Adjustments to section Role in the VO Architecture

[mbtaylor]

See issue #56.

[pdowler]

I think we could nominally add DAP (prerec) since I think the WG has agreed to pursue this in favour of more SIA, SSA, etc. And there is an ivoa-std repo where the work is happening.

I also just transferred a SLAP2 repo to ivoa-std, so that's nominally another... maybe too far in future?

[msdemlei]

On Mon, May 12, 2025 at 11:17:47AM -0700, Patrick Dowler wrote: I think we could nominally add DAP (prerec) since I think the WG has agreed to pursue this in favour of more SIA, SSA, etc. And there is an ivoa-std repo where the work is happening. I also just transferred a SLAP2 repo to ivoa-std, so that's nominally another... maybe too far in future?

Well, I think SLAP2 is what'll happen in the end; my main concern -- that we won't build an SSLDM2 against VAMDC -- is addressed, and I admit it's more realistic that the VAMDC people. But: after looking at things, I don't think we should have SLAP2 in the architecture diagram. So far, we don't talk about major versions in the architecture diagram, and SLAP is already in there. However, I have reshuffled the S-Protocols in <ivoa-std/ivoatex#158>, partly because I'd like to have DAP near the top (btw, wat about renaming ConeSearch to SCS? DAP could move to the top then), partly because we then *could* have SLAP and SLAP2 next to each other.

[mbtaylor]

My idea was to show a representative list not a comprehensive one, but maybe showing all DALI-using standards would be an informative thing to do.

If so there are more required: alongside DAP and SLAP (I agree with Markus let's not put major version numbers into this diagram): SODA, SLAP, ObsObjSAP, have I missed any?

[pdowler]

DataLink?

I didn't catch the "representative list" at first, but I see what you are thinking there. It's probably hard/misleading to do the equivalent of

DAP DataLink ... SODA

in a diagram. I actually like the idea if it could be done

[gmantele]

I also agree: let's not have major version in the role diagram. DALI won't apply to SLAP1 anyway thanks to the sentence you've just added:

but the details of compliance are the responsibility of those
specific standards, and may not in all cases conform exactly
to the recommendations here, for instance in the case of ConeSearch 1.03
because the text predates DALI.

Anyway, SLAP(2) will be a thing. So it can be included.

DAP, on the other hand, is not even a draft, so I have mix feelings here. Maybe, it is too soon. DAP should be standard at some point, but maybe we will have DALI-1.3 at the time DAP becomes a REC. As one of the main editor of DAP, what do you think, @pdowler ?

[pdowler]

DAP hasn't been the highest priority for anyone to really push it, but I could see some activity coming this summer/fall. Mainly I think people from Rubin were interested, but they just starting getting really data from the telescope so I suspect have been too busy to think about it.

As I said, I think we decided that DAP would come instead of SIA-2.1+ and it would provide functionality for other data types (thinking mainly spectra, so SSA replacement). We'd also have to figure out how DAP interacts with ObsCore extensions...

[gmantele]

If you think you can have at least a WD by the end of the year, fine with me. You can include it in the role diagram along with SLAP, DataLink and others.

[msdemlei]

On Mon, Jun 02, 2025 at 08:49:46AM -0700, Grégory Mantelet wrote: If you think you can have at least a WD by the end of the year, fine with me. You can include it in the role diagram along with SLAP, DataLink and others.

In that case, would one of you comment/approve ivoa-std/ivoatex#158 Thanks!