MIWP-8 4th virtual meeting 2015-01-08 14.00-16.00 CET

Connection details

 

Agenda

  • Status of work in ongoing tickets (group-leaders)
  • Managing copies of masterdocument (Michael Ö)
  • Upcoming meeting dates (Michael Ö)

 

 

DRAFT Minutes

 
 
 

Managing of template 

Suggestion is that each group have a copy of the word-template.
 
Ine: There will be some groups working on same paragraphs.
So maybe we need some kind of versioning of documents.
We have to identity subgroups that will work on same chapters.
eg information on referencesystems.  
Michael Ö with get back with details on this.
Meanwhile work can start on the contens
 

(J) Metadata for SDS


Ine:
Tickets are added to Redmine for each element that has to be added to 
make it easier to work separetely on each element.
There are now 7 additional tickets.
 
Ine: Complex issues are Service Locator where IR and TG have different definitions.
We need to update tickets together in group.
 
Michael: The suggested implementation for SDS is defining an extention to ISO19115.
Have there been a discussion on possible way   
 
Ine: We have an overlap between Service type and Service Category defined in SDS TG.
 
We have now a draft SDS TG v 3.1. Will that Technical guidelines be published  in parallell with the updates of TG Metadata.
 
 
Marc: Official European profile of ISO19115.  !!!
 
Ine: To be able to implement SDS without creating an extention to iso1915 we have to look into the 
IR for SDS and see if that can be changed. 
 

(D) Integrate metadata for interoperability into the Metadata TG

 
Michael: This ticket is only for the six interoperability elements that are already specified 
in dataspecifications. The other optional elements that are 
in ticket  MIWP-8 (E-1) Integrate Theme specific metadata
James has some connection issues so Peter reports on this issue.
Peter: We have uploaded a number of tables showing how this information have been dealt with.
James uploaded a template similar to TG that we have  inserted the information into.
We took the information from tables and matched them against ISO 19115.
A draft mapping-table is created (2014-12-16) for this six elements mapped against ISO .
 

MIWP-8 (I) Language neutral identifiers

 
Michael: This relates mainly to element 2.9.2 Conditions applying to access and use 
The TG defines a fixed text in English for defining 
"no conditions apply"
"conditions unknown"
 
Marc: Heavy discussion in France. Is metadata for machines or humans ? In France it was defined as mainly for humans.
There are no plans to do any translations to English.
 
Michael: The issue here is not about translating metadata to any langauge (eg English). The problems is when querying the inspire Geodataportal 
For the values "no conditions apply" "conditions unknown" the translation between European languages does not give a consistant result.
The current suggestion on wiki is to handle the values in mutilingual codelist.
Martin Seiler: This type of content should go into a codelist
Marc Leobet: +1 Martin
Antonio Rotundo: +1 Martin
Javier Nogureas: +1 Martin
Ine de Visser: +1 Martin
Manfred Mittlboeck: +1
Peter Kochmann: +1
 
So the proposed way forward is to create a multilingual codelist for this and put that into the registry.
We will continue that in next steps for the issue.

 

MIWP-8 (L) Unique Resource Identifier

 
Martin: There are multiple discussions ongoing.
A discussion have been on using RS_Identifier or MD_Identifier.
It is proposed to use MD_Identifier
 
Martin: Also some broader discussions on identifiers have been done on
- creating identifiers using the http:scheme
- stable  persistant identifiers
- resolvable identifiers. 
 
Michael: These are all important but is right now not directly in scope for TG Metadata since there are a an other MIG-t subgroup (MIWP-4: Identifiers) that will work on these issues.
 
Chris: This group have currently not started and will probably not start before the first release of TG Metadata is published. It will work high-level use-cases for Inspire-identifiers.
 
Martin: We should discuss the wider scope and define usecases for these identfiers and how it relates to eg Linked data. Its important that we know how technical descions ar taken.
It's unfortunate if this groups make updates on Coupled resources and the later the  patterns for identifiers are changed in later work.
 
Chris:The desicions on major changes will probably have to be done in MIG-t
Martin: We don't want to make changes that later are declined by MIG-t.
Michael: This is the way our sub-groups should work in generall. Always lift the discussion to MIG-t if we do changes that could have influence that change the architecture for Inspire.
 
Michael: We should first finalize the issue on MD_Identifier and RS_Identifier, then we can lift the wider issues on the syntax for identifers that should be stored in this element.

 

MIWP-8 (M) Coupled resources
 

Michael: We suggest use the xlink:href points 
 
Kristian: The  initial CSW ISO AP defined that the uuidref should be referenced. Then later within the Inspire-work the idea came up to reference the metadatarecord using the xlink:href. Within the Inspire its preferred to use the xlink:href.
 
Michael: So by using the xlink:href we should be fine.
 
Martin: It's important to note that the uuidref should point to the resourceidentifier and not the fileidentifier.  In germany we have no a serviceinterface wheer we can send in 
the resourceidentifier and get metadata returned for that. From reading the standards it's right now very hard for anyone not beeing an standards-expert.
 
Marc: In France we use in practice UUID of metadatafile to get the resource.  We are planning to use process our services to find all records that the services work on.
It's large task to find a metadatarecord for each layer.
 

Michael: If an xlink:href is pointing to the metadatarecord then a tool can generate the uuidref that should point to the resourceidentifier.
Kristian: There is a clause in page 39 of  OpenGIS® Catalogue Services Specification 2.0.2 - ISO Metadata Application Profile
That says

To simplify catalogue mining each MD_DataIdentification instance being part of aMD_Metadata instance must have an
identifier having a code value that is equal to thefileIdentifier of the owning MD_Metadata instance.

 
Martin: We should create usecase for this case also so we know how the standards intentions are followed.
 

MIWP-8 (D) Integrate metadata for interoperability into the Metadata TG

 
James has some connection issues so Peter reports on this issue.
Peter: We have uploaded a number of tables showing how this information have been dealt with.
James uploaded a template similar to TG that we have  inserted the information into.
We took the information from tables and matched them against ISO 19115.
A draft mapping-table is created (2014-12-16) for this six elements mapped against ISO .
Marc: From French side we have had some issues mapping the six elements against ISO.
The complicated  element among them are Topological Consistency.
Michael: The way forward here is to complete the template doc with tabular schemas as in existing TG Metadata.
 
Michael: When it comes to Topological Consistency it is currently not known how we can integrate quality reports based on ISO19157
together with qualityreports based on ISO19115. If no new information is introduced we should not use them for reporting quality.
 
 

MIWP-8 (E-1) Integrate Theme specific metadata

 
Michael: What are our solution here? Shoudl we make an overview of existsing elements like in attached Excel-sheets or
should we do a detailed mapping for each element that occur in dataspecifications ?
Peter: We can do the same way as for the six interoparability elements. 
We can in an annex to the TG add all additional elements that exists. There are though some open issues regarding some elements since they are not
very clearly described in the data specifications. We can add a detailed datadicionary for all additional elements. They are 20 elements on total.
 
 
 

MIWP-8 (A) Conditions applying to access and use

 
Marc: We have different interpretation of how to use MD_Constraints and MD_LegalConstraints
Everybody is using the example of MD_Constraints//useLimitation
The only case in France where we have a legal useLimitation is for navigation.
Next step would be to discuss in group all the different interpretations of these elements and see of we can find a common solution.
We may need some support from ISO to get feedback on how these elements should be interpreted.
 
Geraldine: We can not abuse the ISO-elements to get it fit Inspire
Martin: +1 to Geraldine.
Michael: I could see a possible solution to use MD_LegalConstraints/useLimitation for Inspire Conditions applying to access and use and
MD_Constraints/useLimitation as long it does not abuse the ISO-intention. We should investigate the original modelling of ISO for 
the semantic difference between MD_Constraints/useLimitation and MD_LegalConstraints/useLimitation
 

Upcoming meeting dates

Are published on the wiki
 
Communication technology
We have had some issues with connecting through Adobe Connect.
It would be great if members could connect ca 10 mins before meeting.