[WG] Final Quarter Metaticket

[lawrence]

(Which will be complete when all the activities listed below have been properly ticketed, and the final milestones reorganised).

• CSML Issues
  • Scanning tool (csml.py)
    • CSMLid needs to be fixed so scanning doesn't need redoing. Need this to be done asap so CSML production can begin. #788
      • CSML production needs to be moved to an operational activity for numerical datasets. #48
    • NOCS need the scanning tool too, but can't really do any work on this til the HDF read methods are available.
  • HDF support in CSML (read methods), expect these by end of JUNE. #187
  • trajectory ... methods are important for intercomparison ... expect by end of JULY #765
  • QXF finished ... expect within next fortnight. #186 #498 #753
• Indexing the spatial stuff correctly ...
  • Area and centroids in postgres and ordering ...
    • Matt needs to be made aware and it needs doing by end of JULY. #821
  • Not in NDG2:
    • Need to change ingest to support multiple spatial bounding boxes. #822
    • Browse would then need to take the bounding box search ...
• VocabServer? API for broader/narrower (V2, due mid-August)#685
  • Discovery Support
    • Ajax in Discovery (probably done in next few weeks)
    • Vocabserver definitions on parameters in browse (last half of August)
    • Narrower broader on results page (last half of August)
• Granules in MOLES
  • BADC: COAPEC (scanning+catalogue issues, need by end OF JUNE) #48 #785 #784; DDC (end of August)
  • BODC:
    • RAL Network Jade (end of June)
    • Some work needed by Roy to support granules (end of August) #559
      • Oracle -> Roy's Schema -> MOLES Granules ...
      • Some dataset reorganisation needed ... (a few days)
  • PML: you're ok for granules ... (a bit of linking ...) #385
  • NOCS:
    • depends on HDF read methods ... #187
    • got the MOLES, need a procedure ... (need to automate production from underlying database).
    • Fabio needs to go through all his tickets before leaving.
  • Service will be generated by config file, not all granules will be available at those addresses, need to fail gracefully.
    • Ideally there would be a place in MOLES to indicate whether a service could exist. This could be achieved by having a dummy entry, that if present, would be replaced by the real thing by the GUI ...
• Security
  • Big push with Phil, next weekly meeting, but most of the work in July.
  • Low hung fruit: links between user id and role ... and datasets ...
  • User databases; ready and working by end of July.
    • Role Mapping depends on key datasets.
• Datasets:
  • BADC:
  • COAPEC + DDC (get someone else to be repsonsible for moles)
  • BADC needs to get Granule IDs into the system.
  • PML:
    • All data available ...
      • Link granules into moles
    • Ongoing issue
      • too many datasets, need iso multiple bounding boxes.
  • BODC:
    • Point timeseries at sea level
    • profile reads ...
  • NOCS
    • HDF read methods ... 1/4 degree data centre ...
      • Needs a couple of weeks work ... by the end of June should have some python code to read HDF(4).

[selatham]

Everything seems to be ticketed elsewhere.