Minutes from SSS Mtg 2008-May-13
Attendees: D.Harland, G.v.Moorsel, F.Owen, J.Rochford, L.Sjouwerman, B.Truitt
As a group we reviewed the changes that were introduced into the latest
version of the software (see list at bottom of this page) earlier today.
Frazer discussed the trouble he had had with the tool timing out and the
subsequent inability to even close the browser. David mentioned that
JIRA EVL-561 (see below) should go part of the way towards helping
the situation. We then discussed other strategies to take when
there is an undisciplined exit (time-out, browser crash, browser
closed w/out exiting tool, etc.). Currently any changes that were
not explicitly saved are lost. Things discussed:
+ Longer period before time-out.
+ No time-out at all.
+ Change default to save all changes on undisciplined exit.
+ Institute a backup / recovery scheme: system autosaves but
does not automatically replace already saved changes.
User has choice at next start-up to keep or discard the autosave.
[BT, JR, & DH met afterwards and the result will be a couple of
JIRA issues that incorporate some of the above.]
For JIRA EVL-567 the consensus was to ~not~ clear selected items on copy.
We discussed JIRA EVL-572 (feedback to user on cross-field validation).
We will pursue the idea of an error console at the bottom of the page.
We discussed JIRA EVL-574 (associating configurations into groups).
Brian has a prototype in the source catalog that we were not able
to demonstrate. We should all take a look at this.
We discussed disabling the browser's tool bar so that users don't
see the foward and back buttons. It looks like using these buttons
causes no harm, but we might want to hide them any way.
All acknowledged that the hardest part for new users will be understanding
just how to get started with the OPT suite of tools. We had no good solutions
yet, other than using documentation to get people started.
The programming staff asked that the scientists test the issues that
we claim to have completed. Testers were also encouraged to continue
to work with the instrument configuration tool for the next week or so.
We think that one week from now we will ask people to begin testing the
source catalog tool.
The following JIRA issues are in the version of the RCT that was placed on webtest earlier today.