Discography YAML Schema Proposal

[push...@gmail.com]

Hi all,

While working on Pushtape, an open source Drupal distribution for building music websites, I came across a feature request to import/export discographies. I know there are some existing specs based around XML, JSON, Microdata, but felt it might be useful to come up with a self contained file format for discographies using YAML (a superset of JSON), which would allow a discography.yaml file to be modified by both human and machines. This would allow an artist to manage their discography either through software or just by manually editing the file, but allow various software (music players, discography databases) to read and parse the information. I posted more about my reasoning for YAML here.

I posted a quick example implementation over on github:

https://github.com/zirafa/discography-yaml/blob/master/discography.yaml

Any feedback is appreciated. Thanks!

zirafa

[Bo Ferri]

Hi zirafa,

interesting pragmatic proposal. I would recommend to extend or refactor
the credits and notes section a bit to make these parts a bit more
expressive for machine consumption (free text parts are always more
difficult to parse).
You maybe have a look at the comprehensive example [1] of the Music
Ontology for inspiration ;)

Cheers,


Bo


PS: it would be cool, if you could list the Music Ontology in your blog
post as well :)


[1]
http://wiki.musicontology.com/index.php/Mo_-_levels_of_abstraction_-_whole_way_down

On 2/21/2013 8:38 AM, push...@gmail.com wrote:
> Hi all,
>

> While working on Pushtape <http://www.pushtape.com>, an open source

> Drupal distribution for building music websites, I came across a feature
> request to import/export discographies. I know there are some existing
> specs based around XML, JSON, Microdata, but felt it might be useful to
> come up with a self contained file format for discographies using YAML

> <http://www.yaml.org/start.html> (a superset of JSON), which would allow

> a discography.yaml file to be modified by both human and machines. This
> would allow an artist to manage their discography either through
> software or just by manually editing the file, but allow various
> software (music players, discography databases) to read and parse the
> information. I posted more about my reasoning for YAML here.

> <http://zirafaworks.com/note/making-music-portable-yaml-discography-schema-proposal>

>
> I posted a quick example implementation over on github:
>
> https://github.com/zirafa/discography-yaml/blob/master/discography.yaml
>
> Any feedback is appreciated. Thanks!
>
> zirafa
>
>

> --
> You received this message because you are subscribed to the Google
> Groups "Music Ontology Specification Group" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to music-ontology-specific...@googlegroups.com.
> To post to this group, send email to
> music-ontology-sp...@googlegroups.com.
> Visit this group at
> http://groups.google.com/group/music-ontology-specification-group?hl=en.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

[Pushtape]

Thanks, updated my post. You make a good point about having finer
granularity on credits and notes. It's useful having these be text
blobs because it invites the user a bit of freedom to write in a more
conversational style (similar to liner notes included with
recordings), but I can see the benefit of having arrangers, engineers,
etc as more structured elements. It is a bit of a challenge for this
application, but in general it would seem useful to specify both a
minimal type template and a more comprehensive one (or some method of
extending the base template).