Re: One question about "more expressiveness"

[Yves Raimond]

Hello!

On Tue, Mar 22, 2011 at 5:40 PM, Marcelo Miranda
<j.marcel...@gmail.com> wrote:
> Hi, I’m new in music ontology.
>
> "The Music Ontology is divided in three levels of expressiveness - from the
> simplest one to the more complex one. Everything is clustered around the
> following categories:
>     - Level 1: aims at providing a vocabulary for simple editorial
> information (tracks/artists/releases, etc.)
>     - Level 2: aims at providing a vocabulary for expressing the music
> creation workflow (composition, arrangement, performance, recording, etc.)
>     - Level 3: aims at providing a vocabulary for complex event
> decomposition, to express, for example, what happened during a particular
> performance, what is the melody line of a particular work, etc."
>
>  The musical information types and their description forms go beyond the
> internal characteristics traditionally defined (MCLANE, 1996; DOWNIE, 2003)
> as they incorporate association with people, with extra-musical informations
> and with other musics. Music Ontology embraces these description forms.
>
>  “The viewpoint makes the object”. (Saussure).
>
> "The epigraph above is an indication of how the question about the
> information works, indicating, upward the object complexity and fluidity, a
> viewpoint that assumes that every scientific object is a construction
> operated by the researcher and, in special, when this is about social and
> human phenomena. However, this construction isn’t arbitrary. It becomes
> possible only because of the social-historical realities, ie, by considering
> not only the subjects, their practices and their representations, but also
> their structures and the situations they’re involved." (Marteleto, 1995).
>
> So, my question is: Wouldn’t be missing more than one level to provide a
> vocabulary for expressing information and feelings experienced by the user?
> For example, “this song makes me feel sad”, or “this music reminds me of a
> beach”, or “this song puts my dad in my mind”.

I don't think this has been fully tackled before - so research in this
area would definitely be welcome!

Best,
y

>
> Research to meet this demand would be welcome?
>
> I understand that these attributes should be availabe for the users, so they
> could enroll these information elements.
>
> Cheers,
> José Marcelo
>
>

[Bob Ferris]

Hi Marcelo,

Am 28.03.2011 11:17, schrieb Yves Raimond:
>> So, my question is: Wouldn�t be missing more than one level to provide a

>> vocabulary for expressing information and feelings experienced by the user?

>> For example, �this song makes me feel sad�, or �this music reminds me of a
>> beach�, or �this song puts my dad in my mind�.

Such knowledge representations are not really in the scope of the Music
Ontology. They address a more orthogonal issue. I tried to capture such
things with the Association Ontology [1]. This ontology utilizes the
association concept [3] of the Similarity Ontology [4].
To express personal feelings you can utilize ao:mood [2], although, one
might tend to vote for a more specific mo:mood. I hope the Association
Ontology is able to express your concerned knowledge representation.
Don't hesitate to ask further questions, if something is unclear to you.

Cheers,

Bob

[1] http://purl.org/ontology/ao/core#
[2] http://purl.org/ontology/ao/associationontology.html#mood
[3] http://kakapo.dcs.qmul.ac.uk/ontology/musim/0.2/musim.html#Association
[4] http://purl.org/ontology/similarity/

[Bob Ferris]

Hi,

I created a quite comprehensive example [1] (in its basic structure)
that should represent a kind of reference example to illustrate how one
can utilize the Music Ontology and related ontologies to cover each
level of abstraction that is introduced due its inheritance from FRBR.

Please, let me know what you think about this example and especially
recommend enhancements. If the initial basic structure is accepted, then
I will continue to add some more information that is already available
at the different refered websites from Discogs, MusicBrainz, AllMusic
and DBPedia.

Cheers,

Bob

PS: Besides, this example should especially cover the raised
issues/problems with the modelling of releases/release events. Thereby,
I tried to cover different possible use cases.

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