2023-04-24 DM-SST Agenda and Meeting Notes

Short : ls.st/dmsst 

Zoom: https://noirlab-edu.zoom.us/j/97839121776?pwd=K1JPeUpSMXFvSXJSa2xORGkyVk5zdz09

Zoom Meeting ID: 978 3912 1776 
Zoom Password: 512314

Dial closest IP: 162.255.36.11 (east coast) and 162.255.37.11 (west coast) then use the Zoom meeting ID 978 3912 1776  as the dialing extension.

For example: 978 3912 1776 @162.255.37.11 or: 162.255.37.11##978 3912 1776  Password: 512314

Dial-in numbers:

+1 346 248 7799 (US Toll)

+1 669 900 6833 (US Toll) Meeting ID: 936 2540 1560 International numbers available: https://gemini.zoom.us/u/adcUNrbXzS

PST meeting 2023-04-19: 

• TMA Test Campaign review
• Keith presented on  Proposed Policies for Rubin Observatory Commissioning-Era Information Sharing (not yet public, still under discussion)
• Topic suggestions for Building momentum to early science
  • Plans for ~ monthly releases on a range of topics to build momentum in advance of first light and the start of LSST

Focus SST meeting next week on support for brokers in commissioning

Discussed need for a policy for how we're all going to communicate as a team (esp. keeping in mind embargo, etc.). Plan taking shape, but not ready to share with community yet. Ensure continued free flow of info within the commissioning team, while being as transparent as possible to the community. 

Lot of work going on toward celebratory milestones, and in particular building momentum toward first light/early science. ~monthly "release" of a feature – communications is seeking science topic suggestions. 

DMTN-049

https://lsstdescrail.readthedocs.io/en/latest/

DMTN-049 describes roadmap for photo-z. Plan for science community to bring their own PZ estimators (~DP2 – PZ validation cooperative) and we (PZ Commissioning team) vet them for future use. Working to decide on training and validation datasets. 

Q: will folks need support from pipelines when testing their estimators? A: goal is to embed PZ estimator in the pipelines. But unknown – could be an afterburner. (Has to use Object catalog as input.)

Because people volunteering to do this work will be running it, will need to use RAIL. Jim B. has concerns about making sure that RAIL can be used without our middleware (or making sure the algos get directly into pipelines). Plenty of open questions re: how to run and how much middleware work will need to be done to shoehorn RAIL analysis into pipelines...

Zeljko is working with Croatian team that's going to derive stellar distances from Object table, and will use RAIL to validate them. 

Q: does the DPDD already allocate what we'll need? A: it specifies a 2x100 float, which may include the sampled PDF and flags?

 80h already anounced: https://www.lsst.org/about/dm/data-products

dmtn-199 for latency requirements (work in progress) 

Could start taking data even if we're not meeting latency reqs. Proposed letting DMS-REQ-0004 slip to pri. 2. GPDF note: hasn't been edited to reflect 80-hr embargo. Also - may want to split into images vs. catalogs, since catalogs aren't embargoed.

REQ-0319 – in Y1 there won't be 1 yr of prior data, so can't do this until later. Propose 1b → 2. But may keep it 1b since we'll have tooling to characterize variability.

REQ-0271 – in Y1 won't be official DRs to associate to. Split out the matching as a 3rd requirement with priority 2.

REQ-0272 – won't have baseline to derive plx/pm, so could downgrade to pri 2. Or separate to a new requirement?

REQ-0273 – propose 2 → 1b. We'll be ready, and it's a core piece of SS science.

REQ-0323 – propose 3 → 2, since we're already saying we'll do this. In practice it'll probably be done quickly/easily.

REQ-0324 – propose 1b → 2 because matching to Object can't happen until we have DR. Perhaps merge with previous DIA-Obj matching.

REQ-0284 – catch-up processing, so keep at 1b (was proposed to move to 1a).

REQ-0348 – currently 2, could be a 3. Keep it as a 2 since it's in SRD.

REQ-0287 – precovery. Currently 1b, could be pri 2? Mechanism to go into 30-day cache and extract things may be a challenge. Could give us breathing room.

REQ-0288 – could be 1b as this is now the baseline.

REQ-0089 – can start 10yr survey even if SS Objects aren't available within spec latency. Move from 1b → 2?

MLG: Re. DMS-REQ-0271, we added new host association parameters to DIAObjects via the DPDD (Table 3, page 23) but not to DMS. Probably the requirement doesn't need to change but ?

Eric will draft an RFC and LCR to propose changes.