ANZLIC The Spatial Information Council  
[Jurisdictions][Home][Contact Us][Site Map][Site Search][Glossary]  
 

 

ASDI-L Mailing List Archive

From: Rob Atkinson ([email protected])
Date: Wed Jun 18 2003 - 11:20:37 EST


David's point is right about "information science" - one typically
starts with the premise that you understand the problem. To the best of
my knowledge, no-one has formally modelled the inter-relationships and
transactions that occur in the life-cyle of datasets and services, and
thius the metadata that describes this.

for example, the "FreeText" content model is obviously going to be
useless in itself to create any form of long-term usability of the
content of fields we expect to be machine traversable. ( c.f. Davids
experiment to define something to go in such fields. )

So, some sort of profile is required. So thats now an additional object
that has to be modelled throughout the lifecycle !

If we are going to create a short-term solution, it should be based on
the immediate needs of real services infrastructures. Look forward to
having an opportunity to share experiences with these, although there is
no obvious programme to engage with available expertise (e.g. Canada's
GeoConnections).

Regard
Rob Atkinson

David Crossley wrote:

>Bleys, Evert - BRS wrote:
>...
>
>
>>As you may be aware we have been playing with ISO 19115
>> (but that is of no major concern)
>>As a result we approve of any use of the
>>on-line resource options that are compatable with ISO
>>
>>What is also needed (and I am happy to try to coordinate)
>>is the dictionary of terms that enumerate the fields
>>The need to define the elements within the DTD
>>These elements are
>> TransOnLine (Container Element O N -)
>> Linkage (Simple Element M 1 FreeText )
>> Protocol (Simple Element O 1 FreeText )
>> ApplicationProfile (Simple Element O 1 FreeText )
>> Name (Simple Element O 1 FreeText )
>> Description (Simple Element O 1 FreeText )
>> Function (Simple Element O 1 Text from List)
>>The optional obligation is a hassle
>>I think most if not all should be mandatory
>>(if you are going to play - then get real)
>>
>>
>
>Yes, most should be mandatory within TransOnLine. However, we could
>just have guidelines to say: if you want to link to something, then here
>is how to. The key is to have very clear definitions of these fields and
>various real-life examples of their use.
>
>For the demo, the only two that we actually used are Linkage and
>ApplicationProfile. At the bottom of the interface page
>(http://www.indexgeo.net/zap/getmap/) there is a link to explain how
>we implemented this capability.
>
>
>
>> The ISO definitions are not overly helpful
>>I suggest that
>>Linkage
>> (ISO defines as "location for on-line access using URL or similar")
>> be the URL eg if wms then it is the address of the service
>>
>>
>
>Yes. Since each of our example metadata records describe individual
>layers, we link directly to the layer in our WMS service.
>
>
>
>>Protocol
>> (ISO defines as "Connection protocols
>> to be used")
>> be something like a list of mechanisms
>> e.g. FTP, WMS, WFS (need help here)
>>
>>
>
>Not sure what this field is meant for. It does look like
>it could be useful, given well-constrained content.
>
>
>
>>ApplicationProfile
>> (ISO defines as "Name of Application
>> profile that can be used with the
>> on-line Resource")
>> be something like a list of formats
>> e.g. OGC (need help here)
>>
>>
>
>Here are two examples of what we are using in that field:
> ogc:WMS-1.1.0-http-get-map
> ogc:WMS-1.1.0-http-get-capabilities
>I think that this is dangerous - too much information in one field.
>It would be better to have separate metadata fields.
>
>For our ASDD demo, we search in this field for the term "WMS" and
>the term "get-map". As you can see, we are using it as a catch-all
>... not good information science, but fine for a demo.
>
>
>
>>Name
>> (ISO defines as "Name specified
>> for the on-line resource within
>> the capability of the Protocol")
>> be the name of the layer or file
>> (Note that in the case of
>> FTP this would be redundant,
>> but for WMS critical)
>>
>>
>
>We have this a long text name for the resource. The actual layer
>name is embedded in the Linkage URL for WMS. Why do you say that
>this field is critical?
>
>
>
>>Description
>> (ISO defines as a "Text description
>> of what the on-line resource is/does")
>> (Note - this would be useful if/when
>> things go wrong)
>>
>>
>
>Yes, useful.
>
>
>
>>Function
>> (ISO has defined as one of:
>> download; information; offlineAccess; order; search
>> These are of little real value in this context.
>>
>>
>
>I agree that the authority list is way too limiting. With better
>code list values, this field might be useful.
>
>--David
>
>
>
>


[#top]