[kepler-dev] [Bug 5362] Create a GUI so user can easily archive a senosr's data from a dataturbine to a metacat

[bugzilla-daemon at ecoinformatics.org]

http://bugzilla.ecoinformatics.org/show_bug.cgi?id=5362

--- Comment #5 from Jing Tao <tao at nceas.ucsb.edu> 2011-05-03 13:11:16 PDT ---
Hi, matt:

Derik, Dan and I talked about the issue couple times. We've had to drop the
idea of a new Sensor actor context menu for archiving, and a visual indicator
on the  Sensor actor that shows it's being archived.

Here are two options we've come up with:

Options 1. Add a menu item named "Sensor Archiving..." in the Tools menu. This
action will bring up the GUI shown in the attachment (dan did it).

a. Filling this GUI involves these actions: search kar files in a remote data
repository which archives the data from specified sensor; search existing
schedules for the given kar file id in a scheduler server.

b. Clicking OK button will involve these actions:
*) If there is no existing kar file associated with the sensor, the action will
edit the archiving workflow by passing sensor id, DT server address, and
dataset destination parameters to it. Then the customized kar file is uploaded
to a remote repository.
*) Save the changes of the schedule into a scheduler server

c. Clicking Cancel button will close the GUI

pros:
*) easy for user, user doesn't have to really know anything about the archival
WF

cons:
*) user doesn't know anything about archival WF :) -- conceptual disconnects:
streaming data shown in site layout vs where the data actually comes from for
archival (DataTurbine). Unclear when data being archived starts (beginning
of time of channel in DT). Auto-scheduled workflow kars are generated but user
doesn't know their names.
*) we're giving the archival WF special status and tying it to this specialized
scheduler dialog (dialog needs certain wf
params to exist). However a user might want to archive with their own workflow,
in which case they must learn how
to use the regular scheduler instead of this dialog.
*) 2 interfaces to scheduler from within kepler could be confusing

Option 2 No new features. User must simply edit the archival workflow and
schedule it. We could simply add a small
annotation on site layouts "To archive this data, open
sensor-view/archivalWorkflow.kar". The archive WF
could in turn have a small annotation explaining how to configure and schedule
it "Tools=>Scheduler".

pros:
*) no new work
*) avoids all cons of (1)

cons:
*) user must search use scheduler to check for pre-existing scheduled archival
workflow themselves
*) less 'guided' - user must open WF and config it themselves.  User may find
it is not easy to find a sensor id.

-- 
Configure bugmail: http://bugzilla.ecoinformatics.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.