content, format, ctype, or xtype ?

[Mark Taylor]

On Mon, 11 May 2009, Mark Taylor wrote:

> The usefulness of the proposal is not specific to TAP, but I believe 
> it can address some of the data type flagging issues which have 
> been raised in relation to TAP, which is why I brought it up in the
> context of this discussion (http://www.ivoa.net/forum/dal/0903/1073.htm).
> It does not preclude using utypes appropriately, though I agree with 
> earlier posters that pass-through is the main thing that TAP should 
> be aiming for with utypes.

I'll just expand on my own comment slightly:

Specifically, the job that ctype (or whatever we call it) does for
TAP is to bridge the gap between the RDBMS type system and the VOTable
type system.  The former is slightly richer than the latter in that
it can contain things like POINT, REGION, TIMESTAMP, and some 
DB-specific items (PostgreSQL:MAC).  An alternative approach would
be to extend the VOTable type system to include these, but I believe
that would be the wrong way to go, since VOTable is and should be
a low-level container; changing the VOTable type system would
require upheaval in a lot of infrastructure software, and it's
really not necessary.

Mark

-- 
Mark Taylor   Astronomical Programmer   Physics, Bristol University, UK
m.b.taylor at bris.ac.uk +44-117-928-8776 http://www.star.bris.ac.uk/~mbt/