Updated VOSI schema files

[Mark Taylor]

On Mon, 22 Jan 2018, Markus Demleitner wrote:

> > As a separate issue: the document currently residing at
> > http://ivoa.net/xml/VOSITables/VOSITables-v1.0.xsd and hence
> > http://www.ivoa.net/xml/VOSITables/v1.0 doesn't look right to me:
> > I'd expect to see the schema from A.3 of VOSI 1.0,
> > but it's different, e.g. the top-level xsd:schema/@targetNamespace is
> > http://www.ivoa.net/xml/VOSICapabilities/v1.0 rather than
> > http://www.ivoa.net/xml/VOSITables/v1.0.
> > Am I going mad?  (quite possible - I feel my sanity draining
> > every time I think about XML namespaces).
> 
> Well, I might feel compelled to defend namespaces as such -- we just
> shouldn't have included minor version numbers in them.

[I wasn't railling against XML namespaces only cursing my own
mental feebleness]

> But that's for another day -- the particular riddle you encountered
> has an easy solution. Essentially, we didn't have a Erratum process
> in place when you noted the schema in VOSI 1.0 in the document is
> wrong and contradicts the spec -- see
> http://mail.ivoa.net/pipermail/grid/2014-October/002677.html.  As a
> consequence, the schema in the repo was updated with what's there now
> and what's correct for VOSI 1.0, while the document text remained
> unmodified and still contradicts itself.

Are you sure?  It looks to me like that's exactly what hasn't happened;
the schema at http://www.ivoa.net/xml/VOSITables/VOSITables-v1.0.xsd
(still) does not match the text from VOSI 1.0 Appendix A.3, it still
targets the VOSICapabilities namespace.  Your GWS list email cited
above says:

Markus wrote to grid at ivoa.net on 29 Oct 2014:

> Unless anybody comes up with a reason why not in the next few days,
> I plan to ask the IVOA Document Coordinator to replace the existing
> copy on the IVOA web site with a version cut'n'pasted from the
> VOSI appendix, as per the attachment to this message.

I don't know whether you did make that request to ivoadoc, but it
doesn't look like any such change was actually performed.

--
Mark Taylor   Astronomical Programmer   Physics, Bristol University, UK
m.b.taylor at bris.ac.uk +44-117-9288776  http://www.star.bris.ac.uk/~mbt/