Proposed SSA Errata

[Markus Demleitner]

Hi James, Dear DAL,

On Tue, Dec 05, 2023 at 05:27:00AM +0000, Dempsey, James (IM&T, Black Mountain) wrote:
> In a belated response to that I have drafted three proposed errata
> for Simple Spectral Access:

Thanks for these.  Brief comments:

>   *   SSA-1.1 Erratum  3
>   <https://wiki.ivoa.net/twiki/bin/view/IVOA/SSA-1_1-Err-3>:
>   Correct XMLNS in examples

The opening VOTABLE tag on these examples was broken in quite a
number of ways; hence, we are currently discussion what's the best
way to fix them without inflating the erratum too much in the
respective github bug, <https://github.com/ivoa-std/SSA/issues/2>.

I think we have converged on:

  <VOTABLE version="1.1" xmlns="http://www.ivoa.net/xml/VOTable/v1.1">

(i.e., keep the outdated version, have no schemaLocation).  If you
disagree, join the fray.

>   *   SSA-1.1 Erratum 4
>   <https://wiki.ivoa.net/twiki/bin/view/IVOA/SSA-1_1-Err-4>: Make
>   coverage optional

I have made a few editorial changes to that, in particular noting
that we have quite a few SSA theory services already, so we're really
only updating the standard to practices present since even before
REC.

>   *   SSA-1.1 Erratum 5
>   <https://wiki.ivoa.net/twiki/bin/view/IVOA/SSA-1_1-Err-5>:
>   Clarify utype prefixing

Here, I think we really must say

  All UTYPEs defined in this document must [instead of "should" right
  now] be presented with a namespace of "ssa:".

to make this useful.  For one, if the validator fails without the ssa
prefix, it's already assuming a MUST, and since the validator here is
saner than the spec, let's fix the spec rather than the validator.

Also, looking at pyvo, anything without the ssa prefix is broken
there, too (although I note that pyvo is probably broken, too,
because it doesn't seem to case-normalise the utypes; that's my
chance to put in my ceterum censeo: case-insensitivity is evil and
leads to breakage left and right.  Let's not do it ever again).

Thanks,

         Markus