> Suppose that we throw _all_ the data formats out of the VOTable spec
> s.t. it is just a metadata format for describing tables in external
> files...and we keep it general enough that it can describe all 2D
> tables.
>
> Suppose we then define another format for tables that are all XML (by
> implication small tables) that is self-describing to the level of
> current VOTable but also parsable by "the standard XML tools".
>
> Suppose we make sure that VOTable can describe this new format.
>
> Are we OK then?
Yes, this would work. However we also need to define a format for spectral and time series data (at a minimum) as well as "normal" catalogues, or our nice new shiny format will quickly be overloaded with lots of nasty syntax tacked on so that it can be used to describe these as well.
Al.
-- Dr. A. Allan, School of Physics, University of ExeterReceived on 2004-01-19Z16:24:37