PREDEPLOYMENT_CALIB_DATE, SENSOR_FIRMWARE_VERSION and other names that aren't in the Argo users manual and perhaps should be

[BrianKingNOC]

There are a variety of metadata that aren't in the Argo users manual, or are in the Argo users manual with names that might no longer be 'optimal'.

The first is SENSOR_FIRMWARE_VERSION which is on the list to be agreed by ADMT

Another example is PREDEPLOYMENT_CALIB_DATE This ought to be controlled as a proper parseable field. At present, some DACs encode this inside PREDEPLOYMENT_CALIB_COMMENT but with no fixed format way of doing so.

Other examples would include FLOAT_FIRMWARE_VERSION At present we have just FIRMWARE_VERSION in the manual, but that name is now incomplete. It was OK before we wanted to keep sensor firmware versions as well. Our notion of what we want to store has evolved, and it would be useful to have more completely descriptive names going forwards. @Annie Wong pointed out that we should have FLOAT_FIRMWARE_VERSION (presently FIRMWARE_VERSION) FLOAT_MANUAL_VERSION (presently MANUAL_VERSION) FLOAT_SERIAL_NO (already OK)

With the json meta files, we have the opportunity to overhaul the list of names that have become out of date.

We would hope that these would be adopted in the manual in due course, and before that there would need to be some translation when generating meta.nc for backwards compatibility.

If ADMT has appetite to change and update some of these names, then now would be a time to have a full review

[SBS-EREHM]

@BrianKingNOC Let's confirm with ADMT (e.g., Megan Scanderbeg)

• add SENSOR_MODEL_FiRMWARE
• add PREDEPLOYMENT_CALIB_DATE
• clarify PREDEPLOYMENT_CALIB_COMMENT as holdin non-parseable descriptive metadata.

Personally, I think PLATFORM/FIRMWARE_VERSION and PLATFORM/MANUAL_VERSION are ok and distinct from SENSOR/SNSOR_MODEL_FIRMWARE. I don't think we need to see every metadata item as a global variable having a unique name in a flat namespace. At this point, is the disruption of changing the name worth it?