[CF-metadata] CF:featureType and Naming Conventions

From: Jonathan Gregory <j.m.gregory>
Date: Tue, 22 Feb 2011 11:10:02 +0000

Dear Martin

> here is my daily fanpost (don't worry I will be quiet again at some point;-):

Thank you for your energetic contributions to CF!
>
> in the discussion document on "discrete sampling geometries" a recommendation is made to introduce a CF:featureType attribute (with values of point, timeSeries, trajectory, etc.). While this is very reasonable in principle, it opens up the name space for CF attributes

In the current draft, the attribute is just featureType, because the
convention of identifying namespaces e.g. with a "CF:" prefix has not yet
been agreed.

> A minor issue: why did you choose attribute names which begin with lower case characters but have upper case chars embedded in the name? Wouldn't it be more logical to say "TimeSeries" instead of "timeSeries"?

The values of the featureType are not case-sensitive. (This is generally the
case with the possible values for keywords in CF attributes.)

Best wishes

Jonathan
Received on Tue Feb 22 2011 - 04:10:02 GMT