So I finally sorted out the catalogue trinity last week in Perth - due
to a good update on the Resource/Publishing split from Justin last
week.
I am still stuck on the naming of things but progress has been made ...
- Workspace - basically folders for DataStores; no real significance
in terms of publication of data
- Maps - this defines how publication happens; the only thing
significant in terms of publication of data
- Profile - sets logging level and other application preferences/settings
My earlier confusion came from thinking the above concepts were tied
together (like you could publish the same map twice using different
workspaces). Instead I find they are a trinity each seperated from the
other that together form the geoserver configuration.
With this in mind:
- Maps are still a very difficult name - but they do each represent a
public "GeoServer" with as many OGC services enabled as needed.
- It would be *so* nice to refer to Map above as a "Service" and refer
to enabling/disabling the various WMS/WFS/WCS things as "Protocols"
enabled for that service. Out of all the naming ideas this one made
the most sense; indeed once we step beyond OGC (with REST etc..>) it
becomes even more clear.
- The concept of a Map is is exactly what RobA was calling a "profile"
(so that one could publish a profile for INSPIRE complete with the
correct FeatureTypes; Styles and so on) which could be downloaded and
added to your geoserver configuration. Backing this profile onto your
own data would be the interesting part.
- The current association of namespace to workspace was causing a lot
of trouble for people; offering lots of opportunities for confusion.
I agree that the terminology "map" is confusing. I like the terminology of "service" and "protocol" as it describes more accurately what the technical entities actually are, although the term service is quite tied to the idea of OGC services and I fear using it in this way would result in even more confusion.
You did bring up the term "context" to replace "map" which I kind of liked as well.
Jody Garnett wrote:
So I finally sorted out the catalogue trinity last week in Perth - due
to a good update on the Resource/Publishing split from Justin last
week.
I am still stuck on the naming of things but progress has been made ...
- Workspace - basically folders for DataStores; no real significance
in terms of publication of data
- Maps - this defines how publication happens; the only thing
significant in terms of publication of data
- Profile - sets logging level and other application preferences/settings
My earlier confusion came from thinking the above concepts were tied
together (like you could publish the same map twice using different
workspaces). Instead I find they are a trinity each seperated from the
other that together form the geoserver configuration.
With this in mind:
- Maps are still a very difficult name - but they do each represent a
public "GeoServer" with as many OGC services enabled as needed.
- It would be *so* nice to refer to Map above as a "Service" and refer
to enabling/disabling the various WMS/WFS/WCS things as "Protocols"
enabled for that service. Out of all the naming ideas this one made
the most sense; indeed once we step beyond OGC (with REST etc..>) it
becomes even more clear.
- The concept of a Map is is exactly what RobA was calling a "profile"
(so that one could publish a profile for INSPIRE complete with the
correct FeatureTypes; Styles and so on) which could be downloaded and
added to your geoserver configuration. Backing this profile onto your
own data would be the interesting part.
- The current association of namespace to workspace was causing a lot
of trouble for people; offering lots of opportunities for confusion.
Jody
------------------------------------------------------------------------------
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geoserver-devel
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
I agree that the terminology "map" is confusing. I like the terminology of "service" and "protocol" as it describes more accurately what the technical entities actually are, although the term service is quite tied to the idea of OGC services and I fear using it in this way would result in even more confusion.
You did bring up the term "context" to replace "map" which I kind of liked as well.
"Maps" are:
- a virtual server
- a publishing point
- a collection of layers
- a collection of OGC services
- a "nested" GeoServer?
- ...
not sure if any of the above helps in finding a better name. But
I agree "map" is confusing and should be replaced.
How does ESRI call the equivalent concept they have in IMS?
Cheers
Andrea
--
Andrea Aime
OpenGeo - http://opengeo.org
Expert service straight from the developers.
the term I have been using is 'service profile' - i.e. how a set of
users expect a service to behave
Context matches the OGC generalisation of a WMS context. It sort of
relates more to an invocation than a configuration though IMHO
so, if you take the context to be the user-centric viewpoint of
collections of resources which together are useful:
a service profile supports one or more user contexts, consisting of a
ordered set of invocations based on protocol bindings to presentations
(eg styling or formats) of the result of operations (filters) against
resources
whew! complicated but at least its a consistent set of terms that
don't seem too far from the status quo.
With a WMS, the ordered set provides value by building a map from
layers. In a WFS, its traversal of a nested graph of related features
to answer a particular question or get specific data. WCS is arguable
a special case of WFS with functions, with more interest in the value
range than relationship between feature types.
Rob
On Tue, Jun 30, 2009 at 3:51 PM, Justin Deoliveira<jdeolive@anonymised.com> wrote:
I agree that the terminology "map" is confusing. I like the terminology
of "service" and "protocol" as it describes more accurately what the
technical entities actually are, although the term service is quite tied
to the idea of OGC services and I fear using it in this way would result
in even more confusion.
You did bring up the term "context" to replace "map" which I kind of
liked as well.
Jody Garnett wrote:
So I finally sorted out the catalogue trinity last week in Perth - due
to a good update on the Resource/Publishing split from Justin last
week.
I am still stuck on the naming of things but progress has been made ...
- Workspace - basically folders for DataStores; no real significance
in terms of publication of data
- Maps - this defines how publication happens; the only thing
significant in terms of publication of data
- Profile - sets logging level and other application preferences/settings
My earlier confusion came from thinking the above concepts were tied
together (like you could publish the same map twice using different
workspaces). Instead I find they are a trinity each seperated from the
other that together form the geoserver configuration.
With this in mind:
- Maps are still a very difficult name - but they do each represent a
public "GeoServer" with as many OGC services enabled as needed.
- It would be *so* nice to refer to Map above as a "Service" and refer
to enabling/disabling the various WMS/WFS/WCS things as "Protocols"
enabled for that service. Out of all the naming ideas this one made
the most sense; indeed once we step beyond OGC (with REST etc..>) it
becomes even more clear.
- The concept of a Map is is exactly what RobA was calling a "profile"
(so that one could publish a profile for INSPIRE complete with the
correct FeatureTypes; Styles and so on) which could be downloaded and
added to your geoserver configuration. Backing this profile onto your
own data would be the interesting part.
- The current association of namespace to workspace was causing a lot
of trouble for people; offering lots of opportunities for confusion.
Jody
------------------------------------------------------------------------------
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geoserver-devel
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
------------------------------------------------------------------------------
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geoserver-devel
I agree that the terminology "map" is confusing. I like the terminology of "service" and "protocol" as it describes more accurately what the technical entities actually are, although the term service is quite tied to the idea of OGC services and I fear using it in this way would result in even more confusion.
You did bring up the term "context" to replace "map" which I kind of liked as well.
"Maps" are:
- a virtual server
- a publishing point
- a collection of layers
- a collection of OGC services
- a "nested" GeoServer?
- ...
not sure if any of the above helps in finding a better name. But
I agree "map" is confusing and should be replaced.
How does ESRI call the equivalent concept they have in IMS?
Not sure, but I think they call what we call a workspace a folder, which is know is something we discussed. As for a replacement name for map... How do others feel about "context". It is suitably abstract so it more or less describes the concept ;), and is not tied to any preconceptions.
Cheers
Andrea
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
On Wed, Jul 1, 2009 at 4:58 PM, Justin Deoliveira<jdeolive@anonymised.com> wrote:
Not sure, but I think they call what we call a workspace a folder, which is
know is something we discussed. As for a replacement name for map... How do
others feel about "context". It is suitably abstract so it more or less
describes the concept ;), and is not tied to any preconceptions.
Cheers
Andrea
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
On Wed, Jul 1, 2009 at 4:58 PM, Justin Deoliveira<jdeolive@anonymised.com> wrote:
Not sure, but I think they call what we call a workspace a folder, which is
know is something we discussed. As for a replacement name for map... How do
others feel about "context". It is suitably abstract so it more or less
describes the concept ;), and is not tied to any preconceptions.
Cheers
Andrea
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
But since that represents a change to every last thing I am not sure
my recommendation is any good. On the bright side I really like the
technical seperation here.
Jody
On Wed, Jul 1, 2009 at 5:31 PM, Justin Deoliveira<jdeolive@anonymised.com> wrote:
Yup, that looks like it, thanks for summing it up concisely Jody I think
I like (b) the best.
On Wed, Jul 1, 2009 at 4:58 PM, Justin Deoliveira<jdeolive@anonymised.com>
wrote:
Not sure, but I think they call what we call a workspace a folder, which
is
know is something we discussed. As for a replacement name for map... How
do
others feel about "context". It is suitably abstract so it more or less
describes the concept ;), and is not tied to any preconceptions.
Cheers
Andrea
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
I think I like (b) the best.