draft data sources schema

[ThePeach]

The main data sources are:

• users
• media contents (images, videos, music, flash?)

This schema should be then used to define the protocol used by the client-side application.

[ThePeach]

found that there's a very interesting metadata description set called VRA currently at version 4 (http://www.vraweb.org/projects/vracore4/), its schema is available at:
http://www.loc.gov/standards/vracore/vra.xsd
and
http://www.loc.gov/standards/vracore/vra-strict.xsd

Still the big problem here is whether the data has to be stored in a single file or not, being the time to read quite a massive problem, on the other hand, searching through split xmls can cause a couple of headaches :(

[ThePeach]

further development:
turns out there's not linked data for the VRA metadata, see http://linglibrit.wordpress.com/2010/12/28/vra-core-4-0-to-simple-dublin-core-conversion-stylesheet/.
The solutions would be to have a super structure in RDF representing the users and their relationship with the images which can be considered leaves in a root based graph (aka tree); while implementing the tree leaves on a later stage into RDF giving us the opportunity to avoid particular mangling on the data to extract the immediate information, and on the opposite side, remove the possibility to perform particular queries on the given data.
This way the image will link to the work record and the image record defined by VRA and the User, Collection and Images are connected between each other with RDF triples.

[ThePeach]

Looks like ARC (https://github.com/semsol/arc2) is a good enough engine to provide reasoning over RDF stuff, it's very nice and simple, only problem it works only on MySql storage engine.