Throwing @nidever in at the deep end!
Are you just referring to Discourse here or do you have a larger scope for this?
Gregory doesn’t have an account here yet but @xiuqin does.
@mjuric We talked about this with Beth last week, we’re happy to help but can you liaise with her to make sure there’s no duplication/overlap/mixed-messaging. She’s spearheading scence communications issues for LSST it seems.
I’m happy to talk about “data we plan to process” as long as we can meet earlier in the week to discuss the plan. 
1 Like
- @jbosch : yes, that would be great. I’d stick to astronomically obvious functionality as much as possible (i.e., multi-band coadds == go, refactoring of module X == pass)
- @ktl : OK, I can prepare a brief summary (I’ll go over it with you).
- @jsick, @frossie : I think we can use this to summarize where the project is going (if a clear direction crystalizes from Beth’s meetings over the past three days), as it pertains to DM. To be honest, I was initially thinking to use this to solicit feedback and/or further testing by the community for things like Discourse. Any other plans SQuaRE may have that impact the community, this may be a good forum to ask about it (e.g., software distribution plans, etc.)
- @timj, @nidever : No worries, we’ve got resuscitation crews on hand :). On a more serious note, yes, we should meet before to discuss benchmark dataset plans. I’ll set something up for Mon, Tue, or Wed (may have to be over lunch or dinner)
- @xiuqin: If we’re ready, it would be good to give the community a flavor of what our data access website may look like or what kind of functionality it may support (and invite feedback). Note the “if we’re ready” qualifier! If it’s better to hold off on doing this for a few more months until the plans for “version 1” crystalize, we should do that. We can chat offline.
PS: I’m worried the list of things we wanted to present is too ambitious for the time allotted (1.5 hrs) – we may need to cut something 
.
Hi all (esp. @jbosch, @nidever, @davidciardi, @jsick),
Just to confirm to everyone we’re sticking with the structure laid out in the opening post of this topic. The times listed there do not include the time for discussion. Also, if there’s any time left in the end, we’ll use it for an open Q&A session.
PS: @davidciardi will try to deliver the talk via videocon; if that doesn’t work out, @xiuqin, could you do it (or nominate someone someone from the SUI team to take over)?
PPS: If you want to dry-run (or have me look at) your slides, happy to do that tomorrow evening.
Thanks!!
@mjuric Is a talk in the same spirit as the one I gave in ‘Communications I’ ok? Essentially:
- introduce my role,
- demo Discourse,
- show slides inviting astronomers to join us and to think about adopting it for their own purposes.
[10 min]? DM software/algorithm highlights in new release
One suggestion from a new-to-DM, science perspective is that it would be nice to have a sort of “checklist” of features that an astronomer might expect from a photometry package (not just a survey pipeline) that have or have not yet been implemented. E.g.,
- PSF photometry - Done
- Coadd creation - Done
- Moving object [something something something] - ??
- Deblending - ??
- Extended source photometry - ??
Overall I’m thinking if-this-was-sextractor, what would your feature list be right now, what is semi-complete, and what is the near term planned feature list. It’s definitely something that would help me understand the current capabilities of DM better.
1 Like
@ctslater: “Done” is a somewhat problematic term. I prefer to think of things as getting to a minimally functional state, then a scientifically useful (state of the art) state, then a meets-the-SRD state. I’d like to eventually get to a color-coded chart that shows our progress on each component; I started working on that but got bogged down in tool issues.
1 Like
Hi everyone,
I have slides from @nidever and @davidciardi, I’m working to finalize my talk, and using the opportunity to ping @jbosch and @jsick to make sure they have theirs ready by ~9:30 am tomorrow :).
I’d love to see that list (or color-coded chart) as well.
Here are my slides; sorry I didn’t send them earlier. I’ll mention a number of changes we’re bringing over from the HSC side, then focus in on the new multi-band processing, which is both the biggest change (IMO) and the one for which I have the best slides.
I think this would be very useful, and I’d be happy to put some effort into it myself - but it won’t happen by tomorrow, and probably not this week. Remind if I forget (unless @ktl or @mjuric want to take the lead on it).
@timj has already been doing some work at https://confluence.lsstcorp.org/display/DM/Data+Processing+End+to+End+Testing and the pages under it which he and I have used to start working on some diagrams. But we need to fix problems with them and then get to coloring.
Thanks everyone,
Here are the presentations:
You have ~40 minutes for a few final updates 
@davidciardi – We’re looking into setting up the videocon system in the ballroom; should have more for you in ~15-20 minutes.
Thanks … I am here and ready when you are.
One question for @nidever : - I did not see panstarrs listed in your benchmark datasets. Did you decide to hold off on those data until they become public?
@mjuric - also, if my slides are too long, I can skip the slides with the demos and let people know to talk to Xiuqin. In my view the most important slides for this audience is the philosophy and the last slide
A note to say this session is going really well - a wonderfully clear overview of DM!
1 Like
Thanks to everyone for contributing and attending!
The final versions of all talks can be now be found on the workshop pages.