[CF-metadata] Question on WKT representation of CRS

From: Jonathan Gregory <j.m.gregory>
Date: Thu, 6 Oct 2011 16:40:43 +0100

Dear all

I agree that we should not reinvent in CF what has already been thought out
elsewhere if we can help it. This point is often made in the development of
CF. However I think we should restrict our attention to what is really needed
in CF i.e. for which there are already use-cases. In my opinion, importing
another convention "wholesale" without being aware of how it might conflict
with CF, or how it might be internally inconsistent, and without having methods
to identify such conflicts and resolve them, would be a bad move for CF. To
avoid it, we need a thorough understanding of the concepts for which we are
devising a convention. That need makes the development of CF hard work, but
its product is useful, I believe.

Best wishes

Jonathan
Received on Thu Oct 06 2011 - 09:40:43 BST