[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Paper on OpenDAP



Peter-

Peter Cornillon wrote:
I've added dods-tech to this thread. Dods-techers, we are discussing recommendation of a metadata standard to facilitate use of OPeNDAP accessible data in the Earth Sciences. One starting point might be to recommend the use of CF or COARDS. Any thoughts on this?

While COARDS is a good start, it does not always have the
metadata necessary for complete representation of the semantics
of the data.  CF (which is supposed to be a superset of COARDS)
provides more of the information necessary.

For example, COARDS only suggests udunits compliancy for
unit specifications, but does not mandate it as CF does.
That means that users of COARDS could use the non-standard
abbreviation "C" for Celsius or "mb" for millibar and still
be COARDS compliant.  However, in the international standards
that udunits complies with, these are the abbreviations for
Coulomb and millibit.

Another example would be something like atmospheric sigma
coordinates.  COARDS does not specify how one could take
sigma to pressure, nor does it have a way of specifying what
other variables are necessary for this transformation.

For our IDV work, we strongly suggest CF compliance because
it supports more of the semantic information necessary to
use the data.

Don
*************************************************************
Don Murray                               UCAR Unidata Program
address@hidden                        P.O. Box 3000
(303) 497-8628                              Boulder, CO 80307
http://www.unidata.ucar.edu/staff/donm
        "Time makes everyone interesting, even YOU!"
*************************************************************