I am adding my vote for this as well
I have use for it, and not only for classical music (though I admit I am quite an ID3 nazi
), it would tremendously useful if I could search by any jazz musicians, even on the tracks where they do not lead (think of most poor drummers and bass players...), or any other artist for that matter. It is possible to crowd the artist tag and leave album artist for the main band but it is clearly sub-par for esthetic purposes.
-A few examples:
If I want to look for all the tracks where Eric Clapton appears, I have:
Eric Clapton as the main artist
Various Bands: Derek & The Dominos, Blind Faith...
Individual tracks like While My Guitar Gently Weeps
-Another example: Miles Davis second great Quintet: Miles Davis, Herbie Hancock, Wayne Shorter, Ron Carter, Tony Williams.
Most of them also recorded as individual artists. I'd like to be able to list the relevant Davis albums, when I select Herbie Hancock for example, without having to write all the names in the artist field.
-"Feel Good Inc. ft. De La Soul" from Gorillaz could be tagged as follows:
Artist: Gorillaz / De La Soul
Album Artist: Gorillaz
"TPE3": Damon Albarn / Simon Tong / Truguoy / Posdnuos / ...
This way I can also filter or create a smart playlist based on the individuals, without polluting the display of info (the song is still by Gorillaz featuring De La Soul)
-And obviously, the case stands for classical music where you can add additional orchestra/conductor information
The way I see it, the TPE3 tag can be "sort of" hidden, it only needs to be indexed for the filters and smart playlists, and to be included in the queries for artist and album artist.
I also support the composer proposition, where I think a structure composer/"album"/songs is the most flexible (it can work for both classical and non classical music).
And, lastly, I lend my voice for "Original Date" which is great if you want to make Era based playlist, but does not work well with compilations albums.
If somebody wants to add Publisher, my files are ready, but that's not on the top of my list
On a constructive note, I understand some of these feature may not interest more than 10 users, and as a consequence, maybe a fraction of a developper. If somebody like me was to step-in, would such changes to the database (which is never a topic to consider lightly) be ever considered? Or would they be rejected based on the fact that they may be too intrusive for such secondary use cases? (I could understand this position, honestly...)