SemanticsCalls-3 < IVOA

IVOA Web>IvoaSemantics>SemanticsCalls>SemanticsCalls-3 (2021-03-11, MarkusDemleitner) (raw view)

The third edition of the SemanticsCalls series of telecons took
place *Wed, 2021-03-10 at 16:00 UTC*.

---+ Agenda

   * Steps towards a vocabulary of object types
   * Pre-RFC check of Vocabularies in the VO 2
   * AOB: new VEPs in the pipeline?  Anything else?

---+ Minutes


---+++ Steps towards a vocabulary of object types

Use cases:

   * Simbad (TAP queries) – currently, this uses http://simbad.u-strasbg.fr/simbad/sim-display?data=otypes, which is strictly 4-level, and it is being re-engineered.  
   * SSAP/ObsCore target type

Prior Art:

   * Simbad (numeric/symbolic/long/short)
   * Most object types in UAT
   * Ontology of object types: https://ivoa.net/documents/Notes/AstrObjectOntology/

The draft new hierarchy planned for Simbad is: http://simbad.u-strasbg.fr/guide/otypes-new.htx

This doesn't contain the links to the corresponding UAT concepts yet; some things ("blue object") didn't have good matches. 

Comparison of new object types in Simbad and UAT
http://simbad.u-strasbg.fr/otypes-data/newotypes_uat%20-%20newotypes.tmp.tsv.csv

How would using vocabularies this look like for the user in ADQL?

MD: see gavo_vocmatch UDF on http://dc.g-vo.org/tap/capabilities; you would use
=obj_type='double-star'= for an exact match, 
=1=gavo_vocmatch('otype', 'double-star', obj_type)= to match the entire branch.
But of course SIMBAD could use custom ways to make common use cases more straightforward.

Then there's candidate problem: The old hierarchy had some classes that said "we're not sure yet".  

The current thinking is to have a Candidate-X for every X.  MD suggests that perhaps a "we're not sure about the object type yet" flag in the table might work as well?  SD considers if this could perhaps work like the membership likelihood for star clusters and the like?

Action: CDS->MD: prototype hierarchy+prototype mapping to UAT.  MD then tries how far he gets with forming this into an IVOA vocabulary.

---+++ Pre-RFC check of Vocabularies in the VO 2

Implementations:

   * Vocabulary repo http://www.ivoa.net/rdf
   * VOTable validator in TOPCAT validates TIMESYS attributes
   * RofR validator doesn't quite yet validate VOResource attributes (but we're looking into it)
   * Datalink query expansion, pyVO PR https://github.com/astropy/pyvo/pull/241
   * UAT mapping of Registry content: Sembarebro, rr.subject_uat on reg.g-vo.org; see https://blog.g-vo.org/semantics-cross-discipline-discovery-and-down-to-earth-code/.
   * ADQL UDF gavo_vocmatch(vocname, term, col) in DaCHS >2.3
   * Several VEPs
   * UAT adoption PEN

Request: If you know people in the TCG, remind them to review the document when RFC has started (which will happen soon).
    
---+++ AOB

http://volute.g-vo.org/svn/trunk/projects/semantics/veps/VEP-006.txt is still open; ML will voice her concerns on how this might overload #progenitor on-list.







<br/>
<!--
      * Set ALLOWTOPICRENAME = IVOA.TWikiAdminGroup
-->

Topic revision: r2 - 2021-03-11 - MarkusDemleitner

IVOA

Log in or Register

IVOA.net
Wiki Home
WebChanges
WebTopicList
WebStatistics

Twiki Meta & Help
IVOA
Know
Main
Sandbox
TWiki

TWiki intro
TWiki tutorial
User registration
Notify me

Working Groups

Interest Groups

Committees

Stds&Procs

www.ivoa.net
Documents
Events
Members
XML Schema