Inconsistency regarding element and code table 5-14
joergklausen opened this issue · 10 comments
There is an inconsistency between the standard and the corresponding code list regarding WMDS element 5-14. The standard [0] says it is of type Boolean, but the codelist [1] contains elements of type Character. In the WMDR XSD [2, 3], this has been specified as type Boolean. Of course, this is incompatible. As a result, this code list cannot currently be used. The TT-WMD needs to decide if the standard or the codelist should be supported, and then one or the other needs to be updated accordingly. I personally feel that the codelist with primary, additional, unknown, inapplicable gives more flexibility in the future. But it also means a change request for the schema and OSCAR/Surface API, which will take some time. Comments are most welcome!
Kind regards
Jörg
[0] https://library.wmo.int/?lvl=notice_display&id=19925
[1] http://codes.wmo.int/wmdr/ObservationStatus
[2] http://schemas.wmo.int/wmdr/1.0RC9/ (latest version available here, but not the latest version in use and implemented in OSCAR/Surface)
[3] https://github.com/wmo-im/wmdr (NB: the Master branch is ahead of the release, V1.0.1 is implemented in OSCAR/Surface)
I prefer the option for the codelist which as Jörg states provides flexibility for the future.
I recommend the Codelist option. The boolean approach is simplistic and doesn't cater for a range of real-world situations where there this field has not been populated correctly. A default of 'unkown' is useful until the data manager has specifically edited this field.
Agree.
Revised validation report. This needs to be moved as it relates to a WMDR Schema change.
WMDS_Validation_Report_153_StatusOfObs_v0.2.docx
@joergklausen
@joergklausen and @KarlBureau and @IgorZahumensky I have moved this issue on the schemas repository. The team can still discuss it during the meetings. It is just on another repo. We should have the same project structure also in this repo to be able to trace the progress of the issues.
@IgorZahumensky can you create a project identical to the one we have in the other repository?
@chenxiaoxia2019 can you please help @IgorZahumensky in creating the project?
Thank you @KarlBureau, I concur with the validation report.
I also concur with the validation report and what we agreed at yesterdays meeting.
Looks good, sorry for not being able to attend the meeting.
Change of model is proposed. The existing boolean attribute officialStatus under "Reporting" is removed, and a new attribute observationStatus is added to "Deployment". The attribute remains optional and is of type ObservationStatusType, which makes reference to the existing code table at http://codes.wmo.int/wmdr/ObservationStatus.
(In actual fact, this attribute is Public, hence there should be a "+" sign instead of a "-" sign in the figure)
@efucile @toakley76 Please comment.