discrepancy between file checker and manual

[emmerbodc]

we noticed the file checker and manual are not in sync for two variables so just want to check what we agreed on

manual definitions vs the checker error: TIME long_name = “Time elapsed since 1970-01-01T00:00:00Z” VS Required("long_name: Values mismatch: '\"time of measurement\"' vs '\"time of measurement and gps location\"' ") LATITUDE long_name = “Latitude north (WGS84)” VS Required("long_name: Values mismatch: '\"latitude of each measurements and GPS location\"' vs '\"Latitude north\"' ")

@castelao @vturpin @callumrollo

[castelao]

Thanks @emmerbodc . The checker should always respect the manual which has a formal procedure for review, discussion, and approval. Any difference was my mistake or I haven't have the chance to update yet. That said, it is a strange long_name for Time as "Time elapsed since 1970-01-01T00:00:00Z". There is a field for units. The way I think is the long_name would be the axis label in a plot, and might be followed by the units (in the label), such as:

Time [seconds since 1970-01-01T00:00:00Z]

If there is no discussion by the weekend, I'll update the checker to follow the current version of the manual.

[vturpin]

We had this discussion during the final review before Gotheborg. We wanted to align as much as possible location and time variable with the existing vocabulary.

I think we said "long_name" should be the definition from the control vocab and "standard_name" the prefered label of the control vocab.

https://github.com/OceanGlidersCommunity/OG-format-user-manual/issues/252 https://github.com/OceanGlidersCommunity/OG-format-user-manual/issues/253

In fact, we did it for "lat" but not for "lon" and time. Probably in the rush we miss that.

I think we should do the same we did for latitude with longitude (http://vocab.nerc.ac.uk/collection/OG1/current/LAT/) and time (http://vocab.nerc.ac.uk/collection/OG1/current/TIME/) I recall that we have added a "latitude_vocabulary" attribute to link to the vocab (http://vocab.nerc.ac.uk/collection/OG1/current/LAT/)

We should do the same for LON and TIME

[emmerbodc]

Ah yes I thought we had discussed this but couldn't remember what the final decision was. We will update the manual.

[castelao]

I'll wait the document update before updating the checker to conform with the document.

I have a question. Since the long_name includes (WGS84), is that the only acceptable reference? It seems to me a restrictive choice for the long run of the standard. On top of that, there are attributes commonly used to define the reference system used, and using those would make the glider standard closer to conform with other standards and give more flexibility for the OG-1.0.

I have the impression that mixing standard_names, with implicit units and reference in the name would go against the so many attributes and well declared vocabularies been used, thus not a best practice.

[emmerbodc]

Discussed in vocab meeting 06/09/2024 - agreed to remove co-ordinate reference in LAT LON and LAT_GPS and LON_GPS OG1 concepts. Action @danibodc to update OG1 concepts Action @vturpin to update manual Lat long name

[danibodc]

Updates to the definitions of the 4 OG1 terms have been made. Queued for publication and will be live on the NVS from tomorrow morning.