[Geoserver-devel] New community module: CSW

Hi all,
I’m writing to ask space for a CSW (Catalog Services for the Web) community module.

The module should form the core for providing catalog services out of GeoServer.
It won’t be, at least at the beginning, a stand-alone CSW server, so not a competitor
to GeoNetwork or PyCSW, but we are going to try and keep things generic enough that
it can develop in whatever direction we see fit in the future.

What I need to develop is in fact a “proxy front end”, that is, a protocol translator between
the CSW standard and a internal catalog using its own proprietary protocol, so the focus
is on providing all the protocol level implementation, and an interface, similar in the
intents to DataStore, which would allow to plug different kinds of backends onto GeoServer.

Luckily enough it seems that catalog matches well enough a number of subsystems that
we already have in place:

  • querying can be done via OGC Filter of CQL, which we both support
  • records returned can have a varying structure, they can be base dublin core, iso or
    ebRim, in all cases it seems we have a good way of representing them by using
    complex features

(Jody mentioned we have Recod, RecordSchema in gt-api, but I’d prefer to strengthen
the usage of complex features and use one less concept in our coding rather than
to disperse ourselves with new ones that are mere interfaces without an implementation)

The backend plugin may looking something like this:

interface CatalogStore {

/**

  • Well known base record types

*/

FeatureType CSW_RECORD_TYPE = …; /* This one is mandatory for all implementations */

FeatureType ISO_19115_RECORD_TYPE = …;

FeatureType EBRIM_REGISTRY_PACKAGE_TYPE = …;

/**

  • Returns the supported record types

*/

FeatureType getRecordSchemas() throws IOException;

/**

  • Queries a specific record type using the GeoTools Query object (which contains type name,

  • attribute selection

FeatureCollection getRecords(Query q, Transaction t) throws IOException;

List getDomain(Name typeName, Name attributeName);

List addRecord(Feature f, Transaction t) throws IOException;

void deleteRecord(Filter f, Transaction t) throws IOException;

void updateRecord(Name typeName, Name attributeNames, Object attributeValues, Filter filter, Transaction t) throws IOException;

CatalogCapabilties getCapabilities()

}

The CatalogStore would advertise the type of records it offers, including, but not limited to,
CSW (basically, Dublic Core), ISO 19115, and ebRIM, and would return features following
those schemas when getRecords is invoked.

The resulting FeatureCollection would be sent to an output format for encoding, for starters
and CITE certification we’ll need CSW, and I’m going to do some work on ebRIM since
what I have to expose from legacy is a ebRIM for earth observation compliant catalog.

The getDomain operation would support the same named operation, allowing the store
to do anything from a seq scan in the records to a distinct over a database, depending
on the storage used by the backend.

The add/delete/update operations are there to support CSW transactions, which is something
we don’t need out of the box, but it’s nice to have them in the interface anyways.

Finally, the Catalog capabilties class would be a was like QueryCapabilities, for starters
it would advertise support for write methods and the list of record properties that the store can
enumerate the domain of (the ones that can be called on GetDomain).

As you can see there is no harvest operation, this is because a harvest, if one day we will want it,
could be implemented as a service on top of CatalogStore that would look for the record information
remotely and perform transactions on the CatalogStore to update/add/remove the records.

Implementation wise I think we’re are going to create for GeoServer the bridge to the catalog,
but also a sort of “property file store for catalog”, that is a simple implementation reading from
xml files on disk (one per record), probably supporting only dublin core for starters.
This would help in testing, but would also give us a “pure record” storage (without data)
that we can then use to power the CITE tests, which demand for specific records with
a specific structure.
It would also serve as a simple example of how to implement a CatalogStore, just like
property data store did in the past.

CatalogStore at this moment would be part of GeoServer, but I would not mind pushing
it back later into GeoTools, especially if somebody shows up that wants to code a
CSW client around it.

The implementation should start later during this week

Feedback welcomed.

Cheers
Andrea

==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information.

Ing. Andrea Aime
@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 962313
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it


Hi Andrea,

sounds exciting.

By ISO 1915 schema I assume you mean ISO 19139, being it's current
schema realization (19115 is an abstract spec?).

I'll make sure to get some time this week to merge the GSIP69 branch
onto master so that you can experiment with OGC Filter queries on the
internal Catalog.

And good luck with schema conversions, it's gonna be fun :slight_smile:

Cheers,
Gabriel

On Mon, Aug 20, 2012 at 12:11 PM, Andrea Aime
<andrea.aime@anonymised.com> wrote:

Hi all,
I'm writing to ask space for a CSW (Catalog Services for the Web) community
module.

The module should form the core for providing catalog services out of
GeoServer.
It won't be, at least at the beginning, a stand-alone CSW server, so not a
competitor
to GeoNetwork or PyCSW, but we are going to try and keep things generic
enough that
it can develop in whatever direction we see fit in the future.

What I need to develop is in fact a "proxy front end", that is, a protocol
translator between
the CSW standard and a internal catalog using its own proprietary protocol,
so the focus
is on providing all the protocol level implementation, and an interface,
similar in the
intents to DataStore, which would allow to plug different kinds of backends
onto GeoServer.

Luckily enough it seems that catalog matches well enough a number of
subsystems that
we already have in place:
- querying can be done via OGC Filter of CQL, which we both support
- records returned can have a varying structure, they can be base dublin
core, iso or
  ebRim, in all cases it seems we have a good way of representing them by
using
  complex features

(Jody mentioned we have Recod, RecordSchema in gt-api, but I'd prefer to
strengthen
  the usage of complex features and use one less concept in our coding
rather than
  to disperse ourselves with new ones that are mere interfaces without an
implementation)

The backend plugin may looking something like this:

interface CatalogStore {

/**

  * Well known base record types

  */

FeatureType CSW_RECORD_TYPE = ...; /* This one is mandatory for all
implementations */

FeatureType ISO_19115_RECORD_TYPE = ...;

FeatureType EBRIM_REGISTRY_PACKAGE_TYPE = ...;

/**

  * Returns the supported record types

  */

FeatureType getRecordSchemas() throws IOException;

/**

  * Queries a specific record type using the GeoTools Query object (which
contains type name,

  * attribute selection

FeatureCollection getRecords(Query q, Transaction t) throws IOException;

List<Object> getDomain(Name typeName, Name attributeName);

List<FeatureId> addRecord(Feature f, Transaction t) throws IOException;

void deleteRecord(Filter f, Transaction t) throws IOException;

void updateRecord(Name typeName, Name attributeNames, Object
attributeValues, Filter filter, Transaction t) throws IOException;

    CatalogCapabilties getCapabilities()

}

The CatalogStore would advertise the type of records it offers, including,
but not limited to,
CSW (basically, Dublic Core), ISO 19115, and ebRIM, and would return
features following
those schemas when getRecords is invoked.

The resulting FeatureCollection would be sent to an output format for
encoding, for starters
and CITE certification we'll need CSW, and I'm going to do some work on
ebRIM since
what I have to expose from legacy is a ebRIM for earth observation compliant
catalog.

The getDomain operation would support the same named operation, allowing the
store
to do anything from a seq scan in the records to a distinct over a database,
depending
on the storage used by the backend.

The add/delete/update operations are there to support CSW transactions,
which is something
we don't need out of the box, but it's nice to have them in the interface
anyways.

Finally, the Catalog capabilties class would be a was like
QueryCapabilities, for starters
it would advertise support for write methods and the list of record
properties that the store can
enumerate the domain of (the ones that can be called on GetDomain).

As you can see there is no harvest operation, this is because a harvest, if
one day we will want it,
could be implemented as a service on top of CatalogStore that would look for
the record information
remotely and perform transactions on the CatalogStore to update/add/remove
the records.

Implementation wise I think we're are going to create for GeoServer the
bridge to the catalog,
but also a sort of "property file store for catalog", that is a simple
implementation reading from
xml files on disk (one per record), probably supporting only dublin core for
starters.
This would help in testing, but would also give us a "pure record" storage
(without data)
that we can then use to power the CITE tests, which demand for specific
records with
a specific structure.
It would also serve as a simple example of how to implement a CatalogStore,
just like
property data store did in the past.

CatalogStore at this moment would be part of GeoServer, but I would not mind
pushing
it back later into GeoTools, especially if somebody shows up that wants to
code a
CSW client around it.

The implementation should start later during this week

Feedback welcomed.

Cheers
Andrea

--

Our support, Your Success! Visit http://opensdi.geo-solutions.it for more
information.

Ing. Andrea Aime
@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 962313
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it

-------------------------------------------------------

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel

--
Gabriel Roldan
OpenGeo - http://opengeo.org
Expert service straight from the developers.

On Mon, Aug 20, 2012 at 5:59 PM, Gabriel Roldan <groldan@anonymised.com> wrote:

Hi Andrea,

sounds exciting.

By ISO 1915 schema I assume you mean ISO 19139, being it’s current
schema realization (19115 is an abstract spec?).

Eh, you may be right, I did not look much into it, as I have to make a ebRIM
binding, and I think Niels will work on basic CSW/Dublin Core instead (but I may be wrong)

I’ll make sure to get some time this week to merge the GSIP69 branch
onto master so that you can experiment with OGC Filter queries on the
internal Catalog.

I won’t be doing that part, probably Niels will

Cheers
Andrea

==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information.

Ing. Andrea Aime
@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 962313
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it


Very interesting,

+1 Andrea,

2012/8/20 Andrea Aime <andrea.aime@anonymised.com>:

Hi all,
I'm writing to ask space for a CSW (Catalog Services for the Web) community
module.

The module should form the core for providing catalog services out of
GeoServer.
It won't be, at least at the beginning, a stand-alone CSW server, so not a
competitor
to GeoNetwork or PyCSW, but we are going to try and keep things generic
enough that
it can develop in whatever direction we see fit in the future.

What I need to develop is in fact a "proxy front end", that is, a protocol
translator between
the CSW standard and a internal catalog using its own proprietary protocol,
so the focus
is on providing all the protocol level implementation, and an interface,
similar in the
intents to DataStore, which would allow to plug different kinds of backends
onto GeoServer.

Luckily enough it seems that catalog matches well enough a number of
subsystems that
we already have in place:
- querying can be done via OGC Filter of CQL, which we both support
- records returned can have a varying structure, they can be base dublin
core, iso or
  ebRim, in all cases it seems we have a good way of representing them by
using
  complex features

(Jody mentioned we have Recod, RecordSchema in gt-api, but I'd prefer to
strengthen
  the usage of complex features and use one less concept in our coding
rather than
  to disperse ourselves with new ones that are mere interfaces without an
implementation)

The backend plugin may looking something like this:

interface CatalogStore {

/**

  * Well known base record types

  */

FeatureType CSW_RECORD_TYPE = ...; /* This one is mandatory for all
implementations */

FeatureType ISO_19115_RECORD_TYPE = ...;

FeatureType EBRIM_REGISTRY_PACKAGE_TYPE = ...;

/**

  * Returns the supported record types

  */

FeatureType getRecordSchemas() throws IOException;

/**

  * Queries a specific record type using the GeoTools Query object (which
contains type name,

  * attribute selection

FeatureCollection getRecords(Query q, Transaction t) throws IOException;

List<Object> getDomain(Name typeName, Name attributeName);

List<FeatureId> addRecord(Feature f, Transaction t) throws IOException;

void deleteRecord(Filter f, Transaction t) throws IOException;

void updateRecord(Name typeName, Name attributeNames, Object
attributeValues, Filter filter, Transaction t) throws IOException;

    CatalogCapabilties getCapabilities()

}

The CatalogStore would advertise the type of records it offers, including,
but not limited to,
CSW (basically, Dublic Core), ISO 19115, and ebRIM, and would return
features following
those schemas when getRecords is invoked.

The resulting FeatureCollection would be sent to an output format for
encoding, for starters
and CITE certification we'll need CSW, and I'm going to do some work on
ebRIM since
what I have to expose from legacy is a ebRIM for earth observation compliant
catalog.

The getDomain operation would support the same named operation, allowing the
store
to do anything from a seq scan in the records to a distinct over a database,
depending
on the storage used by the backend.

The add/delete/update operations are there to support CSW transactions,
which is something
we don't need out of the box, but it's nice to have them in the interface
anyways.

Finally, the Catalog capabilties class would be a was like
QueryCapabilities, for starters
it would advertise support for write methods and the list of record
properties that the store can
enumerate the domain of (the ones that can be called on GetDomain).

As you can see there is no harvest operation, this is because a harvest, if
one day we will want it,
could be implemented as a service on top of CatalogStore that would look for
the record information
remotely and perform transactions on the CatalogStore to update/add/remove
the records.

Implementation wise I think we're are going to create for GeoServer the
bridge to the catalog,
but also a sort of "property file store for catalog", that is a simple
implementation reading from
xml files on disk (one per record), probably supporting only dublin core for
starters.
This would help in testing, but would also give us a "pure record" storage
(without data)
that we can then use to power the CITE tests, which demand for specific
records with
a specific structure.
It would also serve as a simple example of how to implement a CatalogStore,
just like
property data store did in the past.

CatalogStore at this moment would be part of GeoServer, but I would not mind
pushing
it back later into GeoTools, especially if somebody shows up that wants to
code a
CSW client around it.

The implementation should start later during this week

Feedback welcomed.

Cheers
Andrea

--

Our support, Your Success! Visit http://opensdi.geo-solutions.it for more
information.

Ing. Andrea Aime
@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 962313
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it

-------------------------------------------------------

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel

--
Francesco Izzi
CNR - IMAA
geoSDI
Direzione Tecnologie e Sviluppo

C.da S. Loja
85050 Tito Scalo - POTENZA (PZ)
Italia

phone: +39 0971427305
fax: +39 0971 427271
mob: +39 3203126609
mail: francesco.izzi@anonymised.com
skype: neofx8080

web: http://www.geosdi.org

Geoserver provides some catalog metadata already plus a URL link but a
very uneasy mapping. We have working at trying to sync arcGIS metadata
(ISO or Inspire), geoserver metadata, and Geonetwork catalogues. Better
catalog data in geoserver would surely help

Notice: This email and any attachments are confidential. If received in error please destroy and immediately notify us. Do not copy or disclose the contents.

On Mon, Aug 20, 2012 at 10:41 PM, Phil Scadden <p.scadden@anonymised.com> wrote:

Geoserver provides some catalog metadata already plus a URL link but a
very uneasy mapping. We have working at trying to sync arcGIS metadata
(ISO or Inspire), geoserver metadata, and Geonetwork catalogues. Better
catalog data in geoserver would surely help

The main driver here is not much to provide more metadata, but to provide
the right protocol to discover data in GeoServer without having to hunt it
into a capabilities document (e.g., queriability).

I’m not sure if the team working on exposing the catalog on CSW is going to
add metadata (I’ll be working on the ebRIM/EO proxy model) but I’m pretty
sure having a CSW will trigger more request for more metadata eventually,
so if it’s not going to happen right away I’m confident it will eventually
happen as a side effect of having a CSW around

Cheers
Andrea

==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information.

Ing. Andrea Aime
@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 962313
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it


On Mon, Aug 20, 2012 at 9:36 AM, Andrea Aime <andrea.aime@anonymised.com> wrote:

On Mon, Aug 20, 2012 at 5:59 PM, Gabriel Roldan <groldan@anonymised.com> wrote:

Hi Andrea,

sounds exciting.

By ISO 1915 schema I assume you mean ISO 19139, being it’s current
schema realization (19115 is an abstract spec?).

Eh, you may be right, I did not look much into it, as I have to make a ebRIM
binding, and I think Niels will work on basic CSW/Dublin Core instead (but I may be wrong)

To clarify what Niels is doing (through a contract with OpenGeo) will be adding support for the ISO metadata application profile for CSW 2.0.2. As described here:

http://portal.opengeospatial.org/files/?artifact_id=21460

I’ll make sure to get some time this week to merge the GSIP69 branch
onto master so that you can experiment with OGC Filter queries on the
internal Catalog.

I won’t be doing that part, probably Niels will

Cheers

Andrea

==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information.

Ing. Andrea Aime
@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 962313
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it



Live Security Virtual Conference
Exclusive live event will cover all the ways today’s security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/


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.

Phil, I agree, that was my reaction as well!

On 21/08/12 04:41, Phil Scadden wrote:

Geoserver provides some catalog metadata already plus a URL link but a
very uneasy mapping. We have working at trying to sync arcGIS metadata
(ISO or Inspire), geoserver metadata, and Geonetwork catalogues. Better
catalog data in geoserver would surely help

--
Ben Caradoc-Davies <Ben.Caradoc-Davies@anonymised.com>
Software Engineer
CSIRO Earth Science and Resource Engineering
Australian Resources Research Centre

On Tue, Aug 21, 2012 at 1:13 AM, Justin Deoliveira <jdeolive@anonymised.com> wrote:

On Mon, Aug 20, 2012 at 9:36 AM, Andrea Aime <andrea.aime@anonymised.com> wrote:

On Mon, Aug 20, 2012 at 5:59 PM, Gabriel Roldan <groldan@anonymised.com> wrote:

Hi Andrea,

sounds exciting.

By ISO 1915 schema I assume you mean ISO 19139, being it’s current
schema realization (19115 is an abstract spec?).

Eh, you may be right, I did not look much into it, as I have to make a ebRIM
binding, and I think Niels will work on basic CSW/Dublin Core instead (but I may be wrong)

To clarify what Niels is doing (through a contract with OpenGeo) will be adding support for the ISO metadata application profile for CSW 2.0.2. As described here:

http://portal.opengeospatial.org/files/?artifact_id=21460

Ah nice, so in the end we’ll have some basic version of all three? It is my understanding that in order
to pass certification one also has to support CSW Dublin Core derived basic metadata records

Cheers
Andrea

==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information.

Ing. Andrea Aime
@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 962313
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it


To reiterate a bit - in GeoServer we aren’t trying to replace GeoNetwork, to be a full catalog ‘solution’. We just want to expose GeoServer holdings as CSW. So one would use something like GeoNetwork in a full SDI that requires full metadata and complex editing of that and aggregating many services. But CSW in GeoServer should make it easier for GeoNetwork to harvest information from GeoServers, and to just follow a philosophy of keeping the metadata close to the data. We definitely don’t want to make a full metadata editor like GeoNetwork offers, just want to handle ‘search’ of data layers that can be accessed from GeoServer well.

My personal hope is that GeoNetwork continues its work to split up all the functionality it does, so it can interoperate better. In the GeoNetwork package you get a great CSW (and a number of other protocols to search), a portal builder toolkit, a metadata editor and a set of ‘harvesting’ tools. GeoServer will only look to do basic CSW. So if you want more search protocols you’d probably want to sync your GeoServer with GeoNetwork. And the other 3 things that GeoNetwork does should remain outside of GeoServer. We also want to build better ‘syncing’ with GeoNetwork, so one can edit service level metadata in either place and have it sync.

Chris

On Mon, Aug 20, 2012 at 5:06 PM, Andrea Aime <andrea.aime@anonymised.com> wrote:

On Mon, Aug 20, 2012 at 10:41 PM, Phil Scadden <p.scadden@anonymised.com> wrote:

Geoserver provides some catalog metadata already plus a URL link but a
very uneasy mapping. We have working at trying to sync arcGIS metadata
(ISO or Inspire), geoserver metadata, and Geonetwork catalogues. Better
catalog data in geoserver would surely help

The main driver here is not much to provide more metadata, but to provide
the right protocol to discover data in GeoServer without having to hunt it
into a capabilities document (e.g., queriability).

I’m not sure if the team working on exposing the catalog on CSW is going to
add metadata (I’ll be working on the ebRIM/EO proxy model) but I’m pretty
sure having a CSW will trigger more request for more metadata eventually,
so if it’s not going to happen right away I’m confident it will eventually
happen as a side effect of having a CSW around

Cheers

Andrea

==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information.

Ing. Andrea Aime
@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 962313
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it



Live Security Virtual Conference
Exclusive live event will cover all the ways today’s security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/


Geoserver-devel mailing list
Geoserver-devel@anonymised.comsts.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel

On 23/08/2012 6:31 a.m., Chris Holmes wrote:

But CSW in GeoServer should make it easier for GeoNetwork to harvest
information from GeoServers, and to just follow a philosophy of
keeping the metadata close to the data.

Yes, this is precisely my interest. There is this complicated problem
with metadata - where to keep it. Ideally, you keep it with the data and
let likes of geonetwork harvest it. Ditto for geoserver - if its
metadata handling mapped better to source and to geonetwork, then
keeping things in sync is easily. The problem I have is that geoserver
can use all kind of data sources. Geonetwork can harvest from say SDE
and from geoserver but how to control the harvester when the same
dataset is in arcSDE and also exposed in Geoserver? On the other hand,
for some data sources, geoserver lets us build metadata when there is no
standard close to the data and no harvester for geonetwork. We have been
hacking around with geonetwork harvesters to resolve this. A consistent
picture and a way to expose to geonetwork harvesters whether the
metadata is created in geoserver or fetched from another source would be
good.
--
Phil Scadden, Senior Scientist GNS Science Ltd 764 Cumberland St,
Private Bag 1930, Dunedin, New Zealand Ph +64 3 4799663, fax +64 3 477 5232

Notice: This email and any attachments are confidential. If received in error please destroy and immediately notify us. Do not copy or disclose the contents.