My 2 cents, separate, but related to this discussion
As always, all the different algorithms that can be applied to an audio file, become a nightmare in the real world. I do 4 to 6 backups a day for clients that are migrating data to another computer or simply getting a restore on the same computer. Some use iTunes, some don't... some are macs, some are PCs. Many people have a mixture of m4a, mp3, wma for their "commercial" music libraries, that are usually found via the usual suspects... limewire,bearshare, torrent, youtube convert, etc. some are purchased and protected, many are not.
There are very clear audio artifacts that can be heard on the playback of many of these files. Many have been converted from a lower bitrate to a higher one when protection was removed or many had a "poor original source" like a youtube video. Also, many may have been converted more that once, due to public stupidity and lack of education when organizing large libraries. The point is, my old-school ears "hear" the difference right away; I have heard both phasing and flanging issues in VBR files, High-end frequency decay issues on percussion instruments in files that have lower than a 256KB Bitrate; low end frequency "smearing" (bass notes get less defined and bleed into each-other) in WMA files.
So, it's funny when these clients download free music libraries of 10,000+ songs and wonder why they sound like shit. Guess you get what you pay for.
Now for my musician clients, and myself, I like the old and trusty MP3 (320KBps). These babies play on it all and have excellent fidelity. You can move them to external USB players, many Daws can convert them for manipulation if you are doing background tracks or DJ stuff. Also, many chip-based playback devices accept them, like Tascam SSR1 or units like Cortex HDC-500. Also, Android phones, iphones, ipads, ipods, old Macs, new Macs, old PCs and New all love 320KBps (non-VBR) files and fidelity is extremely close to original and is well-produced over a large PA.