will the whole CD fail to run or just the song or part of it? Or will the computer interpolate the missing data and provide you with a bit perfect copy or an imperfect CD? Isn't this another case against digital perfection? Sometimes there are simply variables beyond the computers control that it must compensate for.
It's funny, hadn't really thought of CD's as 'old' technology yet. Sign of age perhaps. 
