This page is a place to gather up missing or underspecified requirements that would prevent or hinder us from delivering construction.
- What are the open science questions that we must answer in order for the institutional science leads to deliver their products?
- What in the DMSR (i.e. misspecified or underspecified requirements) impairs our ability to deliver on the institutional requirements, and to verify that we've done so?
- How do we want to prioritize these?
|What||Detailed Description||DMSR requirements |
|Clear specification of the capabilities or at least use cases for the Alert Database|
Called the "(Historical) Alert Archive" in requirements. Use cases have never been clearly established.
(Eric Bellm wrote Alert Database Use Cases and Draft Requirements but it has never been formalized)
|Specification of the image and image cutout services|
In progress by Gregory Dubois-Felsmann and Simon Krughoff.
|Specification of user-facing bulk download services, if any|
(data-center-to-data-center bulk transfer is the only part originally captured in the requirements)
|Specification of the user-facing batch services||Is this just "traditional batch" access by ssh-ing to a central batch control node and submitting a traditional job specification (essentially, a request to run a particular command line on a batch worker node)? Or something more API-friendly? A UWS service? Something integrated with notebooks? Something PipelineTask-aware?||In progress https://dmtn-202.lsst.io/|
|Proposed capability for generation of DOIs for queries||It has been proposed to enable the creation of DOIs (possibly by a third party acting on our behalf) for queries: for query specifications and/or for query results (as lists of Object/Source/image IDs).||DMS-REQ-0291 (1b), Query Repeatability, in the discussion states that the ability to create DOIs for query specifications would be desirable|
|Requirements on masking satellite trails||Just masking? Do we need to add model fitting? How will the wings of the trails behave?||None. PST consensus is that we need to add some||Wait til we can get a update from Clare when she is back|
|Alerts on sub-threshold DIASources|
The DPDD says
Also, the system will have the ability to measure and alert on a limited (It will be sized for no less than $\sim$10\% of average \DIASource per visit rate) number of sources detected below the nominal threshold for which additional criteria are satisfied. For example, a $transSNR$ = 3 source detection near a gravitational keyhole may be highly significant in assessing the danger posed by a potentially hazardous asteroid. The initial set of criteria will be defined by the start of LSST operations.
|DMS-REQ-0270(2), "Faint DIASource Measurements"|