Dear Jamie
> > float temperature(realization,time,lat,lon):
> > temperature:coordinates = 'time lat lon metadata1' ;
> > char metadata1(realization,len100):
> > metadata1:standard_name="institution"; // for instance
> > metadata1:external_vocabulary = http://wmo.foo.int/identifierY
> >
> > I proposed that the new standard names would be the same as the new or
> > existing global and local attributes that we use for these purposes
> > (such as institution).
>
> I was wondering do we need _both_ the standard_name and the
> external_vocabulary? For instance is there a risk of redundancy? I
> think the potential for redundancy arises if an application can infer
> the kind of standard name type quantity the vocabulary refers to (either
> by knowing it hard coded, or through vocab URL naming conventions)? The
> main reason I can think of for including both is if an application has
> some idea of what to do with 'institution' without having to know what
> the individual names in the vocab mean (e.g. use it to slice an
> ensemble), but doesn't necessarily know that
> http://wmo.foo.int/identifierY is an institution type label.
Yes, I think that is a good reason. It might be that institution has values
which are sometimes standardised, sometimes not, but it's useful to have the
same standard name in all cases because an application will recognise it, and
so will a human.
> Something else I'm not so clear on is can the providers of a vocabulary
> mix the kind of things they are naming within a vocab list (e.g. have a
> single vocab containing both 'source' and 'institution'?) If the vocabs
> are being maintained externally I don't think you can control whether
> this happens or not.
Yes, I think this is a problem. We can only refer to external vocabularies
if we have negotiated with their owners that they contain the right things
in the format we require. This is why we adopted our own region list, by
reformatting a subset of what GCMD provided.
Cheers
Jonathan
Received on Wed Nov 15 2006 - 08:53:46 GMT
