Re: [mb-style] Works and remixes/covers

[caramel]

Since the remix credits are at the recording level. Remixes are not separate
Works... and should be linked to the original work.

2011/5/18 Lukáš Lalinský 

> The original NGS documentation which was written by me [1] says that
> works with different remixes should be merged together. The Work page
> [2] however says that they should be kep separate. All people I've
> discussed this with agree that they should be merged. Can I change the
> Work page to remove the mention of remixes?
>
> Lukas
>
> [1] http://wiki.musicbrainz.org/Next_Generentation which was written by me
> [1] says that
> works with different remixes should be merged together. The Work page
> [2] however says that they should be kep separate. All people I've
> discussed this with 
> agation_Schema#Work
> [2] http://wiki.musicbrainz.org/Work
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[caller#6]

On 05/17/2011 10:27 PM, Lukáš Lalinský wrote:
> The original NGS documentation which was written by me [1] says that
> works with different remixes should be merged together. The Work page
> [2] however says that they should be kep separate. All people I've
> discussed this with agree that they should be merged. Can I change the
> Work page to remove the mention of remixes?
>
> Lukas
>
> [1] http://wiki.musicbrainz.org/Next_Generation_Schema#Work
> [2] http://wiki.musicbrainz.org/Work
>

Do you disagree with the concept of "Derivative Works" [1]? How many of 
the items listed under "Distinctiveness" should be removed?

Alex / caller#6

[1]"A work can be expressed as being derived from one or more other 
works. Examples: instrumental work with lyrics added later, translation 
of a work into a different language, remix, mashup. " from 
http://wiki.musicbrainz.org/Work#Work-to-Work_relationships

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Frederic Da Vitoria]

I guess I have less than 20 remix tracks in all my CDs, so my point of view
is essentially theoretical, but I believe you are confusing what happens to
be technically implemented in MB and what should be done in order to have a
satisfying database structure. The fact remix credits are currently at
recording level means only that if we decide that remixes should be separate
works, then we have a technical problem (which shouldn't be too difficult to
correct). But IMO it does not mean that remixes are or are not distinct
works.

I always felt that there was no simple answer to that question. Some remixes
are just the same recording subtly modified, so that when you listen to them
you perceive a subtle change in mood but you don't have any problems
recognizing the "original" track. But sometimes the remix is so different
that it is only in the middle of the track that you suddenly recognize it,
and even then you sometimes keep losing it. I experienced this with the
Beatles "LOVE" album. Maybe this is only my classical music lover's point of
view, but I feel that remixes are as different as Ravel's Pictures at an
Exhibition is different from Mussorgsky's. So that if we separate Ravel's
work from Mussorgsky's (we do), I'd feel strange not to distinguish between
some remixes.

I realize that the problem with my position is that this would have to be
decided for each remix.

2011/5/18 caramel 

> Since the remix credits are at the recording level. Remixes are not
> separate Works... and should be linked to the original work.
>
> 2011/5/18 Lukáš Lalinský 
>
>> The original NGS documentation which was written by me [1] says that
>> works with different remixes should be merged together. The Work page
>> [2] however says that they should be kep separate. All people I've
>> discussed this with agree that they should be merged. Can I change the
>> Work page to remove the mention of remixes?
>>
>> Lukas
>>
>> [1] http://wiki.musicbrainz.org/Next_Generentation which was written by
>> me [1] says that
>>
>> works with different remixes should be merged together. The Work page
>> [2] however says that they should be kep separate. All people I've
>> discussed this with 
>> agation_Schema#Work
>>
>> [2] http://wiki.musicbrainz.org/Work
>>
>
-- 
Frederic Da Vitoria
(davitof)

Membre de l'April - « promouvoir et défendre le logiciel libre » -
http://www.april.org
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Lukáš Lalinský]

On Wed, May 18, 2011 at 8:55 AM, caller#6
 wrote:
> Do you disagree with the concept of "Derivative Works" [1]? How many of
> the items listed under "Distinctiveness" should be removed?

My problem is that the situation is not consistent. We are merging
cover recordings into the same work, but remixes are supposed to not
merge remixes. I think they are very similar situations, in both cases
they are derived works. In my opinion, we should either merge remixes
and have a special relationship to link a work to a remixed recording,
or have separate works and have a special relationships to link two
remixed works.

Lukas

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[lorenz pressler]

caller#6 wrote:
> 
> Do you disagree with the concept of "Derivative Works" [1]? How many of 
> the items listed under "Distinctiveness" should be removed?
> 

i think a derivate work has to earn it's status of an own work and (as
Frederic said) this would have to be decided for each derivate work. and i
would decide this on demand: as soon more than one recording would link to a
derivate work it is viable to create it. the whole point in having works
introduced in MB is that you can link multiple records to it. theres no
point in creating tons of new derivate works just to honor their
'sophisticated work'.

eg. "Seven Variations in E-flat major on Mozart's "Bei Männern, welche Liebe
fühlen" (from 'Die Zauberflöte'), WoO 46" by Beethoven might have its own
work entry, but only if there is more than one recording of it available.

this would deny most pop-remixes and the likes an own work entry i guess. 
so i would deny creating independent works for remixes, mashups,
instrumental work with lyrics added later, translation of a work into a
different language in general while making it possible with an exceptional
rule.

greets lorenz/ehrgeiz

 

--
View this message in context: 
http://musicbrainz.1054305.n4.nabble.com/Works-and-remixes-covers-tp3531609p3533497.html
Sent from the Musicbrainz - Style mailing list archive at Nabble.com.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Nikki]

lorenz pressler wrote:

> this would deny most pop-remixes and the likes an own work entry i guess. 
> so i would deny creating independent works for remixes, mashups,
> instrumental work with lyrics added later, translation of a work into a
> different language in general while making it possible with an exceptional
> rule.

Translations have been migrated as different works, they have different 
lyricists and we store lyricists at work level.

Nikki

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Per Starbäck]

> Translations have been migrated as different works, they have different
> lyricists and we store lyricists at work level.

Are covers without lyrics its own work then? Like for instance
"Yesterday" with Oscar Peterson:
http://musicbrainz.org/recording/b0da13b6-12db-4391-9a2c-1dd26fb4cf3a

And is that then the same work as any other cover of the same song
without lyrics, for instance
http://musicbrainz.org/recording/3a21b38a-1e9a-4bfd-a229-6b4cc5a19864
with The Flame All Stars?

I guess that makes sense, even though these two instrumental versions
are two different covers of the original rather than two versions of
the same cover.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Dr Andrew John Hughes]

On 18 May 2011 19:12, lorenz pressler  wrote:
>
> caller#6 wrote:
>>
>> Do you disagree with the concept of "Derivative Works" [1]? How many of
>> the items listed under "Distinctiveness" should be removed?
>>
>
> i think a derivate work has to earn it's status of an own work and (as
> Frederic said) this would have to be decided for each derivate work. and i
> would decide this on demand: as soon more than one recording would link to a
> derivate work it is viable to create it. the whole point in having works
> introduced in MB is that you can link multiple records to it. theres no
> point in creating tons of new derivate works just to honor their
> 'sophisticated work'.
>
> eg. "Seven Variations in E-flat major on Mozart's "Bei Männern, welche Liebe
> fühlen" (from 'Die Zauberflöte'), WoO 46" by Beethoven might have its own
> work entry, but only if there is more than one recording of it available.
>
> this would deny most pop-remixes and the likes an own work entry i guess.
> so i would deny creating independent works for remixes, mashups,
> instrumental work with lyrics added later, translation of a work into a
> different language in general while making it possible with an exceptional
> rule.
>

I think I agree with this logic in general.  Having a work entity has
always meant to me that we have one common place to store information
like composer and lyricist.  Separating every remix out as a separate
work (there can be a multitude for some pieces) pretty much defeats
the point in having a work as distinct from a recording for me.

Remixes essentially differ in production and these credits are held at
recording level, not work level.  The line between a cover and a remix
can be quite thin, and I don't see why a remix should be a new work
but a cover isn't.

> greets lorenz/ehrgeiz
>
>
>
> --
> View this message in context: 
> http://musicbrainz.1054305.n4.nabble.com/Works-and-remixes-covers-tp3531609p3533497.html
> Sent from the Musicbrainz - Style mailing list archive at Nabble.com.
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style



-- 
Andrew :-)

Support Free Java!
Contribute to GNU Classpath and IcedTea
http://www.gnu.org/software/classpath
http://icedtea.classpath.org

PGP Key: F5862A37 (https://keys.indymedia.org/)
Fingerprint = EA30 D855 D50F 90CD F54D  0698 0713 C3ED F586 2A37

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[caramel]

It is always questionable. Other example in Classical:

Satie composed "Gymnopédie No. 1" for piano solo ==> one work.
Debussy orchestrated this work ==> another work (Orch. AR is at work level!)
Jeroen van Veen "arranged" the orchestrated version to get a four-hands
piano version.
Arr. AR is at recording level ==> no work

Since remix ARs is actually at recording level ==> no separate work

Should we change that ?
Not sure... The NGS wanted to present a pyramidal and consolidated view of
the database with N-tracks to 1-recording and N-recordings to 1-work. If we
multiply the number of works, we are loosing this concept and after all, we
should add links between the works to say that it is a cover or a remix...
or a slightly modified version... The principal for cover and remixes is to
set all the credits in the recordings.
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Frederic Da Vitoria]

2011/5/19 caramel 

> It is always questionable. Other example in Classical:
>
> Satie composed "Gymnopédie No. 1" for piano solo ==> one work.
> Debussy orchestrated this work ==> another work (Orch. AR is at work
> level!)
> Jeroen van Veen "arranged" the orchestrated version to get a four-hands
> piano version.
> Arr. AR is at recording level ==> no work
>
> Since remix ARs is actually at recording level ==> no separate work
>
> Should we change that ?
> Not sure... The NGS wanted to present a pyramidal and consolidated view of
> the database with N-tracks to 1-recording and N-recordings to 1-work. If we
> multiply the number of works, we are loosing this concept and after all, we
> should add links between the works to say that it is a cover or a remix...
> or a slightly modified version... The principal for cover and remixes is to
> set all the credits in the recordings.
>

We don't loose the structure by using multiple works, if we link works
together. It's just that rather than having a blurry global work tier, the
work level would itself be a pyramid.

-- 
Frederic Da Vitoria
(davitof)

Membre de l'April - « promouvoir et défendre le logiciel libre » -
http://www.april.org
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[caller#6]

On 05/18/2011 01:00 AM, Lukáš Lalinský wrote:
> On Wed, May 18, 2011 at 8:55 AM, caller#6
>   wrote:
>> Do you disagree with the concept of "Derivative Works" [1]? How many of
>> the items listed under "Distinctiveness" should be removed?
> My problem is that the situation is not consistent. We are merging
> cover recordings into the same work, but remixes are supposed to not
> merge remixes. I think they are very similar situations, in both cases
> they are derived works. In my opinion, we should either merge remixes
> and have a special relationship to link a work to a remixed recording,
> or have separate works and have a special relationships to link two
> remixed works.
>
> Lukas
+1 for consistency.
Could a re-mix be considered special-case of mashup? One which happens 
to include only one song?

c.f. http://forums.musicbrainz.org/viewtopic.php?id=2699

Alex / caller#6



___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Dr Andrew John Hughes]

On 19 May 2011 17:21, caller#6  wrote:
>
>
> On 05/18/2011 01:00 AM, Lukáš Lalinský wrote:
>> On Wed, May 18, 2011 at 8:55 AM, caller#6
>>   wrote:
>>> Do you disagree with the concept of "Derivative Works" [1]? How many of
>>> the items listed under "Distinctiveness" should be removed?
>> My problem is that the situation is not consistent. We are merging
>> cover recordings into the same work, but remixes are supposed to not
>> merge remixes. I think they are very similar situations, in both cases
>> they are derived works. In my opinion, we should either merge remixes
>> and have a special relationship to link a work to a remixed recording,
>> or have separate works and have a special relationships to link two
>> remixed works.
>>
>> Lukas
> +1 for consistency.
> Could a re-mix be considered special-case of mashup? One which happens
> to include only one song?
>
> c.f. http://forums.musicbrainz.org/viewtopic.php?id=2699
>
> Alex / caller#6
>
>
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>

To a degree, yes.  It's essentially different and/or additional
production on the same track in most cases.  There are complex
instances, as mentioned in the forum post you linked to, but IMHO they
are the minority and can be decided if they are due work status on a
case-by-case basis (e.g. if they have an additional lyricist).
-- 
Andrew :-)

Support Free Java!
Contribute to GNU Classpath and IcedTea
http://www.gnu.org/software/classpath
http://icedtea.classpath.org

PGP Key: F5862A37 (https://keys.indymedia.org/)
Fingerprint = EA30 D855 D50F 90CD F54D  0698 0713 C3ED F586 2A37

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Aurélien Mino]

On 05/19/2011 08:57 AM, caramel wrote:
> Jeroen van Veen "arranged" the orchestrated version to get a 
> four-hands piano version.
> Arr. AR is at recording level ==> no work

It was planned to add "Arranger" at work level too after NGS migration, 
exactly for this situation.
See 
http://wiki.musicbrainz.org/Next_Generation_Schema/Track_Relationships_Conversion#Artist-Track

It just has not been added yet.

- Aurélien

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Nikki]

Aurélien Mino wrote:
> It was planned to add "Arranger" at work level too after NGS migration, 
> exactly for this situation.
> See 
> http://wiki.musicbrainz.org/Next_Generation_Schema/Track_Relationships_Conversion#Artist-Track
> 
> It just has not been added yet.

I thought the upgrades script would do it. :/ Oh well, I can add it now.

Should something be added to the relationship type page about when to 
use which?

Nikki

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Nikki]

Per Starbäck wrote:
> Are covers without lyrics its own work then? Like for instance
> "Yesterday" with Oscar Peterson:
> http://musicbrainz.org/recording/b0da13b6-12db-4391-9a2c-1dd26fb4cf3a

Hmm... Not sure. I think I would probably still use the original work, 
the work still has lyrics even if someone decided not to perform them.

Nikki

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[lorenz pressler]

and vice versa you can say the work of a cover where lyrics where added still
has no lyrics. perfect solution imho would be an "additional lyrics" field
on recording lvl. however i guess i could live with an own work-entry and
work-work relationships in these cases although i personally won't credit
covers and remixes the status of a workpiece in most cases.
although it feels quite akward then to grant some tracks an own work-entry
and others not just because "lyrics are stored on work level". 

--
View this message in context: 
http://musicbrainz.1054305.n4.nabble.com/Works-and-remixes-covers-tp3531609p3538864.html
Sent from the Musicbrainz - Style mailing list archive at Nabble.com.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Frederic Da Vitoria]

2011/5/20 Nikki 

> Per Starbäck wrote:
> > Are covers without lyrics its own work then? Like for instance
> > "Yesterday" with Oscar Peterson:
> > http://musicbrainz.org/recording/b0da13b6-12db-4391-9a2c-1dd26fb4cf3a
>
> Hmm... Not sure. I think I would probably still use the original work,
> the work still has lyrics even if someone decided not to perform them.
>

For me, it depends on whether the work sounds like the original or if
removing the lyrics is part of a re-creation of the work. Knowing a little
Oscar Peterson, I'd guess in this example I'd choose the last option.

-- 
Frederic Da Vitoria
(davitof)

Membre de l'April - « promouvoir et défendre le logiciel libre » -
http://www.april.org
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Dr Andrew John Hughes]

2011/5/18 Lukáš Lalinský :
> The original NGS documentation which was written by me [1] says that
> works with different remixes should be merged together. The Work page
> [2] however says that they should be kep separate. All people I've
> discussed this with agree that they should be merged. Can I change the
> Work page to remove the mention of remixes?
>
> Lukas
>
> [1] http://wiki.musicbrainz.org/Next_Generation_Schema#Work
> [2] http://wiki.musicbrainz.org/Work
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>

Can we please come to a decision on this?  I now have a number of
merges being voted down (e.g. http://musicbrainz.org/edit/14496213)
due to this dumb guideline.  It makes no sense to have remixes as
separate works, especially when the remix AR is at recording level and
covers are not separate works but are more different (new vocals, new
production, etc.).  I really don't see the point in works if we're
going to allow an artist's work catalogue to be filled with five or
ten remixes of each song.
-- 
Andrew :-)

Support Free Java!
Contribute to GNU Classpath and IcedTea
http://www.gnu.org/software/classpath
http://icedtea.classpath.org

PGP Key: F5862A37 (https://keys.indymedia.org/)
Fingerprint = EA30 D855 D50F 90CD F54D  0698 0713 C3ED F586 2A37

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[caller#6]

On 05/24/2011 08:02 AM, Dr Andrew John Hughes wrote:
> 2011/5/18 Lukáš Lalinský:
>> The original NGS documentation which was written by me [1] says that
>> works with different remixes should be merged together. The Work page
>> [2] however says that they should be kep separate. All people I've
>> discussed this with agree that they should be merged. Can I change the
>> Work page to remove the mention of remixes?
>>
>> Lukas
>>
>> [1] http://wiki.musicbrainz.org/Next_Generation_Schema#Work
>> [2] http://wiki.musicbrainz.org/Work
>>
> Can we please come to a decision on this?  I now have a number of
> merges being voted down (e.g. http://musicbrainz.org/edit/14496213)
> due to this dumb guideline.  It makes no sense to have remixes as
> separate works, especially when the remix AR is at recording level and
> covers are not separate works but are more different (new vocals, new
> production, etc.).  I really don't see the point in works if we're
> going to allow an artist's work catalogue to be filled with five or
> ten remixes of each song.
For the sake of consistency, I'd rather see the entire list of 
"derivative works"[1] addressed, rather than just re-mixes. But this is 
a good first step.

Putting most re-mix/mashup/medley/ Relationships at the Recording level 
makes sense to me[2]. Or, put another way, some "distinctiveness" 
belongs at the Performance level.

Alex / caller#6


[1] http://wiki.musicbrainz.org/Work#Distinctiveness




___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[MeinDummy]

+1 for merging remix and cover works into the original works

--
View this message in context: 
http://musicbrainz.1054305.n4.nabble.com/Works-and-remixes-covers-tp3531609p3549034.html
Sent from the Musicbrainz - Style mailing list archive at Nabble.com.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Stephen]

I have a question that hasn't been addressed so far: sometimes a cover  
will introduce a specific innovation (could be an extra verse,  
dropping a verse, specific flourishes, restructuring, whatever) into a  
song, and later covers will reproduce that innovation. So those later  
covers aren't covers of the original, but specifically covers of that  
other cover. Is there a way to record this information in either the  
old, current regime or the new proposed one?

I agree that most of the time covers should link to the original, but  
some covers actually are covers of specific later versions. I'm not  
sure how best that might be handled though. Maybe just in annotations  
or something, because it strikes me that the ontological way of  
allowing some covers to generate new works and other covers to not do  
so would likely become a contentious mess.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Frederic Da Vitoria]

2011/5/25 Stephen 

> I have a question that hasn't been addressed so far: sometimes a cover
> will introduce a specific innovation (could be an extra verse,
> dropping a verse, specific flourishes, restructuring, whatever) into a
> song, and later covers will reproduce that innovation. So those later
> covers aren't  covers of the original, but specifically covers of that
> other cover. Is there a way to record this information in either the
> old, current regime or the new proposed one?
>
> I agree that most of the time covers should link to the original, but
> some covers actually are covers of specific later versions. I'm not
> sure how best that might be handled though. Maybe just in annotations
> or something, because it strikes me that the ontological way of
> allowing some covers to generate new works and other covers to not do
> so would likely become a contentious mess.
>

Also, couldn't it happen that a cover takes it's ideas from two distinct
covers ? Even more messy :-)

-- 
Frederic Da Vitoria
(davitof)

Membre de l'April - « promouvoir et défendre le logiciel libre » -
http://www.april.org
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[caramel]

2011/5/25 Frederic Da Vitoria 

> 2011/5/25 Stephen 
>
>> I have a question that hasn't been addressed so far: sometimes a cover
>> will introduce a specific innovation (could be an extra verse,
>> dropping a verse, specific flourishes, restructuring, whatever) into a
>> song, and later covers will reproduce that innovation. So those later
>> covers aren't  covers of the original, but specifically covers of that
>> other cover. Is there a way to record this information in either the
>> old, current regime or the new proposed one?
>>
>> I agree that most of the time covers should link to the original, but
>> some covers actually are covers of specific later versions. I'm not
>> sure how best that might be handled though. Maybe just in annotations
>> or something, because it strikes me that the ontological way of
>> allowing some covers to generate new works and other covers to not do
>> so would likely become a contentious mess.
>>
>
> Also, couldn't it happen that a cover takes it's ideas from two distinct
> covers ? Even more messy :-)
>
>
> The rule could be to merge the works of different covers and remixes... but
allowing exceptions.
The exceptions should be documented in the annotation and in the notes of
the edits. There are too much cases to handle... and the guidelines must
remain simple. If someone think that the works should be distinctive, he has
to report why when creating a new work... in the annotation field AND in the
comment field too.
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[dj empirical]

On Wed, May 25, 2011 at 12:41 PM, Frederic Da Vitoria
 wrote:
> 2011/5/25 Stephen 
>>
>> I have a question that hasn't been addressed so far: sometimes a cover
>> will introduce a specific innovation (could be an extra verse,
>> dropping a verse, specific flourishes, restructuring, whatever) into a
>> song, and later covers will reproduce that innovation. So those later
>> covers aren't  covers of the original, but specifically covers of that
>> other cover. Is there a way to record this information in either the
>> old, current regime or the new proposed one?
>>
>> I agree that most of the time covers should link to the original, but
>> some covers actually are covers of specific later versions. I'm not
>> sure how best that might be handled though. Maybe just in annotations
>> or something, because it strikes me that the ontological way of
>> allowing some covers to generate new works and other covers to not do
>> so would likely become a contentious mess.
>
>
> Also, couldn't it happen that a cover takes it's ideas from two distinct
> covers ? Even more messy :-)

a couple examples, off the top of my head:

Cover of a cover: a few Metallica tribute cd's include performances of
"Am I Evil" and "Blitzkrieg", neither of which is an original
Metallica tune.
example: http://musicbrainz.org/release/487c7cd5-37f5-4b6c-8cd9-a1446d069828

Cover of two songs at once: A Perfect Circle combined Ozzy's "Diary of
a Madman" and The Cure's "Lovesong" into one tune. We have a few
instances of this in MB, mostly (i'm assuming) on bootlegs:
http://musicbrainz.org/artist/078a9376-3c04-4280-b7d7-b20e158f345d/recordings?page=3

-- 
--dj empirical--

http://djempirical.com || http://www.flickr.com/photos/dj_empirical/
|| http://twitter.com/djempirical ||
http://www.facebook.com/pages/DJ-Empirical/497598105223
podcast: wearenotdelicious.blogspot.com/

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Frederic Da Vitoria]

2011/5/25 caramel 

>
> 2011/5/25 Frederic Da Vitoria 
>
>> 2011/5/25 Stephen 
>>
>>> I have a question that hasn't been addressed so far: sometimes a cover
>>> will introduce a specific innovation (could be an extra verse,
>>> dropping a verse, specific flourishes, restructuring, whatever) into a
>>> song, and later covers will reproduce that innovation. So those later
>>> covers aren't  covers of the original, but specifically covers of that
>>> other cover. Is there a way to record this information in either the
>>> old, current regime or the new proposed one?
>>>
>>> I agree that most of the time covers should link to the original, but
>>> some covers actually are covers of specific later versions. I'm not
>>> sure how best that might be handled though. Maybe just in annotations
>>> or something, because it strikes me that the ontological way of
>>> allowing some covers to generate new works and other covers to not do
>>> so would likely become a contentious mess.
>>>
>>
>> Also, couldn't it happen that a cover takes it's ideas from two distinct
>> covers ? Even more messy :-)
>>
>>
>> The rule could be to merge the works of different covers and remixes...
> but allowing exceptions.
> The exceptions should be documented in the annotation and in the notes of
> the edits. There are too much cases to handle... and the guidelines must
> remain simple. If someone think that the works should be distinctive, he has
> to report why when creating a new work... in the annotation field AND in the
> comment field too.
>

Yes, this is why these are called Guides and not Laws or Rules, but it is
always good to remind it.

-- 
Frederic Da Vitoria
(davitof)

Membre de l'April - « promouvoir et défendre le logiciel libre » -
http://www.april.org
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Simon Austin]

Looking at the work-work and work-recording relationships possible, all 
these seem to be covered already... unless I'm missing something.

For example, wouldn't a cover from two (or more) distinct covers just be 
a medley?

- Si

On 25/05/2011 19:34, dj empirical wrote:
> On Wed, May 25, 2011 at 12:41 PM, Frederic Da Vitoria
>   wrote:
>> 2011/5/25 Stephen
>>> I have a question that hasn't been addressed so far: sometimes a cover
>>> will introduce a specific innovation (could be an extra verse,
>>> dropping a verse, specific flourishes, restructuring, whatever) into a
>>> song, and later covers will reproduce that innovation. So those later
>>> covers aren't  covers of the original, but specifically covers of that
>>> other cover. Is there a way to record this information in either the
>>> old, current regime or the new proposed one?
>>>
>>> I agree that most of the time covers should link to the original, but
>>> some covers actually are covers of specific later versions. I'm not
>>> sure how best that might be handled though. Maybe just in annotations
>>> or something, because it strikes me that the ontological way of
>>> allowing some covers to generate new works and other covers to not do
>>> so would likely become a contentious mess.
>>
>> Also, couldn't it happen that a cover takes it's ideas from two distinct
>> covers ? Even more messy :-)
> a couple examples, off the top of my head:
>
> Cover of a cover: a few Metallica tribute cd's include performances of
> "Am I Evil" and "Blitzkrieg", neither of which is an original
> Metallica tune.
> example: http://musicbrainz.org/release/487c7cd5-37f5-4b6c-8cd9-a1446d069828
>
> Cover of two songs at once: A Perfect Circle combined Ozzy's "Diary of
> a Madman" and The Cure's "Lovesong" into one tune. We have a few
> instances of this in MB, mostly (i'm assuming) on bootlegs:
> http://musicbrainz.org/artist/078a9376-3c04-4280-b7d7-b20e158f345d/recordings?page=3
>


___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Dr Andrew John Hughes]

On 25 May 2011 17:12, Stephen  wrote:
> I have a question that hasn't been addressed so far: sometimes a cover
> will introduce a specific innovation (could be an extra verse,
> dropping a verse, specific flourishes, restructuring, whatever) into a
> song, and later covers will reproduce that innovation. So those later
> covers aren't  covers of the original, but specifically covers of that
> other cover. Is there a way to record this information in either the
> old, current regime or the new proposed one?
>
> I agree that most of the time covers should link to the original, but
> some covers actually are covers of specific later versions. I'm not
> sure how best that might be handled though. Maybe just in annotations
> or something, because it strikes me that the ontological way of
> allowing some covers to generate new works and other covers to not do
> so would likely become a contentious mess.
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>

This is a concern of mine too and one I mentioned before NGS happened.
It's actually one of the reasons I find remixes being new works so odd, as
they are essentially a different mix of the same work (hence the name) while
covers involve at least a new artist and in some cases, a different
tune or arrangement
or different lyrics (e.g. parodies).

An example of what Stephen mentions would be the track "Mad World".
The original
by Tears for Fears is quite different to the cover by Michael Andrews (see
http://en.wikipedia.org/wiki/Mad_World).  The Michael Andrews version has since
been covered itself (e.g. on reality shows like The X Factor).
There's also a very different
toned down version of "I Want to Hold Your Hand" by The Beatles which
I believe was
first performed on the soundtrack to "Across the Universe" and then
covered by Chris
Colfer on Glee.

I think what these need is for a cover recording to be able to link
back to either a work or a recording.
At the moment, it seems I can't do a recording<->recording cover link
but funnily enough
I can do a recording<->recording AR for a remix...
-- 
Andrew :-)

Support Free Java!
Contribute to GNU Classpath and IcedTea
http://www.gnu.org/software/classpath
http://icedtea.classpath.org

PGP Key: F5862A37 (https://keys.indymedia.org/)
Fingerprint = EA30 D855 D50F 90CD F54D  0698 0713 C3ED F586 2A37

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[lorenz pressler]

+1

--
View this message in context: 
http://musicbrainz.1054305.n4.nabble.com/Works-and-remixes-covers-tp3531609p3551452.html
Sent from the Musicbrainz - Style mailing list archive at Nabble.com.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Stephen]

On May 25, 2011, at 8:42 PM, Dr Andrew John Hughes wrote:

> I think what these need is for a cover recording to be able to link
> back to either a work or a recording.
> At the moment, it seems I can't do a recording<->recording cover link
> but funnily enough
> I can do a recording<->recording AR for a remix...


A recording--recording link makes sense for a remix, because usually a  
remix is actually a differently edited version of the same recorded  
audio, whereas a new cover version does not usually incorporate audio  
from other versions of the song being covered. Work<->recording seems  
like it would be the appropriate link for a cover to me, even a cover  
based on a version not the original.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Dr Andrew John Hughes]

On 26 May 2011 07:00, Stephen  wrote:
>
> On May 25, 2011, at 8:42 PM, Dr Andrew John Hughes wrote:
>
>> I think what these need is for a cover recording to be able to link
>> back to either a work or a recording.
>> At the moment, it seems I can't do a recording<->recording cover link
>> but funnily enough
>> I can do a recording<->recording AR for a remix...
>
>
> A recording--recording link makes sense for a remix, because usually a
> remix is actually a differently edited version of the same recorded
> audio

I agree.  That's why I don't think they should be new works.

, whereas a new cover version does not usually incorporate audio
> from other versions of the song being covered. Work<->recording seems
> like it would be the appropriate link for a cover to me, even a cover
> based on a version not the original.

But how do you create such an AR if the cover is not a work?

>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>



-- 
Andrew :-)

Support Free Java!
Contribute to GNU Classpath and IcedTea
http://www.gnu.org/software/classpath
http://icedtea.classpath.org

PGP Key: F5862A37 (https://keys.indymedia.org/)
Fingerprint = EA30 D855 D50F 90CD F54D  0698 0713 C3ED F586 2A37

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Nicolás Tamargo de Eguren]

Now for some conflicting opinion: the Spanish work database includes
"Asómate" – http://musicbrainz.org/work/148859be-4c6e-30f1-9558-3f48a42c8101
– and "Asómate (remix)" –
http://musicbrainz.org/work/8eb73a24-e600-46a5-af42-0314e65775d2 – as
different works, with different ISWCs :)

On Thu, May 26, 2011 at 10:49 AM, Dr Andrew John Hughes
 wrote:
> On 26 May 2011 07:00, Stephen  wrote:
>>
>> On May 25, 2011, at 8:42 PM, Dr Andrew John Hughes wrote:
>>
>>> I think what these need is for a cover recording to be able to link
>>> back to either a work or a recording.
>>> At the moment, it seems I can't do a recording<->recording cover link
>>> but funnily enough
>>> I can do a recording<->recording AR for a remix...
>>
>>
>> A recording--recording link makes sense for a remix, because usually a
>> remix is actually a differently edited version of the same recorded
>> audio
>
> I agree.  That's why I don't think they should be new works.
>
> , whereas a new cover version does not usually incorporate audio
>> from other versions of the song being covered. Work<->recording seems
>> like it would be the appropriate link for a cover to me, even a cover
>> based on a version not the original.
>
> But how do you create such an AR if the cover is not a work?
>
>>
>> ___
>> MusicBrainz-style mailing list
>> MusicBrainz-style@lists.musicbrainz.org
>> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>>
>
>
>
> --
> Andrew :-)
>
> Support Free Java!
> Contribute to GNU Classpath and IcedTea
> http://www.gnu.org/software/classpath
> http://icedtea.classpath.org
>
> PGP Key: F5862A37 (https://keys.indymedia.org/)
> Fingerprint = EA30 D855 D50F 90CD F54D  0698 0713 C3ED F586 2A37
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style



-- 
Nicolás Tamargo de Eguren

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Kuno Woudt]

Hello,

On 26/05/11 10:49, Dr Andrew John Hughes wrote:
> On 26 May 2011 07:00, Stephen  wrote:
>>
>> On May 25, 2011, at 8:42 PM, Dr Andrew John Hughes wrote:
>>
>>> I think what these need is for a cover recording to be able to link
>>> back to either a work or a recording.
>>> At the moment, it seems I can't do a recording<->recording cover link
>>> but funnily enough
>>> I can do a recording<->recording AR for a remix...
>>
>>
>> A recording--recording link makes sense for a remix, because usually a
>> remix is actually a differently edited version of the same recorded
>> audio
>
> I agree.  That's why I don't think they should be new works.

I think the keyword in the above statement is "usually".  Perhaps most
remixes fit this description, but there are also cases where a track
is reworked so thoroughly it's more akin to a new track which uses a
few samples from the old one.

 From that perspective I would prefer to allow both.  But that of course
invites edit-note bickering about whether a particular remix constitutes
a new work or not.

Perhaps only allow works for remixes if the liner notes specify certain 
roles for artists which differ from the original work and can only be 
represented with work relationships.

-- kuno / warp.


___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Paul C. Bryan]

Sorry for jumping into the discussion so late in the game—real life has
been demanding much of my time lately.

I would suggest the test could boil-down to whether a new artist work AR
is required to distinguish one remix from another (if a particular remix
contains new content, and has additional composer/lyricist/etc. credit).
If so, then a separate (but related) work would seem to be in order. If
not, then the original work would be used.

Thoughts?

Paul

On Sun, 2011-05-29 at 08:42 +0200, Kuno Woudt wrote:

> Hello,
> 
> On 26/05/11 10:49, Dr Andrew John Hughes wrote:
> > On 26 May 2011 07:00, Stephen  wrote:
> >>
> >> On May 25, 2011, at 8:42 PM, Dr Andrew John Hughes wrote:
> >>
> >>> I think what these need is for a cover recording to be able to link
> >>> back to either a work or a recording.
> >>> At the moment, it seems I can't do a recording<->recording cover link
> >>> but funnily enough
> >>> I can do a recording<->recording AR for a remix...
> >>
> >>
> >> A recording--recording link makes sense for a remix, because usually a
> >> remix is actually a differently edited version of the same recorded
> >> audio
> >
> > I agree.  That's why I don't think they should be new works.
> 
> I think the keyword in the above statement is "usually".  Perhaps most
> remixes fit this description, but there are also cases where a track
> is reworked so thoroughly it's more akin to a new track which uses a
> few samples from the old one.
> 
>  From that perspective I would prefer to allow both.  But that of course
> invites edit-note bickering about whether a particular remix constitutes
> a new work or not.
> 
> Perhaps only allow works for remixes if the liner notes specify certain 
> roles for artists which differ from the original work and can only be 
> represented with work relationships.
> 
> -- kuno / warp.
> 
> 
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style


___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Dr Andrew John Hughes]

On 6 June 2011 18:22, Paul C. Bryan  wrote:
> Sorry for jumping into the discussion so late in the game—real life has been
> demanding much of my time lately.
>
> I would suggest the test could boil-down to whether a new artist work AR is
> required to distinguish one remix from another (if a particular remix
> contains new content, and has additional composer/lyricist/etc. credit). If
> so, then a separate (but related) work would seem to be in order. If not,
> then the original work would be used.
>
> Thoughts?
>

Pretty much what I think too.  It has the benefit of being a very
straight-forward test based on clear technical grounds (do I need an
AR specific to this mix --> Yes, add a new work).
Now it would be nice if we actually had a work to work remix AR to go
with that...

> Paul
>
> On Sun, 2011-05-29 at 08:42 +0200, Kuno Woudt wrote:
>
> Hello,
>
> On 26/05/11 10:49, Dr Andrew John Hughes wrote:
>> On 26 May 2011 07:00, Stephen  wrote:
>>>
>>> On May 25, 2011, at 8:42 PM, Dr Andrew John Hughes wrote:
>>>
 I think what these need is for a cover recording to be able to link
 back to either a work or a recording.
 At the moment, it seems I can't do a recording<->recording cover link
 but funnily enough
 I can do a recording<->recording AR for a remix...
>>>
>>>
>>> A recording--recording link makes sense for a remix, because usually a
>>> remix is actually a differently edited version of the same recorded
>>> audio
>>
>> I agree.  That's why I don't think they should be new works.
>
> I think the keyword in the above statement is "usually".  Perhaps most
> remixes fit this description, but there are also cases where a track
> is reworked so thoroughly it's more akin to a new track which uses a
> few samples from the old one.
>
>  From that perspective I would prefer to allow both.  But that of course
> invites edit-note bickering about whether a particular remix constitutes
> a new work or not.
>
> Perhaps only allow works for remixes if the liner notes specify certain
> roles for artists which differ from the original work and can only be
> represented with work relationships.
>
> -- kuno / warp.
>
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>



-- 
Andii :-)

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Nicolás Tamargo de Eguren]

On Tue, Jun 7, 2011 at 1:50 AM, Dr Andrew John Hughes
 wrote:
> On 6 June 2011 18:22, Paul C. Bryan  wrote:
>> Sorry for jumping into the discussion so late in the game—real life has been
>> demanding much of my time lately.
>>
>> I would suggest the test could boil-down to whether a new artist work AR is
>> required to distinguish one remix from another (if a particular remix
>> contains new content, and has additional composer/lyricist/etc. credit). If
>> so, then a separate (but related) work would seem to be in order. If not,
>> then the original work would be used.
>>
>> Thoughts?
>>
>
> Pretty much what I think too.  It has the benefit of being a very
> straight-forward test based on clear technical grounds (do I need an
> AR specific to this mix --> Yes, add a new work).
> Now it would be nice if we actually had a work to work remix AR to go
> with that...

Make an RFC for it then ;)
("Do I need an AR specific to this mix OR does this mix have its own
ISWC" would be my test though)

>> Paul
>>
>> On Sun, 2011-05-29 at 08:42 +0200, Kuno Woudt wrote:
>>
>> Hello,
>>
>> On 26/05/11 10:49, Dr Andrew John Hughes wrote:
>>> On 26 May 2011 07:00, Stephen  wrote:

 On May 25, 2011, at 8:42 PM, Dr Andrew John Hughes wrote:

> I think what these need is for a cover recording to be able to link
> back to either a work or a recording.
> At the moment, it seems I can't do a recording<->recording cover link
> but funnily enough
> I can do a recording<->recording AR for a remix...


 A recording--recording link makes sense for a remix, because usually a
 remix is actually a differently edited version of the same recorded
 audio
>>>
>>> I agree.  That's why I don't think they should be new works.
>>
>> I think the keyword in the above statement is "usually".  Perhaps most
>> remixes fit this description, but there are also cases where a track
>> is reworked so thoroughly it's more akin to a new track which uses a
>> few samples from the old one.
>>
>>  From that perspective I would prefer to allow both.  But that of course
>> invites edit-note bickering about whether a particular remix constitutes
>> a new work or not.
>>
>> Perhaps only allow works for remixes if the liner notes specify certain
>> roles for artists which differ from the original work and can only be
>> represented with work relationships.
>>
>> -- kuno / warp.
>>
>>
>> ___
>> MusicBrainz-style mailing list
>> MusicBrainz-style@lists.musicbrainz.org
>> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>>
>>
>> ___
>> MusicBrainz-style mailing list
>> MusicBrainz-style@lists.musicbrainz.org
>> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>>
>
>
>
> --
> Andii :-)
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style



-- 
Nicolás Tamargo de Eguren

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Dr Andrew John Hughes]

2011/6/6 Nicolás Tamargo de Eguren :
> On Tue, Jun 7, 2011 at 1:50 AM, Dr Andrew John Hughes
>  wrote:
>> On 6 June 2011 18:22, Paul C. Bryan  wrote:
>>> Sorry for jumping into the discussion so late in the game—real life has been
>>> demanding much of my time lately.
>>>
>>> I would suggest the test could boil-down to whether a new artist work AR is
>>> required to distinguish one remix from another (if a particular remix
>>> contains new content, and has additional composer/lyricist/etc. credit). If
>>> so, then a separate (but related) work would seem to be in order. If not,
>>> then the original work would be used.
>>>
>>> Thoughts?
>>>
>>
>> Pretty much what I think too.  It has the benefit of being a very
>> straight-forward test based on clear technical grounds (do I need an
>> AR specific to this mix --> Yes, add a new work).
>> Now it would be nice if we actually had a work to work remix AR to go
>> with that...
>
> Make an RFC for it then ;)

I would if I knew of an example.  All the remixes I've seen don't
represent new works, but people keep claiming they exist.

> ("Do I need an AR specific to this mix OR does this mix have its own
> ISWC" would be my test though)

I hadn't even heard of an ISWC until I saw it under works, and still
haven't seen one.

>
--

Andii :-)

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Simon Austin]

On 07/06/2011 03:03, Dr Andrew John Hughes wrote:
> 2011/6/6 Nicolás Tamargo de Eguren:
>> On Tue, Jun 7, 2011 at 1:50 AM, Dr Andrew John Hughes
>>   wrote:
>>> On 6 June 2011 18:22, Paul C. Bryan  wrote:
 Sorry for jumping into the discussion so late in the game—real life has 
 been
 demanding much of my time lately.

 I would suggest the test could boil-down to whether a new artist work AR is
 required to distinguish one remix from another (if a particular remix
 contains new content, and has additional composer/lyricist/etc. credit). If
 so, then a separate (but related) work would seem to be in order. If not,
 then the original work would be used.

 Thoughts?

>>> Pretty much what I think too.  It has the benefit of being a very
>>> straight-forward test based on clear technical grounds (do I need an
>>> AR specific to this mix -->  Yes, add a new work).
>>> Now it would be nice if we actually had a work to work remix AR to go
>>> with that...
>> Make an RFC for it then ;)
> I would if I knew of an example.  All the remixes I've seen don't
> represent new works, but people keep claiming they exist.
>

Global Communication remixed Chapterhouse's "Blood Music" 
(http://musicbrainz.org/release/769aee1c-37b6-4788-8686-82277f2a2132) 
into "Pentamerous Metamorphosis" 
(http://musicbrainz.org/release/cd14f8ba-a764-49a8-a032-761bc917914e). 
It's usually called a reworking: 
http://en.wikipedia.org/wiki/Pentamerous_Metamorphosis

Might also consider Spacemonkeyz' "Laika Come Home", 
(http://musicbrainz.org/release/ce666d18-80b8-47d5-83d5-64e168a80f40) 
which is a remix of Gorillaz eponymous first album into reggae/dub 
versions. http://en.wikipedia.org/wiki/Laika_Come_Home

- Si

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Nicolás Tamargo de Eguren]

On Tue, Jun 7, 2011 at 5:03 AM, Dr Andrew John Hughes
 wrote:
> 2011/6/6 Nicolás Tamargo de Eguren :
>> On Tue, Jun 7, 2011 at 1:50 AM, Dr Andrew John Hughes
>>  wrote:
>>> On 6 June 2011 18:22, Paul C. Bryan  wrote:
 Sorry for jumping into the discussion so late in the game—real life has 
 been
 demanding much of my time lately.

 I would suggest the test could boil-down to whether a new artist work AR is
 required to distinguish one remix from another (if a particular remix
 contains new content, and has additional composer/lyricist/etc. credit). If
 so, then a separate (but related) work would seem to be in order. If not,
 then the original work would be used.

 Thoughts?

>>>
>>> Pretty much what I think too.  It has the benefit of being a very
>>> straight-forward test based on clear technical grounds (do I need an
>>> AR specific to this mix --> Yes, add a new work).
>>> Now it would be nice if we actually had a work to work remix AR to go
>>> with that...
>>
>> Make an RFC for it then ;)
>
> I would if I knew of an example.  All the remixes I've seen don't
> represent new works, but people keep claiming they exist.

This comes to mind:
Wiz Khalifa - Black and Yellow http://www.youtube.com/watch?v=UePtoxDhJSw
vs.
Wiz Khalifa - Black And Yellow [G-Mix] ft. Snoop Dogg, Juicy J &
T-Pain http://www.youtube.com/watch?v=nWAGLkyxQG0

>> ("Do I need an AR specific to this mix OR does this mix have its own
>> ISWC" would be my test though)
>
> I hadn't even heard of an ISWC until I saw it under works, and still
> haven't seen one.

They're unique identifiers for works, and you can see some here
http://iswcnet.cisac.org/iswcnet/
(those are mostly for US works though, France, Spain, etc, have their
own databases… but the ISWCs themselves are universal)

>>
> --
>
> Andii :-)
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style



-- 
Nicolás Tamargo de Eguren

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Calvin Walton]

On Mon, 2011-06-06 at 23:50 +0100, Dr Andrew John Hughes wrote:
> On 6 June 2011 18:22, Paul C. Bryan  wrote:
> > Sorry for jumping into the discussion so late in the game—real life has been
> > demanding much of my time lately.
> >
> > I would suggest the test could boil-down to whether a new artist work AR is
> > required to distinguish one remix from another (if a particular remix
> > contains new content, and has additional composer/lyricist/etc. credit). If
> > so, then a separate (but related) work would seem to be in order. If not,
> > then the original work would be used.
> >
> > Thoughts?
> >
> 
> Pretty much what I think too.  It has the benefit of being a very
> straight-forward test based on clear technical grounds (do I need an
> AR specific to this mix --> Yes, add a new work).
> Now it would be nice if we actually had a work to work remix AR to go
> with that...

One thing that concerns me a bit with a work→work remix ar: Remixes are
typically based on a specific recording of a work. How do we represent
this in Musicbrainz? Should we continue to use the recording→recording
AR alongside the work→work AR on any remixes which are based on a
specific recording, but also add e.g. new lyrics?

-- 
Calvin Walton 


___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Pete Marsh]

This makes my head hurt, but here's a couple of questions just to make sure I'm 
kind of understanding it

1) how do we ascertain that a remix has enough new content to make it a new 
work? (it's going to be the exception rather than the rule to have new 
lyricist/composer credits). it would seem to me that most contemporary remixes 
would fall under that category. some are almost indistinguishable from the 
original (Aphex Twin has been known to pass off completely original new works 
as remixes)

2) wouldn't the relationship be to a recording, not a work?  so the remix is a 
work extrapolated from a recording?

cheers

P

-Original Message-
From: musicbrainz-style-boun...@lists.musicbrainz.org 
[mailto:musicbrainz-style-boun...@lists.musicbrainz.org] On Behalf Of Calvin 
Walton
Sent: 07 June 2011 12:28
To: MusicBrainz Style Discussion
Subject: Re: [mb-style] Works and remixes/covers

On Mon, 2011-06-06 at 23:50 +0100, Dr Andrew John Hughes wrote:
> On 6 June 2011 18:22, Paul C. Bryan  wrote:
> > Sorry for jumping into the discussion so late in the game-real life 
> > has been demanding much of my time lately.
> >
> > I would suggest the test could boil-down to whether a new artist 
> > work AR is required to distinguish one remix from another (if a 
> > particular remix contains new content, and has additional 
> > composer/lyricist/etc. credit). If so, then a separate (but related) 
> > work would seem to be in order. If not, then the original work would be 
> > used.
> >
> > Thoughts?
> >
> 
> Pretty much what I think too.  It has the benefit of being a very 
> straight-forward test based on clear technical grounds (do I need an 
> AR specific to this mix --> Yes, add a new work).
> Now it would be nice if we actually had a work to work remix AR to go 
> with that...

One thing that concerns me a bit with a work→work remix ar: Remixes are 
typically based on a specific recording of a work. How do we represent this in 
Musicbrainz? Should we continue to use the recording→recording AR alongside the 
work→work AR on any remixes which are based on a specific recording, but also 
add e.g. new lyrics?

--
Calvin Walton 


___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Paul C. Bryan]

On Tue, 2011-06-07 at 07:28 -0400, Calvin Walton wrote:


> One thing that concerns me a bit with a work→work remix ar: Remixes are
> typically based on a specific recording of a work. How do we represent
> this in Musicbrainz? Should we continue to use the recording→recording
> AR alongside the work→work AR on any remixes which are based on a
> specific recording, but also add e.g. new lyrics?



My vote would be yes, they're basically parallel.
Work→Work AR for remix indicates work derivation—where remix requires a
distinct “work” attribution such as composer/lyricist/etc.
Recording→Recording indicates that one recording is derived from the
other (i.e. contains recorded material from the original).

Paul

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Paul C. Bryan]

On Tue, 2011-06-07 at 16:16 +0100, Pete Marsh wrote:


> This makes my head hurt, but here's a couple of questions just to make
> sure I'm kind of understanding it


Take 2 ARs and call us in the morning. ;-)


> 1) how do we ascertain that a remix has enough new content to make it
> a new work? (it's going to be the exception rather than the rule to
> have new lyricist/composer credits). it would seem to me that most
> contemporary remixes would fall under that category. some are almost
> indistinguishable from the original (Aphex Twin has been known to pass
> off completely original new works as remixes)


I suggest it be when the derivative work has work attributions that are
distinct from the original (e.g. additional composer or lyricist).


> 2) wouldn't the relationship be to a recording, not a work?  so the
> remix is a work extrapolated from a recording?


Yes, it would be too. It wouldn't be a remix if it did not contain
recordings from the original it is a remix of. The issue is that a remix
can contain "additional work" of another artist.

Paul
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Pete Marsh]

thanks Paul!
 
i still think the criteria for a remix becoming a work rather than a
recording need a bit more discussion. credits for additional lyrics or
compositions on such things are going to be the exception rather than
the rule. in fact i would argue that in the vast majority of cases
remixes will introduce additional material, yet very rarely is the
remixer given a composition credit. so my feeling is that by default a
remix a should be a new work (a different duration is usually a
giveaway).  those cases where we are merely dealing with just a
rebalancing of the material (ie a remix in the strict sense of the word)
then could be a recording-recording relationship (in the same way a
remastered recording might).

am i being too philosophical here?
:-) 
 


From: musicbrainz-style-boun...@lists.musicbrainz.org
[mailto:musicbrainz-style-boun...@lists.musicbrainz.org] On Behalf Of
Paul C. Bryan
Sent: 08 June 2011 23:46
To: MusicBrainz Style Discussion
Subject: Re: [mb-style] Works and remixes/covers


On Tue, 2011-06-07 at 16:16 +0100, Pete Marsh wrote:



This makes my head hurt, but here's a couple of questions just
to make sure I'm kind of understanding it



Take 2 ARs and call us in the morning. ;-)



1) how do we ascertain that a remix has enough new content to
make it a new work? (it's going to be the exception rather than the rule
to have new lyricist/composer credits). it would seem to me that most
contemporary remixes would fall under that category. some are almost
indistinguishable from the original (Aphex Twin has been known to pass
off completely original new works as remixes)



I suggest it be when the derivative work has work attributions that are
distinct from the original (e.g. additional composer or lyricist).



2) wouldn't the relationship be to a recording, not a work?  so
the remix is a work extrapolated from a recording?



Yes, it would be too. It wouldn't be a remix if it did not contain
recordings from the original it is a remix of. The issue is that a remix
can contain "additional work" of another artist.

Paul 

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Paul C. Bryan]

On Thu, 2011-06-09 at 15:48 +0100, Pete Marsh wrote:

> thanks Paul!
>  
> i still think the criteria for a remix becoming a work rather than a
> recording need a bit more discussion. credits for additional lyrics or
> compositions on such things are going to be the exception rather than
> the rule. in fact i would argue that in the vast majority of cases
> remixes will introduce additional material, yet very rarely is the
> remixer given a composition credit. so my feeling is that by default a
> remix a should be a new work (a different duration is usually a
> giveaway).  those cases where we are merely dealing with just a
> rebalancing of the material (ie a remix in the strict sense of the word)
> then could be a recording-recording relationship (in the same way a
> remastered recording might).
> 
> am i being too philosophical here?
> :-) 


You may be, but it's worthy of discussion IMO. Your position calls into
question the issue of covers then. By your definitions, covers should
also be different works:


  * different recording material 
  * rarely a different composition credit 
  * different duration


I find myself resistant to the idea that covers should be different
works. By the same philosophy, I am resistant to the idea that all
remixes should be different works. 

Paul


>  
> 
> 
> From: musicbrainz-style-boun...@lists.musicbrainz.org
> [mailto:musicbrainz-style-boun...@lists.musicbrainz.org] On Behalf Of
> Paul C. Bryan
> Sent: 08 June 2011 23:46
> To: MusicBrainz Style Discussion
> Subject: Re: [mb-style] Works and remixes/covers
> 
> 
> On Tue, 2011-06-07 at 16:16 +0100, Pete Marsh wrote:
> 
> 
> 
>   This makes my head hurt, but here's a couple of questions just
> to make sure I'm kind of understanding it
>   
> 
> 
> Take 2 ARs and call us in the morning. ;-)
> 
> 
> 
>   1) how do we ascertain that a remix has enough new content to
> make it a new work? (it's going to be the exception rather than the rule
> to have new lyricist/composer credits). it would seem to me that most
> contemporary remixes would fall under that category. some are almost
> indistinguishable from the original (Aphex Twin has been known to pass
> off completely original new works as remixes)
>   
> 
> 
> I suggest it be when the derivative work has work attributions that are
> distinct from the original (e.g. additional composer or lyricist).
> 
> 
> 
>   2) wouldn't the relationship be to a recording, not a work?  so
> the remix is a work extrapolated from a recording?
>   
> 
> 
> Yes, it would be too. It wouldn't be a remix if it did not contain
> recordings from the original it is a remix of. The issue is that a remix
> can contain "additional work" of another artist.
> 
> Paul 
> 
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style


___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Frederic Da Vitoria]

2011/5/20 Nikki 

> Aurélien Mino wrote:
> > It was planned to add "Arranger" at work level too after NGS migration,
> > exactly for this situation.
> > See
> >
> http://wiki.musicbrainz.org/Next_Generation_Schema/Track_Relationships_Conversion#Artist-Track
> >
> > It just has not been added yet.
>
> I thought the upgrades script would do it. :/ Oh well, I can add it now.
>
> Should something be added to the relationship type page about when to
> use which?
>

It's been more than 2 months, but a discussion with Jesus2099 made me
realize this: a new arranger should not be considered as a sufficient reason
to create a new Work. AFAIK the rules for considering 2 recordings are still
being discussed, but a new arranger is not a reason IMO.

-- 
Frederic Da Vitoria
(davitof)

Membre de l'April - « promouvoir et défendre le logiciel libre » -
http://www.april.org
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Alex Mauer]

On 08/05/2011 02:19 PM, Frederic Da Vitoria wrote:
> It's been more than 2 months, but a discussion with Jesus2099 made me
> realize this: a new arranger should not be considered as a sufficient reason
> to create a new Work. AFAIK the rules for considering 2 recordings are still
> being discussed, but a new arranger is not a reason IMO.

+1 to this.

“Orchestrator” is the same way, IMO.
—Alex Mauer “hawke”


___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Frederic Da Vitoria]

2011/8/5 Alex Mauer 

> On 08/05/2011 02:19 PM, Frederic Da Vitoria wrote:
> > It's been more than 2 months, but a discussion with Jesus2099 made me
> > realize this: a new arranger should not be considered as a sufficient
> reason
> > to create a new Work. AFAIK the rules for considering 2 recordings are
> still
> > being discussed, but a new arranger is not a reason IMO.
>

I should read myself more carefully. I meant "AFAIK the rules for
considering 2 Recordings as one or as 2 different Works are still being
discussed, but a new arranger is not a reason IMO."

-- 
Frederic Da Vitoria
(davitof)

Membre de l'April - « promouvoir et défendre le logiciel libre » -
http://www.april.org
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Calvin Walton]

On Fri, 2011-08-05 at 21:19 +0200, Frederic Da Vitoria wrote:
> 2011/5/20 Nikki 
> Aurélien Mino wrote:
> > It was planned to add "Arranger" at work level too after NGS 
> migration,
> > exactly for this situation.
> > See
> > 
> http://wiki.musicbrainz.org/Next_Generation_Schema/Track_Relationships_Conversion#Artist-Track
> >
> > It just has not been added yet.


> It's been more than 2 months, but a discussion with Jesus2099 made me
> realize this: a new arranger should not be considered as a sufficient
> reason to create a new Work. AFAIK the rules for considering 2
> recordings are still being discussed, but a new arranger is not a
> reason IMO.

One of the worst cases for this issue regarding whether arrangements are
new works comes in with a particular subgenre of game remix music -
Touhou arrange music.

An example of a work where this occurs is
http://musicbrainz.org/work/255e9094-da2f-3f1f-bc5b-9258a2a630d7
(Note that the original version of this song was a midi file included
with the game.) Take a look at the artist 'ZUN' for more.

Many of the linked recordings are instrumental arrangements of
instrumental songs. Most are recognizably the same song. Some are
difficult to tell apart. Some sound like they might as well be new songs
- except that the arranger says that it was based on a particular
original track. Sometimes the arrangement is based on multiple original
tracks.

Some of the linked recordings add vocals. Due to the way works on
musicbrainz are set up, all of the arrange tracks that have vocals would
have to be added as a new work.

Due to a bug in the NGS conversion, if a cover version of a song
included a lyrics AR that the original version didn't have, a new work
would be created for the cover, and the cover recording would be linked
to /both/ the new work and the original work. An example of where this
happened is 
http://musicbrainz.org/recording/99da4375-1b86-4dac-96d3-f3dbe7883d21
(This is a new work "Now & Forever" based on two separate original
instrumental works "今昔幻想郷 ～ Flower Land" and "眠れる恐怖 ～
Sleeping Terror", and had lyrics added as well. It needs to be cleaned
up.)

-- 
Calvin Walton 


___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[SwissChris]

We should clearly keep classical out of this debate, since "arrangement" in
classical also stands for everything we'd call "cover performance" in other
music styles. In Classical the ongoing discussion whether a specific
arrangement or orchestration qualifies as a new derivative work tends to ask
for at least two distinct recordings. For non-classical music I could
imagine that one of the criteria for such a decision could be: Are there (or
could there be) *cover performances* of this specific version? Only then
should a new "arrangement" or "orchestration" or "remix" qualify as a new
work. Thoughts?

On Fri, Aug 5, 2011 at 10:19 PM, Calvin Walton wrote:

> On Fri, 2011-08-05 at 21:19 +0200, Frederic Da Vitoria wrote:
> > 2011/5/20 Nikki 
> > Aurélien Mino wrote:
> > > It was planned to add "Arranger" at work level too after NGS
> migration,
> > > exactly for this situation.
> > > See
> > >
> http://wiki.musicbrainz.org/Next_Generation_Schema/Track_Relationships_Conversion#Artist-Track
> > >
> > > It just has not been added yet.
>
>
> > It's been more than 2 months, but a discussion with Jesus2099 made me
> > realize this: a new arranger should not be considered as a sufficient
> > reason to create a new Work. AFAIK the rules for considering 2
> > recordings are still being discussed, but a new arranger is not a
> > reason IMO.
>
> One of the worst cases for this issue regarding whether arrangements are
> new works comes in with a particular subgenre of game remix music -
> Touhou arrange music.
>
> An example of a work where this occurs is
> http://musicbrainz.org/work/255e9094-da2f-3f1f-bc5b-9258a2a630d7
> (Note that the original version of this song was a midi file included
> with the game.) Take a look at the artist 'ZUN' for more.
>
> Many of the linked recordings are instrumental arrangements of
> instrumental songs. Most are recognizably the same song. Some are
> difficult to tell apart. Some sound like they might as well be new songs
> - except that the arranger says that it was based on a particular
> original track. Sometimes the arrangement is based on multiple original
> tracks.
>
> Some of the linked recordings add vocals. Due to the way works on
> musicbrainz are set up, all of the arrange tracks that have vocals would
> have to be added as a new work.
>
> Due to a bug in the NGS conversion, if a cover version of a song
> included a lyrics AR that the original version didn't have, a new work
> would be created for the cover, and the cover recording would be linked
> to /both/ the new work and the original work. An example of where this
> happened is
> http://musicbrainz.org/recording/99da4375-1b86-4dac-96d3-f3dbe7883d21
> (This is a new work "Now & Forever" based on two separate original
> instrumental works "今昔幻想郷 ～ Flower Land" and "眠れる恐怖 ～
> Sleeping Terror", and had lyrics added as well. It needs to be cleaned
> up.)
>
> --
> Calvin Walton 
>
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Frederic Da Vitoria]

2011/8/5 SwissChris 

> We should clearly keep classical out of this debate, since "arrangement" in
> classical also stands for everything we'd call "cover performance" in other
> music styles. In Classical the ongoing discussion whether a specific
> arrangement or orchestration qualifies as a new derivative work tends to ask
> for at least two distinct recordings. For non-classical music I could
> imagine that one of the criteria for such a decision could be: Are there (or
> could there be) *cover performances* of this specific version? Only then
> should a new "arrangement" or "orchestration" or "remix" qualify as a new
> work. Thoughts?


None from me, sorry: not being English, I don't really understand what
"cover" means in English :-/ Which could mean that this is not the ultimate
way: if I don't really understand, probably many non-English users will
hesitate too.

-- 
Frederic Da Vitoria
(davitof)

Membre de l'April - « promouvoir et défendre le logiciel libre » -
http://www.april.org
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[SwissChris]

On Fri, Aug 5, 2011 at 11:44 PM, Frederic Da Vitoria wrote:

> 2011/8/5 SwissChris 
>
>> We should clearly keep classical out of this debate, since "arrangement"
>> in classical also stands for everything we'd call "cover performance" in
>> other music styles. In Classical the ongoing discussion whether a specific
>> arrangement or orchestration qualifies as a new derivative work tends to ask
>> for at least two distinct recordings. For non-classical music I could
>> imagine that one of the criteria for such a decision could be: Are there (or
>> could there be) *cover performances* of this specific version? Only then
>> should a new "arrangement" or "orchestration" or "remix" qualify as a new
>> work. Thoughts?
>
>
> None from me, sorry: not being English, I don't really understand what
> "cover" means in English :-/ Which could mean that this is not the ultimate
> way: if I don't really understand, probably many non-English users will
> hesitate too.


Here's what the french wiki has for "reprise", which is the translation of
"cover version". The article exists in about two dozen languages…

>
>
> --
> Frederic Da Vitoria
> (davitof)
>
> Membre de l'April - « promouvoir et défendre le logiciel libre » -
> http://www.april.org
>
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Ryan Torchia]

On Fri, Aug 5, 2011 at 2:10 PM, SwissChris  wrote:

> We should clearly keep classical out of this debate, since "arrangement" in
> classical also stands for everything we'd call "cover performance" in other
> music styles. In Classical the ongoing discussion whether a specific
> arrangement or orchestration qualifies as a new derivative work tends to ask
> for at least two distinct recordings. For non-classical music I could
> imagine that one of the criteria for such a decision could be: Are there (or
> could there be) *cover performances* of this specific version? Only then
> should a new "arrangement" or "orchestration" or "remix" qualify as a new
> work. Thoughts?


That distinction doesn't only exist between Classical and non-Classical.
"Cover" is mostly a pop music term, it has meant different things at
different times.  The earliest examples were from competing record companies
trying to capitalize on hit songs on the other labels by recording their own
versions.  It's not too far from how the term is currently used, but it
happened much faster and was driven by commercial intrerests rather than any
kind of respect for the song.  It's not really clear though if the actual
term "cover" came from this practice; more likely it came in the 50s and
60s.  By then the practice was more oriented around race.  The term "cover"
referred more to having a white artist record a version of a song by a black
artist, resulting in... pieces of...wonderfulness...such as
this.
(Have ear bleach handy.)

The different recordings were often released around the same time, and
sometimes the "cover" was even released first.  The term "cover" didn't
refer to a later version covering over an earlier version, it was a
reference to covering over the black artist's performance with a white
artist's perfomance.  This two main effects of this were that the label
would make a lot of money off the "white" version (which usually sold much
better -- and yes, the cover in the youtube video above charted at the same
time, and *higher, *than the original), and that it helped segregate white
radio and sales charts from black radio and sales charts.

Gradually after this practice (thankfully) died off, "cover" came to be used
like it is today, to mean recording a version of a song that's already been
recorded, usually well after its original release.  It obviously doesn't
carry the racial overtones it did 60 years ago.

As to the question, the one thing about covers that hasn't changed (as far
as I know) is that a cover of a copyrighted song is always credited to the
songwriter (and they're paid royalties).  I'm not sure if that should be the
only criteria -- it might not be true worldwide -- but it might be helpful
in settling some disputes here since, at the very least, it's objective.

I'm really not sure we can use the "cover" attribute to link to somebody
else's (non-original) version of the song.  While it's clear sometimes that
the performance is a "cover" of another performance, the term has
historically referred to the song itself.  Maybe another term like "remake"
could be used to indicate that situation, or maybe it's just not something
we can track.  Unless the artist confirms they were remaking a specific
version, it seems like that would all be based on speculation anyway.

--Torc.
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Frederic Da Vitoria]

2011/8/6, SwissChris :
> On Fri, Aug 5, 2011 at 11:44 PM, Frederic Da Vitoria
> wrote:
>
>> 2011/8/5 SwissChris 
>>
>>> We should clearly keep classical out of this debate, since "arrangement"
>>> in classical also stands for everything we'd call "cover performance" in
>>> other music styles. In Classical the ongoing discussion whether a
>>> specific
>>> arrangement or orchestration qualifies as a new derivative work tends to
>>> ask
>>> for at least two distinct recordings. For non-classical music I could
>>> imagine that one of the criteria for such a decision could be: Are there
>>> (or
>>> could there be) *cover performances* of this specific version? Only then
>>> should a new "arrangement" or "orchestration" or "remix" qualify as a new
>>> work. Thoughts?
>>
>> None from me, sorry: not being English, I don't really understand what
>> "cover" means in English :-/ Which could mean that this is not the
>> ultimate
>> way: if I don't really understand, probably many non-English users will
>> hesitate too.
>
> Here's what the french wiki has for "reprise", which is the translation of
> "cover version". The article exists in about two dozen languages…

No it isn't the translation, just check the article lengths in the two
languages :-) It is the matching article, but I don't trust much this:
the meanings of a word and it's "official" translation often differ
and understanding the differences is often the key to understanding
the original meaning (as neglecting those differences is often the
cause of complete misunderstandings between people of different
languages). I understand the general concept of cover, but I was not
convinced that I knew enough to answer here.

If we are limiting ourselves to the primary sense of "later
interpretation", then I fear that your definition is not specific
enough. Answers would differ depending on whether the speaker is
someone who barely knows music, has a good musical ear, is a
specialist of this kind of music, is himself a performer, or even
performed himself this particular song. For example two jazz covers of
a rock song could be considered as almost identical by someone who
never listens to jazz but completely different by a jazz specialist.
Just as two rock covers of a classical work would be considered
identically horrible by someone who only listens to classical, but
could be actually completely different :-)

I guess we'll have to accept that the granularity of works hierarchy
will be different from anything any user would like. If I use myself
as a reference (I have difficulties expressing this in general terms),
in some domains we will merge some Works which I would rather
separate, but mostly I believe there will be more Works than I would
need because for example two rap or metal covers will be considered as
completely different while they'll sound identical to my uneducated
ear.

-- 
Frederic Da Vitoria
(davitof)

Membre de l'April - « promouvoir et défendre le logiciel libre » -
http://www.april.org

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[jesus2099]

+1
for keeping remixes, covers and other versions the same 1 work.

-
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/Works-and-remixes-covers-tp3531609p3726979.html
Sent from the Musicbrainz - Style mailing list archive at Nabble.com.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[MeinDummy]

Lukáš Lalinský wrote:
> 
> The original NGS documentation which was written by me [1] says that
> works with different remixes should be merged together. The Work page
> [2] however says that they should be kep separate. All people I've
> discussed this with agree that they should be merged. Can I change the
> Work page to remove the mention of remixes?
> 
> Lukas
> 
> [1] http://wiki.musicbrainz.org/Next_Generation_Schema#Work
> [2] http://wiki.musicbrainz.org/Work
> 
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@.musicbrainz
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
> 

This discussion seems to have faded out now.
There are many unresolved specific issues regarding covers, translations,
instrumentals, ... Most of them are beyond the remix work topic and should
probably be further discussed in separate threads.
Despite of those issues isn't the guideline update that Lukáš proposed ready
for RFC/RFV now?

Christian (MD)

--
View this message in context: 
http://musicbrainz.1054305.n4.nabble.com/Works-and-remixes-covers-tp3531609p4019048.html
Sent from the Musicbrainz - Style mailing list archive at Nabble.com.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Lukáš Lalinský]

On Wed, Nov 9, 2011 at 9:30 AM, MeinDummy  wrote:
> This discussion seems to have faded out now.
> There are many unresolved specific issues regarding covers, translations,
> instrumentals, ... Most of them are beyond the remix work topic and should
> probably be further discussed in separate threads.
> Despite of those issues isn't the guideline update that Lukáš proposed ready
> for RFC/RFV now?

I think adding remixes and covers to the same work as the original
version is now a common practice. Not sure if it makes sense to
explicitly state it somewhere.

Lukas

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[MeinDummy]

Lukáš Lalinský wrote:
> 
> On Wed, Nov 9, 2011 at 9:30 AM, MeinDummy  wrote:
>> This discussion seems to have faded out now.
>> There are many unresolved specific issues regarding covers, translations,
>> instrumentals, ... Most of them are beyond the remix work topic and
>> should
>> probably be further discussed in separate threads.
>> Despite of those issues isn't the guideline update that Lukáš proposed
>> ready
>> for RFC/RFV now?
> 
> I think adding remixes and covers to the same work as the original
> version is now a common practice. Not sure if it makes sense to
> explicitly state it somewhere.
> 
> Lukas
> 
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@.musicbrainz
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
> 

We still have that guideline which explicitly states the opposite:
http://wiki.musicbrainz.org/Work
And I still see editors and voters referring to that guideline every now and
then.
Specifically because of that common practice I think a guideline update is
overdue.

Christian (MD)

--
View this message in context: 
http://musicbrainz.1054305.n4.nabble.com/Works-and-remixes-covers-tp3531609p4019172.html
Sent from the Musicbrainz - Style mailing list archive at Nabble.com.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Nicolás Tamargo de Eguren]

On Wed, Nov 9, 2011 at 11:24 AM, MeinDummy  wrote:

>
> Lukáš Lalinský wrote:
> >
> > On Wed, Nov 9, 2011 at 9:30 AM, MeinDummy  wrote:
> >> This discussion seems to have faded out now.
> >> There are many unresolved specific issues regarding covers,
> translations,
> >> instrumentals, ... Most of them are beyond the remix work topic and
> >> should
> >> probably be further discussed in separate threads.
> >> Despite of those issues isn't the guideline update that Lukáš proposed
> >> ready
> >> for RFC/RFV now?
> >
> > I think adding remixes and covers to the same work as the original
> > version is now a common practice. Not sure if it makes sense to
> > explicitly state it somewhere.
> >
> > Lukas
> >
> > ___
> > MusicBrainz-style mailing list
> > MusicBrainz-style@.musicbrainz
> > http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
> >
>
> We still have that guideline which explicitly states the opposite:
> http://wiki.musicbrainz.org/Work
> And I still see editors and voters referring to that guideline every now
> and
> then.
> Specifically because of that common practice I think a guideline update is
> overdue.
>

While I agree that remixes in the classic sense are usually (always?) the
same work, I won't accept a wording that says simply "all remixes should be
merged into the original work". A lot of things called "remix" involve a
completely new music or new lyrics, especially in hip-hop. Also, I'd leave
them separate if they have different ISWCs, but that's normally not the
case for classic-sense remixes :)  But yeah, if the wording is made in a
way that handles these cases, I think a new "what defines an unique work"
is due.

Christian (MD)
>
> --
> View this message in context:
> http://musicbrainz.1054305.n4.nabble.com/Works-and-remixes-covers-tp3531609p4019172.html
> Sent from the Musicbrainz - Style mailing list archive at Nabble.com.
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style
>



-- 
Nicolás Tamargo de Eguren
___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[MeinDummy]

Nicolás Tamargo de Eguren wrote:
> 
> While I agree that remixes in the classic sense are usually (always?) the
> same work, I won't accept a wording that says simply "all remixes should
> be
> merged into the original work". A lot of things called "remix" involve a
> completely new music or new lyrics, especially in hip-hop. Also, I'd leave
> them separate if they have different ISWCs, but that's normally not the
> case for classic-sense remixes :)  But yeah, if the wording is made in a
> way that handles these cases, I think a new "what defines an unique work"
> is due.
> -- 
> Nicolás Tamargo de Eguren
> 

The proposal is just to remove the mention of remixes from
http://wiki.musicbrainz.org/Work . There won't be something like "all
remixes should be merged into the original work".
But I agree that we need a place where handling of remixes is described in
more detail. "what defines an unique work" seems to be just right for that
(as a new section of the page, to be discussed in one or several separate
RFCs ???).

Christian (MD)

--
View this message in context: 
http://musicbrainz.1054305.n4.nabble.com/Works-and-remixes-covers-tp3531609p4019239.html
Sent from the Musicbrainz - Style mailing list archive at Nabble.com.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[jesus2099]

Lukáš Lalinský wrote:
> Can I change the Work page to remove the mention of remixes?

+1

-
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/Works-and-remixes-covers-tp3531609p4020364.html
Sent from the Musicbrainz - Style mailing list archive at Nabble.com.

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Nicolás Tamargo de Eguren]

On Wed, Nov 9, 2011 at 6:30 PM, jesus2099  wrote:
>
> Lukáš Lalinský wrote:
>> Can I change the Work page to remove the mention of remixes?
>
> +1

Nobody seemed to oppose the change, so I've done it.

> -
> 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/Works-and-remixes-covers-tp3531609p4020364.html
> Sent from the Musicbrainz - Style mailing list archive at Nabble.com.
>
> ___
> MusicBrainz-style mailing list
> MusicBrainz-style@lists.musicbrainz.org
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style



-- 
Nicolás Tamargo de Eguren

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style

Re: [mb-style] Works and remixes/covers

[Per Øyvind Øygard]

2011/11/9 Lukáš Lalinský :
> On Wed, Nov 9, 2011 at 9:30 AM, MeinDummy  wrote:
>> This discussion seems to have faded out now.
>> There are many unresolved specific issues regarding covers, translations,
>> instrumentals, ... Most of them are beyond the remix work topic and should
>> probably be further discussed in separate threads.
>> Despite of those issues isn't the guideline update that Lukáš proposed ready
>> for RFC/RFV now?
>
> I think adding remixes and covers to the same work as the original
> version is now a common practice. Not sure if it makes sense to
> explicitly state it somewhere.

If we're to merge remixes with originals we need a remix attribute for
recording->work relationships like we have for covers. Saying a remix
is a performance of a work seems inaccurate at best.

-- 
Per / Wizzcat

___
MusicBrainz-style mailing list
MusicBrainz-style@lists.musicbrainz.org
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-style