there is long list of individual player and singer credits in addition to the orchestra. In my CD booklet these are explicitly listed as members of “La Chapelle Royale - Collegium Vocale”, which are also credited as orchestra and choir. In effect each is credited twice. Is this in accord with the MB style? Outside MB it is not usual practice to credit all individual members of an orchestra or choir, just any soloists. This is certainly a duplication?
This example also highlights a weakness of using relationships for this purpose. The artist entries for the orchestra and choir lists members - but these lists barely overlaps the explicitly credited members on this release. Presumably this is due to changing membership over time. The logical membership “relationship” of orchestra and player is not time or date-dependent (except as an annotation), so any attempt to keep up would result in an unwieldy list of artist entries of departed members.
On a quick search I didn’t see any previous discussion of this. Perhaps I missed some, but I would be interested to know what the “guru”s think.
In fact, full historical membership lists for orchestras would be fantastic data, even if it takes three pages to display it all
If we know who performed specifically on each track, I see no reason not to credit them. A different question is whether we really know this (are they all on all tracks, or did some members play on some but not others?).
The recording artists are set to the credited ensembles, which seems correct. As per the recent discussion in What to do? Trio name and individual members credited some of us like having the ensembles also linked as relationships even if the members are known, while some others don’t, but it’s in any case not wrong. Ideally we could say “X performed on Y as a member of Z” but that’s not something we can do at the moment.
Adding all the orchestra members at the track level seems both unnecessary and error prone, as you’d have to be sure they all perform on every track. Adding them as release relationships is good, and so is adding a member relationship between orchestra and individuals. There’s no “duplication” here, either. Having the right data in the right place, twice, isn’t always a duplication to optimize away.
Thanks for the comments. I wasn’t suggesting there was much wrong here, just asking what the concensus and guidelines were.
But on reflection, I think there is one piece of data missing in the current approach which does have implications for software downstream using the MB data. Currently there doesn’t appear to be any way of identifying whether artist/performers credited to a recording are members of the orchestra or not.
Perhaps MB needs an additional qualifier type for such relationships to add this missing data?
