[blahblahblah]

I mostly agree with the article in the context of distribution of a final mix. However, the article ignores one glaringly obvious reason to distribute in 24/192 format: to allow the listener to be a participant in the creative process, enabling better results for amateur musician listeners who want to sample or remix the audio or for DJs to get better results when altering the tempo for beat matching one track with another, etc. Of course, if you're going to do that, you might as well distribute in a multi-track format instead to maximize flexibility for the end user (Want to sing karaoke? Just turn off the lead vocal track for playback).

[nullc]

Yea, and and if the bandwidth/storage is at all an issue 6x size bloat from 24/192 pays for 6 separated tracks. (Actually more, because multitrack is more losslessly compressible while 24/192 is less). If you're already providing multitrack then 24 bit audio would make sense... otherwise, meh.