Minutes from SSS Mtg 2008-Jun-03
 
Attendees: B.Butler,    D.Harland,  B.Hesman,
           G.v.Moorsel, J.Rochford, B.Truitt


The programming staff gave no presentation; instead, the floor was
opened for comments.  Most of the items below were as a result
of Brigette's testing; some were comments from other attendees.

  1. SCT, source details: the flux density has too many positions
     after the decimal point.

  2. Go to nth page of a catalog or group.  Now select an item.
     Is there a quick way to get back to the nth page?
     [This is similar enough to EVL-565 that work on this should
      be done under that issue.]

  3. When using Firefox on a Mac, the top level menus appear
     beneath the scroll bar of the left pane.

  4. We had a discussion on copy / paste of catalog items.
     Recall from prior meetings: we changed the way in which
     we associated items (eg, sources or instrument configurations)
     with groups.  We no longer use copy / paste to make or break
     those associations.  A downside of this is illustrated by
     the following use case.

        Astronomer Jones has a personal catalog of sources,
        and this catalog has groups.

        Jones goes to the std NRAO calibrator catalog and
        copies a few sources.

        Jones wants to put those sources in group 2 of the
        personal catalog, so Jones tries to paste them there.
        However, the tool no longer allows pasting as a way to
        put items into a group.  Instead, Jones must paste the
        sources into the personal catalog, locate and select
        those new sources, then associate them with group 2.

     The expectation that Jones had was reasonable and should
     be supported.  However, we do not want to go back to using
     copy / paste as the group-association method because when
     one takes source S from catalog C and pastes it into
     groups G1 and G2 of C, we are NOT really making copies of S.
     Instead, C holds one instance of S, and both G1 and G2 point
     to S.

     It was also made clear from the discussion that if one copies an
     item from any catalog and then pastes that item into a catalog
     that has an exactly-equal item, the pasting process should
     create a new item with a similar name.  (For pasting groups
     we use "GroupName (copy 1)", for example.)
     
     Brian and Dave will get together to see if we can support
     all the user-expected behaviors with the current group
     model.  If not, we may consider replacing it with a tagging
     system.  In this system items could have zero or more labels,
     or "tags", pinned on them.  We could then provide a way
     to filter catalogs on certain tags, perhaps with the current
     catalog / group tree view, or perhaps not.  This is not a
     pressing enough issue, though, to distract us from more
     important ones.

  5. The PdBI calibrator catalog needs RA / Dec groups.

  6. We had a discussion on the telescope motion model and
     the types of time (start time, end time, duration, on-source)
     that can be used for a given scan.  We agreed that it would
     be acceptable if we allowed the use of an optional start time
     for a whole scheduling block, but then outlawed the use
     of start times for individual scans.

  7. What some of the icons mean (eg, the little red "*" that
     shows up on the positions in a table of sources and the
     little yellow dot for unsaved items) is not immediately
     obvious.  We need application help on these things.
     [Note: Lorant is in charge of the help systems for the tools.
      Contact him directly with any suggestions regarding
      user help.]

  8. Users can search for calibrators by completing a form and
     pressing the "Search" button.  It can take the system a little
     while to complete the search.  The user has to watch carefully
     to see if the search results have arrived.  We need to make
     the "here are your results" notification more prominent.

  9. The unsaved-data marker on icons should be more prominent.

 10. The UVmin/max from the old VLA calibrator catalog might
     not have a good home in the source model.  They are
     shown in the "historical record" section, but that text
     is more computer friendly, not human friendly.  Perhaps
     the model needs tweeking, or we could use the 
     user-defined values section.

 11. We should parse the SMA and ATCA calibrator catalogs and
     make them read-only catalogs in the SCT.


Abbreviations used above:

  a. SCT = Source Catalog Tool
  b. EVL-xxx = an EVLA item in "JIRA", our issue tracking software
  c. EVLA = you'd better know that one already
  d. SMA = SubMillimeter Array
  e. ATCA = Australia Telescope Compact Array

We did not update webtest this week, so this note has no list of JIRA issues.

( As a reminder, the HTTP address for the RCT is http://webtest.aoc.nrao.edu/rct/.
You can navigate to this from the main SSS Documentation page,
http://www.aoc.nrao.edu/asg-internal/software/sss/sssDocMain.shtml. )