[CF-metadata] Dimensionality of Label Variables

From: Jonathan Gregory <j.m.gregory>
Date: Mon, 3 Sep 2012 12:53:24 +0100

Dear Michael and Jim

I agree that section 6.1 of the convention and of the conformance document are
inconsistent. I don't have an example of a multi-dimensional string-valued
auxiliary coordinate variable, but the convention does not exclude this and
it might be useful. However the conformance document does exclude it and the
start of the text of 6.1 talks about the use in a 1D case. So it's not clear
to me whether it's better to restrict the convention, or to generalise the
conformance requirements. I'd tend to the latter. What do you think? In either
case I think we could regard this as a defect to be corrected, rather than a
material change to the CF standard.

Cheers

Jonathan
Received on Mon Sep 03 2012 - 05:53:24 BST