Minutes from SSS Mtg 2008-Jul-01

Attendees: B.Butler, M.Claussen, D.Harland, B.Hesman,     G.v.Moorsel,
           F.Owen,   J.Rochford, M.Rupen,   L.Sjouwerman, B.Truitt,    S.Witz

Frazer's discovery regarding cutting the final project from the OPT was demonstrated. The programmers will come up with a better way of handling this. [Created JIRA issue EVL-657 for this.]

* * *

We went through the five points in an email from Gustaaf.

1. In blue banner on top, maybe replace 'Source Catalog' by 'Sources' and 'Instrument Configuration Catalog' by 'Instrument Configurations'
   Brian implemented this.
    
2. In blue banner on top, an exception occurs when clicking on 'Observation preparation' or 'Source Catalog' while in the Instrument Configuration Catalog.
   James fixed this.
    
3. In Source Catalog, search on occurrence in a string, not exact match; i.e. asterisks before and after the search spring should be implied
   The group discussed this point and decided it is a personal preference. We will leave this as is. The documentation will let users know that wildcards "*" and "?" can be used to broaden searches.
    
4. In Source Catalog, don't search in aliases by default, but add the option to extend search to aliases
   The group discussed this for awhile, too. Some conclusions:
   1. We should call the simple search "Name Search".
   2. We should add a check box for whether or not aliases should be included in the name search. (We did not specify what the default should be!) [See JIRA issue EVL-405 for this.]
   
   
5. After clicking on File, Edit, or Help on top, the button still has to be released in order to make a choice: click and drag down still doesn't work
   This is JIRA issue EVL-591. It has a low priority and the developers are not sure whether the 3rd-party software we use will allow us to fulfill this request.

* * *

A few items for the upcoming update of webtest were mentioned:

• The various dialogs associated with pasting in the catalog tools will be in the next software update. We have erred on the side of using these dialogs all the time, as opposed to erring in a manner that left us with out a dialog where one is needed. As we gain experience pasting inside the catalogs we may be able to eliminate in situations where there are no ambiguities.
   
• There is a known issue: If you go to the summary page of the source catalog, change the name, do not hit the enter key, and then leave the name field by some means other than using the tab key, you will see a validation error. The new name, however, has been properly saved. More on this as we figure it out. [JIRA issue EVL-653.]
   
• The exporting of instrument configuration catalogs is no longer a zip file, and no longer gives a clue via its file extension that it is really a disguised XML file.
   
• If a user's browser crashes, the user can now log back in (before the expiration time for their session) and have all their changes preserved. Previously such changes were lost.

* * *

We discussed secondary reference pointing, a topic from last week. Bryan confirmed that the online system does support this feature. The OPT is already capable of getting the needed data from the user. We may need to work on the logic that converts the model to a script in order for secondary reference pointing to work.

We discussed scan options in general and the "Apply last reference pointing" option in particular. It was pointed out that this option feels a little hidden, tucked away as it is on a separate tab. Some felt the apply last ref ptg option should be more prominent. We discussed the difficulty of a generalized solution.

Some ideas for fulfilling the spirit of the original observation were made:

1. Hightlight selected options to make them stand out better.
    
2. Give the OPT or model more "situational awareness". Eg, if a scheduling block has a reference pointing scan, then perhaps those scans created after and placed after the pointing scan could automatically have the apply-ref-ptg option selected.
    
3. Another way to display situational awareness would be to highlight scans in the tree that were using the most recent ref-ptg. If an observer neglected to apply the ptg to a particular scan, that scan would stand out from the others. (We'd need to see if we can generalize this -- eg, if we do a focus scan, do we do the same kind of highlighting? What if we do both a focus and ptg scan?)
    
4. Handle this via validation: if we see a sched block w/ a ptg scan and also see that subsequent scans are not using the latest ptg, we could display a warning. [DMH: this means offers the best hope for incorporating lots of rules in a way that lets us communicate the issue to the user clearly.] [JIRA issue EVL-658.]

* * *

Bryan gave an overall view of the next steps for the project. Initially he and Claire were going to decide whether or not to use the new software for the Ka call. The plan now is that Lorant, Gustaaf, and Mark C. will make an assessment of the likelihood that the software would be ready for widespread use by the end of the year and report their findings to Bob Dickman. Bob & Bryan would then make the go/no-go decision in the middle of this month.

* * *

Mark reported some of his findings:

1. The source catalog gives a number of options for the text formats it will accept for import. However, there is nothing other than some acronyms to let you know just what a particular format is.
    
2. The Advanced Search dialog for sources needs work. [DMH: see JIRA EVL-301.] When doing a cone search, it would be nice to have it center on a source of the user's choosing, and not make the user copy the RA / Dec into the boxes. It would also be nice to display the angular distances of each of the returned sources from the central point.