[CF-metadata] Getting back to ensembles

From: Jonathan Gregory <j.m.gregory>
Date: Fri, 29 Dec 2006 08:40:09 +0000

Dear John

Thanks for your comments.

I think it is possible that the standard name list may be superseded at some
point by a set of vocabularies. This is worth thinking about, but it is a
complex question as you say and would take time. Indeed there are various
characteristics which distinguish different kinds of quantities, but so far
none which seem to argue for clear categories. I don't think we should be
delayed at this point by trying to draw distinctions for which we don't
apparently have a present requirement. I believe the principle of introducing
conventions for which there is actually a need has served us well; once you
need something, you know what it is, so it is possible to design an appropriate
way to do it. Here we seem to be trying to devise a convention when we don't
really have a practical need for it, and hence no clear idea of what it is.

Therefore I favour carrying on as we are. We can continue to add names for
more kinds of quantities to the standard_name list until we can see for sure
that we need something different. At that point we can work out a new
convention and map the old one to the new one.

Best wishes

Jonathan


----- Forwarded message from John Graybeal <graybeal at mbari.org> -----

> I agree that some terms are not useful to describe science values, and you may want to distinguish those terms from the others in the vocabulary. How you do that -- with separate vocabularies, or just by characterizing each term in the one vocabulary according to its ontological class -- is a design decision. Either approach could be used to pre-fill drop-down lists, which is useful. But I don't see how either will prevent selection errors.
>
> Note that this question generalizes -- CF is fundamentally about climate and forecast terms so far, but a lot of us are trying to apply it to more concepts and terms, and not just in-situ or scientific ones. To what extent (and how) will CF evolve into a set of vocabularies? While such an evolution may prove valuable or necessary, I agree you should have a guideline or process that describes how to distinguish terms.
>
> John
Received on Fri Dec 29 2006 - 01:40:09 GMT