Ron,
We are missing an overall Services Architecture, that is the result of the glitch
we got when we went SOA about 2001 or thereabouts. There was one once in Topic 12
<http://portal.opengeospatial.org/files/?artifact_id=1221>
http://portal.opengeospatial.org/files/?artifact_id=1221, but that was never
complete, nor completely SOA and is definitely out of date (2001).
I also agree that WFS should be the core of all data access services,
including coverages, images, sensors, observations, licenses (okay that might
be pushing the issue a bit) and metadata (a catalog entry is also feature).
Which is one of the reasons that WFS might need to be freed up from
restrictions on the type of schema it can use in its responses, since there are
already standard schemata for some of this, at least in the non-geo world, that
could be extended to cover our issues (sort of what we did in GeoREL and
GeoXACML).
Regards,
John
From: owner-galeon@xxxxxxxxxxxxxxxx [mailto:owner-galeon@xxxxxxxxxxxxxxxx] On
Behalf Of Ron Lake
Sent: Thursday, May 10, 2007 1:42 PM
To: John Herring; Carl Reed OGC Account; Simon.Cox@xxxxxxxx;
p.baumann@xxxxxxxxxxxx
Subject: RE: OGC Ottawa TC meeting highlights
John:
I don't disagree - except we don't seem to have a top down model with respect
to services and how those services should fit together and what they are all
for. I think we have a reasonably coherent model for data in the abstract
specification. That is the top down part that I see as missing. I do agree
that everything is a feature - and most especially coverages and observations -
and to me a consequence of that ought to be that a WCS is a kind of WFS as is a
SOS.