<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Thanks to Markus for this new version.<br>
<br>
Statistics have been implemented in a beta version of the CDS
registry.<br>
<br>
Version 1.3 has other features such as those you listed in the
announcement of VODataService, but also the possibility now to
link capabilities and interfaces with tables declarations and thus
solve a historical issue.<br>
<br>
This capability is interesting for VizieR, but also for any other
data centers that provide in a registry record (CatalogResource
or CatalogService) a collection of tables that can be queried via
dedicated services. <br>
(eg: a service described in a capability is only used for a
subset of tables).<br>
This is possible with simple Cone Search (but also in the future
with HaTS, HiPS or SCS2).<br>
<br>
I like it! .. and, I suggest adding a section in the PDF document
explaining this feature.<br>
<br>
Another point, may be a detail. <br>
The XML serialization (v1.3) reuses vs:BaseParam and the 'stats'
element to specify interface-tables relationships. This is correct
for XSD serialization, but the semantic 'stats' sounds bizarre. I
would prefer, if it doesn’t break anything, to use 'values' (as
suggested in the presentation (by Markus) in IVOA).<br>
<br>
Thank again for this new version,<br>
Gilles</p>
<p><br>
</p>
<p><br>
</p>
<div class="moz-cite-prefix">On 7/2/26 14:33, Markus Demleitner via
registry wrote:<br>
</div>
<blockquote type="cite"
cite="mid:rlr3m6na453n66rr4xfy3hq4y7o3igor5nsar6inejpyryslm2@pqn7awpftvkr">
<pre wrap="" class="moz-quote-pre">Dear Registry community,
A month ago I have uploaded a proposed recommendation for
VODataService 1.3 to the document repository:
<a class="moz-txt-link-freetext"
href="https://ivoa.net/documents/VODataService/20260601/">https://ivoa.net/documents/VODataService/20260601/</a>
In this new version 1.3, the most relevant news are:
* column statistics (with a goal to allow plausible guesses of
distributions); sect. 3.5
* declaring which product types (e.g., time series vs. images) are in
a resource; sect. 3.1.1
* declaring the source of the data (experiment vs. "theory"); also
sect. 3.1.1.
We would like to start RFC within the next few weeks. If you have
feedback to give on this until then (see also my Interop talk
<a class="moz-txt-link-rfc2396E"
href="https://wiki.ivoa.net/internal/IVOA/InterOpJune2026Registry/vods13.pdf"><https://wiki.ivoa.net/internal/IVOA/InterOpJune2026Registry/vods13.pdf></a>
and the questions at the end), I'd be very grateful. It's always
nice if documents can stay roughly constant during RFCs...
Thanks,
Markus
</pre>
</blockquote>
</body>
</html>