CDs have to be in quanta of 75ths of a second
Oh yeah; I forgot about CD frames.
It only guarantees sample-accurate gapless with the headers, but the heuristics used on files without the headers are actually pretty good in practice.
Of course, again, unless I'm mistaken, the LAME headers, sadly, predate anyone's heuristics implementations.
other formats (Vorbis, FLAC, WMA) can encode arbitrary-length audio
That's great for the eight people who want to use those formats, own no players that can't support them, and have the proficiency to rip their CDs using them.
any added fail there is purely proprietary
True, and if you had a soup-to-nuts ripping and playback solution that didn't need any compatibility with any other system, or had very little restriction on storage space, there's no excuse for failing gaplessness. Otherwise, it's best-effort. (Of course, that effort should now include the use of LAME headers on MP3s.)
I've lost track of my point....
Oh, yeah. The expectation of fail in this space is derived directly from the mp3 spec. In general, I emphatically agree with your complaint about diminished expectations, but I think we should point fingers in the right direction: the mp3 spec and mediocrity-loving idiots.