[CF-metadata] CF2 and standard names

[Jonathan Gregory]

----- Forwarded message from "Gross, Tom" <grosst at si.edu> -----
> Given the wealth of datums available, I am of the opinion that a surface
> such as, "sea_surface_height", should not include the datum in the
> standard name.  It is an attribute of the variable.

I agree with that. My posting may not have been clear. When I referred to a
standard name parameter, I meant some extra piece of information, *not* in the
standard name, which clarifies it. We have decided in principle that we need
to be able to do this for some standard names, but not agreed the mechanism.

Cheers

Jonathan