One comment following from the GUID meeting is that I believe we need to get serious about being able to represent our data models in RDF. This means that the goal should indeed be for a "flat" (Darwin Core like) standard. If there are elements which hold nested complexity that we wish to represent, we should recognise that these are probably separate data objects which should be modeled as separable components (with their own "flat" standards). The top level object can then have a property whose value is the identifier for one of the lower level components. Even if we choose to compose rich documents with entire trees of object relationships, the underlying model should make these separations clear.
I am concerned about this, to me it seems to block the way into a future for scientific data exchange.
Flat list is great for advertising, (see e.g. DublinCore, DarwinCore or RSS), but not really for exchanging complex data.
Completely forbidding any object aggregation would mean that each measurement of a specimen has a stand-alone resource... Or each author in a publication (because authors have addresses, so they can not be just a list, the list elements have structure).
Where can we discuss this issue which goes far beyond the TDWG-LIT?
Gregor---------------------------------------------------------- Gregor Hagedorn (G.Hagedorn@bba.de) Institute for Plant Virology, Microbiology, and Biosafety Federal Research Center for Agriculture and Forestry (BBA) Königin-Luise-Str. 19 Tel: +49-30-8304-2220 14195 Berlin, Germany Fax: +49-30-8304-2203