RFC-335. Forbid arrange on non classical works

classic Classic list List threaded Threaded
11 messages Options
Reply | Threaded
Open this post in threaded view
|

RFC-335. Forbid arrange on non classical works

jesus2099
Arrange on non classical works leads to systematic wrong inheritances for several later performance recording links like covers, live versions, alternate versions, medleys and any versions you might encounter — some could be right but some could be wrong → all get the wrong or right inheritance from the works anyway, without checking.

Discussion was in http://musicbrainz.1054305.n4.nabble.com/Arrange-on-works-td3709045i40.html

PROPOSAL STARTS HERE ← NEW !

 ☞ For this reason, forbid arrange relationship on non classical works by adding a text in the “Create relationship” page.

http://musicbrainz.org/edit/relationship/create?type0=work&type1=artist&entity0=7ab765cf-2b81-3805-8c31-c3698576b92b&entity1=aab5c954-cabe-432e-899e-1c4f99757327&returnto=http%3A%2F%2Fmusicbrainz.org%2Fwork%2F7ab765cf-2b81-3805-8c31-c3698576b92b%23relate_to

Current text when you select arrange on work is :
« Indicates the artist who arranged a tune into a form suitable for performance. »

Future text when you select arrange on work would be something like (to be written or reviewed by native Englishman) :

PROPOSAL IMPLEMENTATION HERE ↓ WOW ! (thanks to SwissChris)

☞ « Indicates the artist who arranged a tune into a form suitable for performance.
Don’t use this on non classical works — unless you can prove that the same specific arrangement is used by at least two different recordings (by different performers) — link to recordings instead. »
 PATATE12   jesus2099   GOLD MASTER KING   FAKE E-MAIL ADDRESS 
Reply | Threaded
Open this post in threaded view
|

Re: RFC-335. Forbid arrange on non classical works

Ryan Torchia


On Wed, Sep 14, 2011 at 1:57 AM, jesus2099 <[hidden email]> wrote:
Arrange on non classical works leads to systematic wrong inheritances for
several later performance recording links like covers, live versions,
alternate versions, medleys and any versions you might encounter — some
could be right but some could be wrong → all get the wrong or right
inheritance from the works anyway, without checking.


The problem lies with how the AR is designed in MB; the solution should be to fix MB, not to try to skirt the problem with inconsistent practices and contentious guidelines.  Here are some problems with this proposal:
1) Somebody adding an arrangement credit from a release has no idea whether or not that arrangement has been recorded elsewhere, or whether that arrangement will be recorded again in the future.  They also may not know whether it's the same arrangement on a different recording if uncredited, or if it's a different arrangement by the same person on a subsequent recording. 
   Take "The Wall" for example: Michael Kamen is credited for arrangements on the original album, for the film (although probably not on the dozen or so bootlegs taken from the film soundtrack), and orchestrations (and conducting) for the 1990 live album, but not for the 1981-2 archival live album (whose credits are fairly sparse,though it's likely that his arrangements were used at least in some places).  We know these are different recordings; we have no idea if Kamen reorchestrated the parts for the movie or 1990 live show, or if his arrangements were used in the 1981-2 tours, or Waters' recent (ongoing?) Wall tour (and on the numerous bootlegs taken from it).
2) The division of Classical vs. everything else is really kind of arbitrary.  For one thing, I'd hate to leave it up to editors to fight about whether a given work is "Classical" or not.  For another, most pre-WWII non-Classical genres are also frequently arranged, especially jazz, traditional pop, showtunes, marches, dances, etc.

However we do it, arrangement credits should be at the same level regardless of what kind of music it is or how many time it's been recorded.  Until we have something in place that accommodates this, I think it'd be better just to stick all arrangement credits at the Recording level.  It'd be pretty easy to join the matching arrangements later when we have a mechanism in place for that.

--Torc.

_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|

Re: RFC-335. Forbid arrange on non classical works

jesus2099
1) Is fixed by this RFC as when you don’t know you just won’t attach bad arrange links on recording and there won’t be any bad inheritance coming from the work. Here no info → no link. That’s how it should be.

2) In the previous discussion, needs of arrange at work level was for classical people. This is why I don’t think it’s good to forbid arrange-work for classical. It will be the most generic case and there will be some exceptions in non classical but, as said in the RFC, Those who recognise the need will use it and prove it, voters will judge. As for all other edits.

If your more drastic proposal of moving all links from works to recordings is RFCed I’ll be for it, but some people don’t want that.

Me.


Ryan Torchia wrote
On Wed, Sep 14, 2011 at 1:57 AM, jesus2099 <[hidden email]>wrote:

> Arrange on non classical works leads to systematic wrong inheritances for
> several later performance recording links like covers, live versions,
> alternate versions, medleys and any versions you might encounter — some
> could be right but some could be wrong → all get the wrong or right
> inheritance from the works anyway, without checking.
>
>
The problem lies with how the AR is designed in MB; the solution should be
to fix MB, not to try to skirt the problem with inconsistent practices and
contentious guidelines.  Here are some problems with this proposal:
1) Somebody adding an arrangement credit from a release has no idea whether
or not that arrangement has been recorded elsewhere, or whether that
arrangement will be recorded again in the future.  They also may not know
whether it's the same arrangement on a different recording if uncredited, or
if it's a different arrangement by the same person on a subsequent
recording.
   Take "The Wall" for example: Michael Kamen is credited for arrangements
on the original album, for the film (although probably not on the dozen or
so bootlegs taken from the film soundtrack), and orchestrations (and
conducting) for the 1990 live album, but not for the 1981-2 archival live
album (whose credits are fairly sparse,though it's likely that his
arrangements were used at least in some places).  We know these are
different recordings; we have no idea if Kamen reorchestrated the parts for
the movie or 1990 live show, or if his arrangements were used in the 1981-2
tours, or Waters' recent (ongoing?) Wall tour (and on the numerous bootlegs
taken from it).
2) The division of Classical vs. everything else is really kind of
arbitrary.  For one thing, I'd hate to leave it up to editors to fight about
whether a given work is "Classical" or not.  For another, most pre-WWII
non-Classical genres are also frequently arranged, especially jazz,
traditional pop, showtunes, marches, dances, etc.

However we do it, arrangement credits should be at the same level regardless
of what kind of music it is or how many time it's been recorded.  Until we
have something in place that accommodates this, I think it'd be better just
to stick all arrangement credits at the Recording level.  It'd be pretty
easy to join the matching arrangements later when we have a mechanism in
place for that.

--Torc.

_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
 PATATE12   jesus2099   GOLD MASTER KING   FAKE E-MAIL ADDRESS 
Reply | Threaded
Open this post in threaded view
|

Re: RFC-335. Forbid arrange on non classical works

Alex Mauer
In reply to this post by Ryan Torchia
On 09/16/2011 01:31 AM, Ryan Torchia wrote:

> The problem lies with how the AR is designed in MB; the solution should
> be to fix MB, not to try to skirt the problem with inconsistent
> practices and contentious guidelines.  Here are some problems with this
> proposal:
> 1) Somebody adding an arrangement credit from a release has no idea
> whether or not that arrangement has been recorded elsewhere, or whether
> that arrangement will be recorded again in the future.
> 2) The division of Classical vs. everything else is really kind of
> arbitrary.
>
> However we do it, arrangement credits should be at the same level
> regardless of what kind of music it is or how many time it's been
> recorded.  Until we have something in place that accommodates this, I
> think it'd be better just to stick all arrangement credits at the
> Recording level.

+1 to all this


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|

Re: RFC-335. Forbid arrange on non classical works

swisschris
I think we had reached a consensus in a previous thread [1] that, as a general rule, arrangement (and orchestration!) credits should  be linked to recordings, not to works, for every kind of music (not only for non-classical).

There are known cases (particularly in score-based classical), where specific arrangements or orchestrations have become works in their own right, being recorded by various performers over and over. Several examples have been quoted before [1]. This is not necessarily limited to classical: one could also think of arrangements of some popular tunes for brass bands, or of parody covers of a given performance by a given artist, re-using the specific arrangements…

For such cases – and only for such cases – an "arrange" AR would still be needed at work level. A new work based on arranger or orchestrator credits should not be allowed to be created unless one can prove that there are at least two distinct recordings by two different performers of this specific arrangement/orchestration.

With such a guideline there is no need to make an arbitrary distinction between classical and "all the rest".

[1] http://musicbrainz.1054305.n4.nabble.com/Arrange-on-works-td3709045i20.html

On Sat, Sep 17, 2011 at 6:06 PM, Alex Mauer <[hidden email]> wrote:
On 09/16/2011 01:31 AM, Ryan Torchia wrote:

> The problem lies with how the AR is designed in MB; the solution should
> be to fix MB, not to try to skirt the problem with inconsistent
> practices and contentious guidelines.  Here are some problems with this
> proposal:
> 1) Somebody adding an arrangement credit from a release has no idea
> whether or not that arrangement has been recorded elsewhere, or whether
> that arrangement will be recorded again in the future.

Where's the problem?  If he does not know, he'll simply use the generic rule and add the arranger and/or orchestrator at recording level. If someone knows better - or finds out later – the new work can always be created then ;-)

> 2) The division of Classical vs. everything else is really kind of
> arbitrary.

Agreed. See above. 

>
> However we do it, arrangement credits should be at the same level
> regardless of what kind of music it is or how many time it's been
> recorded.  Until we have something in place that accommodates this, I
> think it'd be better just to stick all arrangement credits at the
> Recording level.

+1 to all this


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|

Re: RFC-335. Forbid arrange on non classical works

Nicolás Tamargo de Eguren
Anyone feels like actually writing + proposing this?

On Sat, Sep 17, 2011 at 8:27 PM, SwissChris <[hidden email]> wrote:

> I think we had reached a consensus in a previous thread [1] that, as a
> general rule, arrangement (and orchestration!) credits should  be linked to
> recordings, not to works, for every kind of music (not only for
> non-classical).
> There are known cases (particularly in score-based classical), where
> specific arrangements or orchestrations have become works in their own
> right, being recorded by various performers over and over. Several examples
> have been quoted before [1]. This is not necessarily limited to classical:
> one could also think of arrangements of some popular tunes for brass bands,
> or of parody covers of a given performance by a given artist, re-using the
> specific arrangements…
> For such cases – and only for such cases – an "arrange" AR would still be
> needed at work level. A new work based on arranger or orchestrator credits
> should not be allowed to be created unless one can prove that there are at
> least two distinct recordings by two different performers of this specific
> arrangement/orchestration.
> With such a guideline there is no need to make an arbitrary distinction
> between classical and "all the rest".
> [1] http://musicbrainz.1054305.n4.nabble.com/Arrange-on-works-td3709045i20.html
>
> On Sat, Sep 17, 2011 at 6:06 PM, Alex Mauer <[hidden email]> wrote:
>>
>> On 09/16/2011 01:31 AM, Ryan Torchia wrote:
>>
>> > The problem lies with how the AR is designed in MB; the solution should
>> > be to fix MB, not to try to skirt the problem with inconsistent
>> > practices and contentious guidelines.  Here are some problems with this
>> > proposal:
>> > 1) Somebody adding an arrangement credit from a release has no idea
>> > whether or not that arrangement has been recorded elsewhere, or whether
>> > that arrangement will be recorded again in the future.
>
> Where's the problem?  If he does not know, he'll simply use the generic rule
> and add the arranger and/or orchestrator at recording level. If someone
> knows better - or finds out later – the new work can always be created then
> ;-)
>>
>> > 2) The division of Classical vs. everything else is really kind of
>> > arbitrary.
>
> Agreed. See above.
>>
>> >
>> > However we do it, arrangement credits should be at the same level
>> > regardless of what kind of music it is or how many time it's been
>> > recorded.  Until we have something in place that accommodates this, I
>> > think it'd be better just to stick all arrangement credits at the
>> > Recording level.
>>
>> +1 to all this
>>
>>
>> _______________________________________________
>> MusicBrainz-style mailing list
>> [hidden email]
>> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>
>
> _______________________________________________
> MusicBrainz-style mailing list
> [hidden email]
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>



--
Nicolás Tamargo de Eguren

_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|

Re: RFC-335. Forbid arrange on non classical works

jesus2099
In reply to this post by swisschris
So is proposed this swisschris’ new text for the docs instead :

« Arrangement (and orchestration!) credits should  be linked to recordings, not to works, for every kind of music (not only for non-classical).

There are known cases (particularly in score-based classical), where specific arrangements or orchestrations have become works in their own right, being recorded by various performers over and over. { here we’ll paste several examples as quoted before [1] in a bullet list }. This is not necessarily limited to classical: one could also think of arrangements of some popular tunes for brass bands, or of parody covers of a given performance by a given artist, re-using the specific arrangements…

For such cases – and only for such cases – an "arrange" AR would still be needed at work level. A new work based on arranger or orchestrator credits should not be allowed to be created unless one can prove that there are at least two distinct recordings by two different performers of this specific arrangement/orchestration. »

The new RFC-335 delay re-starts from today.
Feel free to modify it a little bit.
 PATATE12   jesus2099   GOLD MASTER KING   FAKE E-MAIL ADDRESS 
Reply | Threaded
Open this post in threaded view
|

Re: RFC-335. Forbid arrange on non classical works

Nikki-3
In reply to this post by jesus2099
Where is the new version of the arranger wiki page?

Nikki

jesus2099 wrote:

> Arrange on non classical works leads to systematic wrong inheritances for
> several later performance recording links like covers, live versions,
> alternate versions, medleys and any versions you might encounter — some
> could be right but some could be wrong → all get the wrong or right
> inheritance from the works anyway, without checking.
>
> Discussion was in
> http://musicbrainz.1054305.n4.nabble.com/Arrange-on-works-td3709045i40.html
>
> PROPOSAL STARTS HERE ← NEW !
>
>  ☞ For this reason, forbid arrange relationship on non classical works by
> adding a text in the “Create relationship” page.
>
> http://musicbrainz.org/edit/relationship/create?type0=work&type1=artist&entity0=7ab765cf-2b81-3805-8c31-c3698576b92b&entity1=aab5c954-cabe-432e-899e-1c4f99757327&returnto=http%3A%2F%2Fmusicbrainz.org%2Fwork%2F7ab765cf-2b81-3805-8c31-c3698576b92b%23relate_to
>
> Current text when you select arrange on work is :
> « Indicates the artist who arranged a tune into a form suitable for
> performance. »
>
> Future text when you select arrange on work would be something like (to be
> written or reviewed by native Englishman) :
>
> PROPOSAL IMPLEMENTATION HERE ↓ WOW ! (thanks to SwissChris)
>
> ☞ « Indicates the artist who arranged a tune into a form suitable for
> performance.
> Don’t use this on non classical works — unless you can prove that the same
> specific arrangement is used by at least two different recordings (by
> different performers) — link to recordings instead. »
>
> -----
> jesus2099 × Ti = Tristan + patate12 ÷ saucisson7
> mb : http://musicbrainz.org/user/jesus2099
> mb userscripts : http://userscripts.org/users/31010/scripts
> --
> View this message in context: http://musicbrainz.1054305.n4.nabble.com/RFC-335-Forbid-arrange-on-non-classical-works-tp3812157p3812157.html
> Sent from the Musicbrainz - Style mailing list archive at Nabble.com.
>
> _______________________________________________
> MusicBrainz-style mailing list
> [hidden email]
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style


_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|

Re: RFC-335. Forbid arrange on non classical works

CallerNo6
In reply to this post by jesus2099
I agree with the need for a "test of Work-ness" but would (personally)
expect the bar to be a little bit higher than "two performances by
different artists". Not a deal-breaker for me, though.

---

On a related note, I feel like this whole discussion is based on the
idea that a Work is a Composition (which is the usual meaning, more or
less). I understand the desire to keep an artist's Work/Composition list
manageable.

Would the problem (of Arrangements mucking up Works Lists) disappear if
we moved away from the Work=Composition equivalence? That's what I
/hope/ we can do using Work Types[1] and a move comprehensive set of
Work-Work relationships.

Or is the real problem one of complexity vs usablility? If the goal (of
proposals like this) is to limit the complexity of Work-Work
relationships then I'm completely off-track.

Alex / caller#6


[1] http://wiki.musicbrainz.org/User:PBryan/Work/Draft_1#Attributes

On 09/21/2011 08:06 AM, jesus2099 wrote:

> So is proposed this swisschris’ new text for the docs instead :
>
> « Arrangement (and orchestration!) credits should  be linked to recordings,
> not to works, for every kind of music (not only for non-classical).
>
> There are known cases (particularly in score-based classical), where
> specific arrangements or orchestrations have become works in their own
> right, being recorded by various performers over and over. { here we’ll
> paste several examples as quoted before [1] in a bullet list }. This is not
> necessarily limited to classical: one could also think of arrangements of
> some popular tunes for brass bands, or of parody covers of a given
> performance by a given artist, re-using the specific arrangements…
>
> For such cases – and only for such cases – an "arrange" AR would still be
> needed at work level. A new work based on arranger or orchestrator credits
> should not be allowed to be created unless one can prove that there are at
> least two distinct recordings by two different performers of this specific
> arrangement/orchestration. »
>
> The new RFC-335 delay re-starts from today.
> Feel free to modify it a little bit.
>
> -----
> jesus2099 × Ti = Tristan + patate12 ÷ saucisson7
>

_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
Reply | Threaded
Open this post in threaded view
|

Re: RFC-335. Forbid arrange on non classical works

jesus2099
In reply to this post by Nikki-3
Nikki-3 wrote
Where is the new version of the arranger wiki page?

Nikki
http://wiki.musicbrainz.org/Arranger_Relationship_Type
is already OK actually as it doesn’t mention works arranger at all
(because the work arranger relationship was added without RFC/RFV).
The proposed text is intended to show up in the the edit page helper when you try to make a work arrange edit.
I don’t know where do we change this text.
 PATATE12   jesus2099   GOLD MASTER KING   FAKE E-MAIL ADDRESS 
Reply | Threaded
Open this post in threaded view
|

Re: RFC-335. Forbid arrange on non classical works

Nikki-3
jesus2099 wrote:
> http://wiki.musicbrainz.org/Arranger_Relationship_Type
> is already OK actually as it doesn’t mention works arranger at all
> (because the work arranger relationship was added without RFC/RFV).
> The proposed text is intended to show up in the the edit page helper when
> you try to make a work arrange edit.
> I don’t know where do we change this text.

The wiki is the primary place for guidelines, so any guidelines about
which level arrangers should credited at belong on that page. I've
updated it to reflect how it's implemented in NGS so now you can make a
copy and add the guidelines you'd like to add.

Once the wiki page has guidelines, then a (preferably short) bit of text
can be added to the relationship description on the site too. Based on
the previous posts in this thread, I'd suggest something like "Arrangers
should normally be credited on recordings! Please read (link to wiki
page with guidelines) before adding an arranger to a work" (bold too, if
you want ;)).

Nikki

P.S. I'm sorry if you don't like arranger being work level, but that's
just what was decided as part of the NGS migration. Originally it was
going to be work-level only and it's only because there was disagreement
about which level is better that it was decided to have it on recordings
at all.

_______________________________________________
MusicBrainz-style mailing list
[hidden email]
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style