>Date: Mon, 30 Mar 1998 11:54:09 -0600
>From: Tom Rink <rink@xxxxxxxxxxxxx>
In the above message, you wrote:
> > ((scalar1, scalar2), field)
> I think one thing to consider is if (scalar1, scalar2) could be
> interpreted as having a default CoordinateSystem. In that case,
> I dont thing they should be broken out.
In general, the scalar variables in a netCDF dataset won't have an
associated CoordinateSyste. Thus, should they be broken out?
Steve Emmerson <http://www.unidata.ucar.edu>