Re: [thredds] NcML and the <geospatialCoverage> element

The document says to use Metadata_Conventions, not Conventions.  While what
you did makes (more) sense, it is not "correct".

On Jan 14, 2010 12:44 PM, "Roy Mendelssohn" <Roy.Mendelssohn@xxxxxxxx>
wrote:

There are metadata in our files that follows all of those conventions.  In
trying to relay that act, we felt this was the best way to relay that fact.
 So you will find the metadata required for CF, the Coastwatch conventions,
and the Unidata Data Discovery Conventions.

-Roy

On Jan 14, 2010, at 9:40 AM, Benno Blumenthal wrote: > OK, again trying
desperately to understand...