El 23/06/2008, a las 12:10, Mark Taylor escribió:
> I think one general point which may be appropriate to make in some
> cases is that application interoperability is not strictly VO
> business in
> the sense that, say, DAL protocols are. Despite the IVOA provenance,
> developers of applications which do not view themselves as virtual
> observatory tools should not feel that this technology is not for
> them.
I completely agree with this view.
However, we have to keep in mind that some particular data models are concerned with "data provenance", and that keeping metadata, specially "VO provenance metadata", should be something messaging protocols should be able to do.
ps. I'm using "data provenance" in an ambiguous way, meaning both "data source" provenance and instrumental provenance. The latter is probably kept in the binary data, so sending messages with such datasets trivially keeps it, while the data source of a data providing message should be kept around maybe as a parameter for all rows of the message, or a property of the message.
-- Juan de Dios Santander Vela Diplomado en CC. Físicas, Ingeniero en Electrónica Doctorando en Tecnologías Multimedia Becario Predoctoral del Instituto de Astrofísica de Andalucía Miguel de Cervantes Saavedra: No hay libro tan malo que no contenga algo bueno.Received on 2008-06-27Z14:18:26