Minutes from SSS Mtg 2007-Sep-20

Attendees: B.Butler, D.Harland, B.Hesman, J.Rochford, M.Rupen, L.Sjouwerman, S.Witz


* A suggestion was made to move this meeting to Tuesdays.
  Dave will look for open conference rooms and report new date / time
  to everyone.


* SSS group will meet tomorrow at 9:00am to discuss agenda for
  videocon w/ EOD & OpenSky.


* Discussion on JIRA

  - We talked about making it easier for astronomers to separate noise
    (which for them is the technical hooha of how we meet the req'ts)
    from the useful discussions on the current and desired behavior
    of the applications.

  - When updating the apps on webtest, SSS will continue to publish a
    list of items that the newest apps address.

  - When the developers are seeking help from astronomers, we'll start
    the comment in JIRA with "Astronomers:".  When the astronomers see
    this in the emails from JIRA, they'll treat it like the Bat Phone.

  - Dave will continue to plow through the existing issues and put
    astronomers on the watch list of some of those issues.  (Others
    are invited to do the same!)


* Scan Times

  - We spent a bunch of time discussing temporal specification of scans:
    the use of start times, stop times, on-source duration, and total duration.
    There was certainly not complete agreement among the astronomers
    regarding the need to mix durations and points in time.

  - We discussed a couple different ways in which coordination with
    another telescope is needed.  In one way ("synchronized"), each
    and every scan must be coordinated with the other scope.  In the
    other way ("simultaneous") we merely need to be observing in the
    same window of time as the other scope, but both scopes are free
    to wander around a bit (eg, to do calibrations) during that window.

  - It might do me (Dave) some good if i wrote everything down in a
    document and sent this to a few people.  My goal is to make sure
    the model of a scheduling block can tell the obs prep tool, scheduler,
    or anyone else, how long the block will take to run -- given
    some information about initial conditions (what time it is, where do
    we assume the telescopes are pointing, what wrap are we on).

  - The model will need the telescope motion model in place in order
    to be able to provide the true duration of an SB.
    + Craig Walker is a good person to talk to about this.

  - FYI: the old executor worked in stop times.
    The transition executor works in start times.