New ProvenanceDM working draft released

[Kristin Riebe]

Dear Data Modelers,

hooray! The provenance work group has finished a new version of the
ProvenanceDM working draft, see
http://wiki.ivoa.net/internal/IVOA/ObservationProvenanceDataModel/WD-ProvenanceDM-1.0-20170921.pdf

It's going to appear soon at the IVOA documents page as well.

The main changes are:
- Complete rewrite of the section on Accessing provenance: it's now
split into two sections, one on serializations, the other on access
protocols ProvDAL and ProvTAP, which were much extended, based on our
experience from implementations
- New section on How to use the model
- Proposed PROV-VOTable/TAP_SCHEMA tables in appendix
- Many more examples

Unfortunately with all the new additions we reached a rather high number
of pages now, but it's a lot of examples, tables and a long
introduction, so I hope it's not too daunting.

There is still some work to do, especially on finishing our
implementations and the accompanying implementation note, but we still
hope to bring the standard soon to the Proposed Recommmendation phase.

We ask your comments on the following issues (as soon as possible):

1) ProvDAL:
The parameter name "AGENT" (values 'true'/'false') can be misleading,
since AGENT=false only means that tracking provenance should stop at the
agent and it shouldn't be followed any further.
Any ideas for a better name? Maybe EXPLORE_AGENT or TRACKAGENT?
Or we could use more useful parameter values like:
AGENT=END: no further tracking after an AGENT is found
AGENT=EXPLORE: track further (i.e. also find all the other things that
an AGENT was responsible for)
Any other ideas?

2) PROV-VOTable table descriptions, TAP_SCHEMA:
We propose to make the column *names* normative, so the same ADQL query
can be done on multiple provenance services
- Do we need unique column names? (Because there will be table joins!)
- Prefix column names with an abbreviation for the table to which they
belong?

3) UML diagram
Can someone with more UML knowledge please check, if the relations and
compositions make sense? We changed some relations to avoid double
compositions and get a VO-DML compliant UML diagram, see Figure 5.

Many thanks for your input!

Cheers,
Kristin

-- 
-------------------------------------------------------
Dr. Kristin Riebe
Press and Public Outreach

Email: kriebe at aip.de, webmaster at aip.de
Phone: +49 331 7499-377
Room:  Bib/3
-------------------------------------------------------
Leibniz-Institut für Astrophysik Potsdam (AIP)
An der Sternwarte 16, D-14482 Potsdam
Vorstand: Prof. Dr. Matthias Steinmetz, Matthias Winker
Stiftung bürgerlichen Rechts
Stiftungsverzeichnis Brandenburg: 26 742-00/7026
-------------------------------------------------------