2020 02 26

2020-02-26 Environmental Data Retrieval SWG Meeting 6

18:00 UTC (19:00 CET, 13:00 EST, 11:00 MST, 05:00 AEDT, 02:00 CST)

gotomeeting in OGC Portal

Agenda:

1. Agree Agenda and IPR Call;

2. Triage of open issues

3. Discuss and Resolve any identified convergence issues with other OGC API standards and drafts;

4. Any Agenda items for the Face-to-Face meeting in Ottowa on 3 March 2020? Change time? (Tuesday, Ottawa, 14:45-16:15 EST local (20:45 CET, 19:45 UTC, 12:45 MST, 06:45(Wed) AEDT, 03:45(Wed) CST)

5. Any issues for Sprint in Washington 18-20 March 2020?

6. Report on EGOWS/ECMWF workshop/telco on Time Series held on Wed 26 Feb 2020:

   • Interest in implementing from ECMWF, Norway, Netherlands;
   • Keen on conformance testing for payloads;
   • Interested in how to handle ensembles of data;
   • They are aware of need for interoperability of metadata frameworks;
   • They are aware of the distinction between JSON, geoJSON, CoverageJSON and TimeSeriesML;
   • They will collect samples of current JSON payloads for time series data.

7. Any Other Business?

   • Does OpenAPI V3.0.3 cause any problems?

Minutes:

Attending: Chris Little UK Met Office
Dave Blodgett USGS
Boyi Shangguan WHU
Lei Hu WHU
Clemens Portele (interactive instruments)
Mark Burgoyne
Mingda Zhang WHU
Shane Mill
Peter Trevelyan

Agreed to agenda. Noted a couple new callers from WHU -- Welcome!

Open Issues:

• Issue #16 needs some minor work in the spec.
• Issue #12 will likely need to be discussed in detail at the sprint. Open PR related to it can be merged.
• Issue #9 needs some text in the document.
• Issue #8 still to address.

Future issues?

• Chris had a discussion with ECMWF -- they are interested in ensembles and conformance tests.
• Re conformance testing: Recent testbed developed CITE tests -- there will be a vote to move new OGC-API tests forward. Expect JSON is supported, not a problem here. Q: will there be a way to have tests for "common"? A: Should be doable, but it is something to determine going forward. Q: Were conformance tests done in the context of openAPI or using the TEAM engine? A: The testbed used an existing library (not Swagger) -- need to review ER for details.

Agenda for Ottawa:

• Discuss scope of V1 and beyond (roadmap)
• Switch time? -- probably not.

Sprint planning:

• UK Met Office and NOAA had planning meeting on Tuesday.
• Chris has requested a sprint repository to use.
• Will have remote participation from ECMWF and WHU.
• Wondering about after hours work space? A hotel or pub might have space?
• End of each day will have a closing session to hand off to WHU participants.
• Once repo is up -- will elicit input from participants about what their plans are.

ECMWF Meeting from 2/25:

• Discussing some software that's been used for EDR-type use cases in the past.
• ECMWF are very interested in time-series services for various slices and profiles of forecast data.
• A number of meteorological services are interested in EDR implementation.
• Conformance testing for payloads are important!
• ECMWF will provide an issue on github for their requirements.
• Had some discussion of metadata frameworks. Meteorological metadata code-lists are tricky in this area.
• Code lists and metadata will be a key part of the hackathon.
• ECMWF will collect a number of example JSON payloads for us to look at.

New version of OpenAPI:

• Mostly bug fixes and minor fixes.
• Next version will be a more major update.
• JSON-schema will be included in the next version.

Actions:

• Need to raise an issue to plan out conformance testing -- probably at the sprint.
• Mark will get merge conflicts fixed on open PR.
• Chris will post another document to pending.
• Dave will raise a github issue for listing explicit EDR patterns for V1, V2, and beyond?
• Shane will work with Steve to get the sprint planning slides distributed.