Unknown User (mjuric-admin) will take over the editing at 5pm and produce a version to be submitted to the TCT next week, together with a list of "Known Issues" to be resolved at a latter date
Mario Juric: "I've lost track of all the CRs that we (the SST) are driving or contributing to (e.g., retaining all DRs, keeping calexps, keeping footprints, ...). I'd like to spend a few minutes to enumerate all of them and get a better idea of the TODO list."
Discussed LCR-908, "Keep all Data Releases loaded in databases" and the associated DM ticket
RFC-336
-
Getting issue details...STATUS
. Held up on providing requirements to the database team on expected performance of databases in older DRs.
Discussed RFC-329 (Heavy Footprints); Jim Bosch thinks the motivation to close this has dropped once everyone was in rough consensus it's doable; Unknown User (mjuric-admin) argues that's all the more reason to close it. Action on Jim to add it to the decision-making session for the DMLT next week.
Discussed
RFC-325
-
Getting issue details...STATUS
(keeping all PVIs). Initial costing for this returned high numbers, but the sense of the group is that this change would be tremendously useful for science and we should explore in depth different ways how it may get done (including, e.g., lossy compression; Robert Gruendl reported that DES successfully uses lossy-compressed images with 6.8x compression factor). Unknown User (mjuric-admin) will comment on the RFC, and organize follow-up.
Also discussed the work that Melissa Graham started on better defining the requirements on the DM System and likely processing scenarios that will need to be supported for Special Programs (DDFs included).
Eric Bellm discussed four likely CRs coming out of our interactions with the UK community at the Edinburgh LSST:UK meeting:
CR to broadcast upper limits on previous (non)observations of newly discovered transients.Eric Bellm thinks this is extremely important for enabling follow-up (with others concurring).
CR to fix the language in the DPDD to make it clear we're not deblending on Level 1 images
CR to review and change 24 hour requirements to "next night" requirements. A proposal was to make that X hours (8 was mentioned)
CR/study to define our policy wrt. what happens when the maximum processing time is exceeded for alerts (i.e., when the L1 system cannot keep up with processing in real time all images coming from the telescope – in crowded fields, for example). Mario Juric thinks this is an important thing to get studied and resolved on a short timescale; it will have implications on system design.