In many places WCS already uses GML. However, we (ie: Arliss) had to
simplify handling a little. It might turn out that GMLifying further
adds complexity at a stage where we still have conceptual discussion. I
would see this as a downstream encoding question.
-Peter
Ron Lake wrote:
It would be good then to have, for this limited case, a canonical
response encoding for a DescribeCoverage and GetCoverage operations
which is GML, thus laying the foundations for future extensions. If
this requires modifications to GML, these can be undertaken. This would
also bring WCS and WFS inline with one another.
Ron
