mbjones
commented
3 years ago These look great @twhiteaker . I'd like to hear from others whether they think having separate formatIds for each of the CF versions is useful, or whether your proposal that all that is needed is to know it is a NetCDF file. I think the original logic was to be able to help software know that it would have to deal with different CF conventions, but that may indeed be overkill. Nevertheless, we never remove existing formats from the vocabulary that could have been used in some system. But we certainly don't have to continue all of the version-specific ones if we prefer a generic netcdf format. Let's get input from @datadavev @csjx @jclark @laijasmine @amoeba and others on their houghts on this.
datadavev
Format Metadata
Provide the standard metadata for the proposed format, ensuring that the id and name are unique and appropriate to the version of the format being proposed.
1.5:
CF-1.5NetCDF Climate and Forecast Metadata Convention, version 1.5DATAapplication/netcdfnc1.6:
CF-1.6NetCDF Climate and Forecast Metadata Convention, version 1.6DATAapplication/netcdfnc1.7:
CF-1.7NetCDF Climate and Forecast Metadata Convention, version 1.7DATAapplication/netcdfnc1.8 and up:
netCDF-CFNetCDF Climate and Forecast Metadata ConventionDATAapplication/netcdfncFormat description
Describe why a new format is needed, including items such as where the format type has been encountered, what software produces it, and what software can read it.
CF 1.5 through 1.8 are released. Since we have 1.0 through 1.4, we should add the rest. Starting with CF 1.8, a
Conventionsattribute indicating the convention used is required (previously it was optional) in the netCDF file, and so indicating that this is a CF file via formatName is enough, and you can leave it up to CF-aware software to interpret the version upon inspecting the file itself.Specification / Namespace documentation
Provide the location(s) of the documentation of the format specification or the namespace for the format or vocabulary.
http://cfconventions.org/
Checklist
image/pngis specific to one format, whereastext/xmlis not specific to one format)DATA,METADATA, orRESOURCEConsiderations
Describe or list any considerations that might impact the use of the format, or related issues that we should consider.
The
1.8 and upentry does not indicate in the format name that this is for 1.8 and up. Will this be confusing to users who are trying to choose a formatName? I was trying to be efficient and save future work with the generic1.8 and upversion, but I also have no guarantee that CF won't drop the conventions-identification requirement in the future (though I am almost certain they would not drop it).Stepping back a bit, I question whether the CF Conventions entries are even needed. I think when choosing a format, specifying nteCDF would be enough. Software that can read netCDF files, is either CF-aware or not. If aware, it will know what to do with the file when it sees it. If not, the end user may have to hold the software's hand a bit. But either way, I don't see how knowing whether CF is used ahead of time, is helpful.