Edit #48817616 has got me thinking, since the ability to use artist credits on relationships was added a few years ago, we haven’t really addressed what’s appropriate to use and what’s not. I personally don’t like the idea of using only artists’ surnames as songwriting credits, for the reasons outlined in the linked edit note. Thoughts/suggestions?
My take would be that relationships with works should not be customized by artist credits, unless the artist credit is of a specific pseudonym. That pseudonym should be listed as an alias of that artist.
To me, there are three types of aliases that artists may use for AC. The first is the Snoop Dogg / Snoop Lion distinction, where they represent different personas of the same person used for different projects. The second is pen names that artists use for different purposes (name as lyricist and name as composer), but the artist does not purport to be different personas or characters when using different names. The third type is linguistic variations on the name (ex: name in different scripts, romanizations, special characters, etc).
AC should only be used for the second type of aliases. For the first type of aliases, a separate artist should be created (see style guidelines). We should not use the third type of aliases as AC because work relationships should be universal in that it is not restricted by how the work-related information is printed on releases. So I think the second should be the only alias type to be used as AC.
This could be framed thusly:
“as credited” vs “artist’s common name”.
I usually have a preference for " as credited" which I keep in check and instead follow community standards.
In this context though?
If I was using an encyclopedia I’d prefer to read that a song was written by “Jane Smith” or “Urich Smith” over just “Smith” cause it would help me stay orientated better.
And I would be thrown out of the library for howling, “But which *©¶℅∆ Smith?!!”, less often, cause sometimes Ulrich or Jane would be a songwriter I already know.
The problem with works is that a writer may be credited as “Smith” on one release, “B. Smith” on another and “Beelzebub Smith” on a third release. So the usefulness of artist credits is limited for works.
That doesn’t mean they are completely useless of course, for example when you add a work credited as “B. Smith” and you are pretty sure it must be Beelzebub Smith. You can use the existing artist with the artist credit to be not much more specific than your source, without creating a probably duplicate artist.
I wouldn’t change an established full name credit to a shortened version just to match the artist credit on one source though.
There is a fourth kind of alias as well: names that were changed without being seen as starting a new project (one I was working on a bit ago went from Ivory Keys to IvoryKeysADSR to Aivri, but they don’t draw any other line between what was released under the different aliases). I do think that would be another use for the relationship ACs as well, but it’s not relevant in the particular case that started the thread – I’m voting “no” for the reason @mfmeulenbelt gives.
I may be misreading you here. Please correct me if that is so.
If I am only pretty sure that that B. Smith is Belzeebub Smith then, lacking further evidence, the correct course is to create a new (possibly duplicate) artist and annotate.
I must be more specific than my hypothetical “artist = B. Smith” source.
This is why I typically check multiple sources if writers’ full names are not available via one source. I have done what you described in cases where such searches came up empty.
As I started my email-client today, I got a bit surprised about all the replies on my edits:wink: As I started entering one of my first detailed relationships/work edits, I thought that information would be on a “per release basis” as it is by per track artist credits (featuring -> “Artist as credited”). For beginning work editors the field naming “credited as” is a bit misleading in my opinion (and I found no help in documentation).
Now to the discussion:
I agree that works should have the full artist names, not shortened.
On the other hand it would be nice to have that shortened (or even faulty) “credited as” information on a per release basis (future feature?).
With this information kept identifying correct AC would be easier.
Although I remembered the following edit/discussion with faulty credits. Not really matching, but maybe it helps. Please take a look: https://musicbrainz.org/edit/40133285
The discussion so far has not considered the needs of classical music. According to the style guidelines, only the composer gets entered as a track artist. The “performers” are therefore entered as artist-recording relationships. This is therefore the only place where you can add the as-credited name for the performer (unless they are also shown as, for example, release artist). This can be particularly relevant where the MB name is in non-Latin script or where the performer has changed their name since the recording date.
This is specifically about songwriting (work-level) info. Usually, for classical music, I’ll follow the printed credits for recording-level data, while using the standard name for work-level data.
The credits on the work should probably always use the legal name (at the time of the writing).
I agree that it would be nice if they could be AC’d at the recording or release level, since those can vary. But then what do you do if a release doesn’t credit all authors, or credits the writing to [Band Name]?
This is certainly viable for artists who perform under their legal names (or variants thereupon; see below), but there are other cases where this isn’t workable without data loss:
- Artists who use stage names often have songwriting credits listed under their stage names.
- Some artists use pen names that are separate from their usual artist names.
Additionally, if this were to become a guideline there ought to be some clarification on what we mean by “legal name.” Do we prefer full names (e.g. “Bruce Howard Kulick” instead of Bruce Kulick)? Would shortened names (e.g. Tommy Thayer) or nicknames (e.g. “Fast” Eddie Clarke) be permitted?
In that case I normally do some digging elsewhere (e.g. a database) to figure out who all the contributors are.
For the first point: maybe IPI instead of legal name then, on the assumption that the various works databases use that to identify authors.
In many cases, while a release might list DJ Foo as writer, the relevant works database would list Mark Aurelius Foofty; I wiuld consider the latter more correct on the work’s ARs.
For the second point: my point is that if you want to preserve the credit as on the release, that would include not expanding band names to the probably-intended list of members. And that is not something that ACs would remedy nicely (unless you’d set the individual ones to Band Name, or [not credited] if not included in the releases’s credits).
I agree with @Zastai if you can find the artist in a works database use that as the reference.
See https://musicbrainz.org/doc/Work for a list of databases from rights societies where you can look up these writing credits.
There are cases where the same person has multiple ipi numbers (and artist credits) in these databases such as https://musicbrainz.org/artist/83886397-adf2-431a-b841-dc4af744a6cc where there are 4 ipi numbers and incorrect spelling of his name in these databases.
The whole problem is people using what’s printed on a release.
THESE ARE NOT RELEASE BASED CREDITS.
The don’t link to any specific releases, they link to multiple recordings, which in turn are linked to multiple releases, so this shouldn’t even be an argument at all.
Work credits may often be consistent among official releases.
If it is not the case then ok to choose best option.
That’s bringing it to the point.
Exactly what I expected from my former release track artist credited as edits
My interpretation of drsaunde is that for Works we are better using the more authorative sources we have available (eg encyclopedia’s, scholarly works, rights databases) to determine Work artist names. As opposed to seeing ourselves confined to what is on releases.
And yes, often times releases will be the most authorative sources we have. But if we have conflicting data then we are better going with the more authorative.