Date July 01, 2014
Attendees
Unknown User (robyn), Unknown User (abecker), Kian-Tat Lim, Jim Bosch
on vacation: Jacek Becla, Simon Krughoff
- missing: Mario Juric,
Discussion Items
Andy
- Andy is working on machine learning of estimates of diffraction in DCR and I'm having to look at how quasars have way different spectra than stars in practice in terms of being able to predict what the refraction is. Andy is essentially using machine learning to predict what refraction of DCR is of a given object given it's color and airmass. He found he could do this within tolerance for all the given bands except for u-bands. U-bands only kicks in at about 23 airmass. There are limits on universal cadence fields in airmass for about 1.5 for most passbands. Does anyone know if we have different limits for u bands on the airmass for the universal cadence field.
- K-T: Cadence is all up in the air. There is a huge workshop at the August meeting to discuss cadence. K-T doesn't know if anobody really knows.
- Andy: I've found that beyond 20 degrees, we're having a tough time modeling the refraction in DCR in u-band.
- K-T: If there are some constraints you want to put on the cadence then that would be good to know.
- Andy: My first suggestion would to limit u-band to less than 20 degrees from zenith.
- K-T: Then they'll have to stick that into OpSim to see what that means in terms of getting coverage.
- Andy: It's going to be impossible for things in the LMC.
- Andy is handling quasars right now. There should be a small number of them in our fields; a lot of them are probably going to provide DCR dipoles.
- Andy is working on machine learning of estimates of diffraction in DCR and I'm having to look at how quasars have way different spectra than stars in practice in terms of being able to predict what the refraction is. Andy is essentially using machine learning to predict what refraction of DCR is of a given object given it's color and airmass. He found he could do this within tolerance for all the given bands except for u-bands. U-bands only kicks in at about 23 airmass. There are limits on universal cadence fields in airmass for about 1.5 for most passbands. Does anyone know if we have different limits for u bands on the airmass for the universal cadence field.
K-T
- He sent out the Middleware Meeting Notes 2014-07-01 to 'lsst-data'.
- No blockers. The team is working on getting some demos to illustrate that they've completed Nightly Computing simulations.
- K-T is doing work on the butler this week.
Jim
- Jim went to a review for Perry on trying to get the new measurement framework ready to get merged into the stack and used as the default option. Well, it's been merged into the stack; we're trying to get the changes merged into everything else.to make it a useful to use both options. They found two more things they want to do before they do that. It's getting very close.Jim thinks he's enough for Perry to do before Jim finishes his own task.
- K-T: So this will make it into the August release?
- Jim: Yes. He had anticipated this going in sometime in the next one to two weeks. When it's closer to release, they'll send a note to the list telling everyone about it.
Robyn
- Robyn plans on turning off the shadow buildbot after the meeting in order to transition buildbot over to running on the DM stack.
- K-T: The test buildbot is working fine, master is building, you can do tickets on demand, and all that stuff?
- Robyn: Yes. The one thing we don't have is ldap authentication because buildbot doesn't have real support for it. So we're using htpasswd using a well-known account. Because we need to send personalized email messages to users who trigger a build (with or without tickets), there is a required parameter to enter your user name and email account.
- K-T: is that address saved?
- Robyn: Yes, it's permanently recorded in the metadata for the build.