search

ivoa-std / ObsCoreExtensionForRadioData

ObsCore model extension for radio data
Creative Commons Attribution Share Alike 4.0 International
0 stars 6 forks source link

tracking_type #57

Open alle-ira opened 6 months ago

alle-ira commented 6 months ago

Sect. 4.5 - The parameter tracking_type as defined in ObsLocTAP pertains more to observation planning and execution that to data discovery. Additionally, its scope is wider than the radio extension since it pertains data irrespectively of their energy/frequency domain. Moreover, as it is defined now tracking_type does not comprise all the values needed to describe the variety of tracking modes associated to the sky scan modes (for instance: skydip, on the fly/raster maps).

BaptisteCecconi commented 6 months ago

Anyhow, this kind of list of terms should go into a vocabulary in Semantics... especially since it is used in 2 recommendations (ObsLocTAP and this one).

Bonnarel commented 6 months ago

I agree with Baptiste here. the parameter name should be the same whatever the situation : observation planning or description of achieved observations. this is good for interoperability. And if terms are missing they should be added in the external generic vocabulary

alle-ira commented 6 months ago

Apologies, what we intended is that we are not sure this parameter should be listed in the ObsCore radio extension given its (more) general scope. Is there a use case motivating its inclusion in the radio extension?

loumir commented 6 months ago

Agreed with Alexandra . Will users query an archive using this term? Is this term tracking_type really designating the same concept between ObsLocTap and the tracking into a radio instrument ? Expanding, and standardizing vocabularies takes time to build consensus , so is it really a key parameter ?

mservillat commented 4 months ago

Note that tracking_type is indeed used in different contexts, including in our current High Energy discussions. Cherenkov telescope arrays could use this concept, as well as other ways to scan the sky (we discuss a convergent or divergent mode for example) or run observations sequences (where wobbling between 2 or 4 positions around a source in one special technique). So the High Energy domain could benefit from those descriptions for discovery purposes as well, or for observations characterisation (which to me is also a use case for ObsCore)