[mb-style] RFC-321: Work parts relationship

[Frederic Da Vitoria]

2011/6/7, Maurits Meulenbelt <maurits.meulenbelt at gmail.com>:
> I don't know if this has been mentioned before (in that case I've missed
> it), but I think we should be careful about inheriting ARs from the
> aggregate work to the subworks.
> Some works are left unfinished by a composer and finished by another,
> which leads to different composers on different parts of a work (Mozarts
> Requiem is a well known but little disputed example:
> http://en.wikipedia.org/wiki/Requiem_%28Mozart%29 ).
> Other works have instrumental parts that don't need a librettist.
>
> Maurits Meulenbelt (mfmeulenbelt)
>
> Op 7-6-2011 7:53, caramel schreef:
>> 2011/6/6 Simon Reinhardt <simon.reinhardt at koeln.de
>> <mailto:simon.reinhardt at koeln.de>>
>>
>>     Hi
>>
>>
>>     caramel wrote:
>>
>>         This AR could be useful but mostly when there are only few
>>         works related together. I would prefer the concept of "Work
>>         group" instead with the possibility for the works to inherit
>>         the ARs at the work group level.
>>         For CSG, the number of sub-works can be several tens.
>>
>>
>>     I'd prefer relationships to a grouping type. What's the problem
>>     with having loads of sub-works?
>>
>> St-John Passion (Bach)  can be split into 68 named subworks, 78 for
>> "St-Matthew Passion", 21 for "Sports et divertissements" from Erik
>> Satie,...
>> All individual sub-works of a composition or all works of a collection
>> have the same ARs (composer, lyricist, score, web site link...)
>> It is obviously faster to set the ARs at a work group than at a work
>> level and after add additional ARs to link them together. And as
>> mentioned by Symphonik, it would be good to have the possibility to
>> order the subworks... in fact to get a work list as we do it for the
>> track lists.
>>
>> The concept of a classical composition split into sub-works or works
>> in a collection is missing in the actual NGS but I really think that
>> it is required if we want to fix and maintain the millions of records
>> of the database.
>>
>> A non-order work group is useful too for cover and remixes or
>> different orchestrations when additional works are created. Individual
>> ARs are useful too but not usable when the number of relations exceeds
>> several items.

What about manual inheritance? I suggest an option to copy the ARs
from a main Work to the sub-Works or from a sub-Work to the main Work.
The user could be shown a list of current ARs about to be copied.
Maybe he could even uncheck the ARs he thinks shouldn't be copied. Of
course there could still be errors, but these would be human errors,
and even with completely manual duplication, human errors are bound to
happen.
-- 
Frederic Da Vitoria
(davitof)

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