Reported by steven.c.hankin on 5 Jun 2009 16:11 UTC
Following the meeting with Chris and Cathy yesterday, the following discussion/suggestion on how to improve the tracking of changes to the data files. Cathy raised the question -- if a scientists has repaired a QC problem and submitted the updated cruise file, but that file does not become a visible part of SOCAT until the next version, how do other users know that the cruise has been updated? Here's my suggestion:
Remove the "SUSPEND" and "EXCLUDE" QF flag
Instead add a new "Status" constraint with possible states of "NEW", "UPDATED", "EXCLUDE", "NEEDFIX", "ISFIXED".
"NEEDFIX" is equivalent to our previous "SUSPEND". "ISFIXED" means that a SVN has the repairs, but only in the next version of SOCAT will the status change from "ISFIXED" to "UPDATED".
Unlike QC flags, the status flags should be maintained exclusively by you -- not by the scientists. Basically -- the status flags reflect the state of SVN.
All of this implies a new database field (or table?) to keep track of status, as well as QC flag.
Comment by steven.c.hankin on 12 Jun 2009 14:16 UTC
LETS POSTPONE THIS UNTIL AFTER NORWICH. IT DOESN'T REALLY BECOME RELEVANT UNTIL THE CRUISE FILES START GETTING CHANGED.
Reported by steven.c.hankin on 5 Jun 2009 16:11 UTC Following the meeting with Chris and Cathy yesterday, the following discussion/suggestion on how to improve the tracking of changes to the data files. Cathy raised the question -- if a scientists has repaired a QC problem and submitted the updated cruise file, but that file does not become a visible part of SOCAT until the next version, how do other users know that the cruise has been updated? Here's my suggestion:
"NEEDFIX" is equivalent to our previous "SUSPEND". "ISFIXED" means that a SVN has the repairs, but only in the next version of SOCAT will the status change from "ISFIXED" to "UPDATED".
Unlike QC flags, the status flags should be maintained exclusively by you -- not by the scientists. Basically -- the status flags reflect the state of SVN.
All of this implies a new database field (or table?) to keep track of status, as well as QC flag.
See improvements?
Migrated-From: http://dunkel.pmel.noaa.gov/trac/las/ticket/703