Discussion #3025

Usage of data set series in INSPIRE implementation and metadata

Added by Michael Lutz almost 3 years ago. Updated almost 3 years ago.

Status:New
Priority:Normal
Assignee:Robert Tomas
INSPIRE Theme:

Description

Prompted among other things by this discussion on the Thematic Clusters, we would like to discuss at the 44th MIG-T meeting about whether and how data set series are used in the implementation of INSPIRE in the Member States, and how these are documented in the metadata.

We would therefore like to ask you to provide the following input:

  • Are there any guidelines or common practices for the use of data set series in your countries?
  • For what purpose(s) are they used (e.g. time series, series of map sheets, collections of INSPIRE data sets from different regions in your country, …)
  • Have you encountered any implementation issues related to data set series?

qualityInstancesScope.png (185 KB) Michael Lutz, 28 Oct 2017 02:46 pm

DS_Series_Belgium.docx (318 KB) Michael Lutz, 28 Oct 2017 02:51 pm

2186

History

#1 Updated by Michael Lutz almost 3 years ago

  • Description updated (diff)

#2 Updated by Michael Lutz almost 3 years ago

Please find my answers below.
Are there any guidelines or common practices for the use of data set series in your countries?
Yes. There is an regulation regarding organization and maintenance of state geodetic and cartographic data sets.
For what purpose(s) are they used (e.g. time series, series of map sheets, collections of INSPIRE data sets from different regions in your country, …)
Regulation defines data set series for:
·       Aerial and satellite imagery,
·       Orthoimagery,
·       DEM.
In addition to this we use data set series concept for data sets that are not mentioned in regulation like:  
·       Series of topographic map sheets,
·       Series of thematic map sheets.
Have you encountered any implementation issues related to data set series?
Initially, there was a discussion how to define data set series for different types of spatial data. When this was sorted out we haven’t encountered any significant implementation issues so far.
 
Best regards,
Marcin Grudzień

#3 Updated by Michael Lutz almost 3 years ago

I am going to try to answer your questions, although I am afraid, the responses won’t be very useful:
 
•             Are there any guidelines or common practices for the use of data set series in your countries?
 
There aren´t guidelines.
 
•             For what purpose(s) are they used (e.g. time series, series of map sheets, collections of INSPIRE data sets from different regions in your country, …)
 
There aren´t guidelines.
 
•             Have you encountered any implementation issues related to data set series?
 
Several comments:
 
·         It’s very difficult to maintain thousands of sheet metadata so we decided to remove it from our catalogue. And, as far as I know, it is not mandatory.
·         Atom feeds are not fit for providing thousands of sheet files.
·         Map sheets are not part of INSPIRE themes.
·         We don’t know how to deal properly with time series (for instance, orthoimagery), regarding network services.
 
Best regards,
 
Emilio López Romero

#4 Updated by Michael Lutz almost 3 years ago

In Cyprus’ INSPIRE implementation we have both cases:
 
1.      Data for an INSPIRE theme (service) come from several sources/stakeholders.
2.      A single dataset/stakeholder cover one theme (service).
 
In the first case, metadata are separated so that there is an appropriate coverage per dataset/stakeholder.
 
In the second case, metadata correspond to the single dataset and service.
 
In addition, we have the case of Imagery (aerial ortho-photos and satellite images), where for the same service we have several time coverages per theme (service). It is important in this case to use this information for historical purposes.
 
Another example is the Geographical Grid Systems, which is not fully implemented yet. If you remember I raised a question during our meeting in Ghent. We intend to insert in this theme all the map grids from all available map series. These grids are very useful information for our users and they provide a good index to specific map series and individual maps. In the meeting I received a response that it is up to us to decide which information to include in the theme, and that we could include as many datasets as needed.
 

Regards,

Andreas Hadjiraftis

 

#5 Updated by Michael Lutz almost 3 years ago

reading the discussion in the thematic cluster,
I think you are searching something like the ISO 19115 DQ_Scope datatype in the context of the Quality scope element.
In this case, is possible to provide detail information and context for the several datasets of the serie in the same metadata file.
I'm sending in attachment a printscreen of the ISO 19115:2003 with the Quality section UML diagram.
The DQ_Scope datatype is composed by level (in use in Portugal), extent (in use in Portugal) and levelDescription, that could be the name of the dataset (not in use in Portugal).
But that kind of relation serie/datasets is not in use in Portugal, as far as I know.
Best regards,
Henrique Silva

#6 Updated by Michael Lutz almost 3 years ago

We do not have any gidelines or common practices for the use of data set series at the antional level. It is up to the organisations so they can have different approaches and purposes.

The biggest data provider in the Czech Republic, the Czech Office of Surveying and Cadastre, use data set series concept for the INSPIRE download services. For several INSPIRE themes is not possible to provide one dataset for whole Czech Republic or it would be so big (several hundred GB in GML) that it would be unusable in SW.

For example the ATOM service can not provide smaller unit than data set and if the data set would be the whole Czech Republic the user would have to download all 13 091 cadastral areas (almost 2 GB) instead of one or several cadastral areas (several hundret kB). Therefore we prefer the user's needs before 100% fulfilling the INSPIRE requirements.

The number of data sets for individual INSPIRE themes:
Theme                                Unit                        Number
Cadastral parcels                 Cadastral area         13091
Addresses                           Municipality              6259
Administrative units              CZ                                1
Buildings                             Municipality              6259
Transport networks               Map sheet 1:10000   4533
Hydrography                        Map sheet 1:10000   4533
Geographical names            Map sheet 1:10000    4533

Kind regards,
Lenka Rejentová
 

#7 Updated by Michael Lutz almost 3 years ago

In France dataset series are defined as “data sharing same characteristics, e.g. specification “  (see http://cnig.gouv.fr/wp-content/uploads/2014/01/2012-08-20_guide-catalogues-md-inspire-v1.0.pdf). It is stated that only series metadata should be reported in the central catalog if the dataset serie is served by one service. However there is no specific recommendation to have a dataset serie gathered in one service. I would say that it mainly depends if the serie is produced by one producer or several.
 
The only implementation issue I know is that it creates a discrepancy in number of metadata between producers that use dataset level and producer that use serie level for the reporting and in the catalogs.
 
Marie

#8 Updated by Michael Lutz almost 3 years ago

Please find enclosed the collated answers of the data set series questions for all implementing bodies in Belgium. 
 
Kind regards
Ouns Kissiyar

#9 Updated by Michael Lutz almost 3 years ago

I have made a request for data set series in metadata in the Danish Geoportal (Discovery Service) and the result was none. I do not know to any Danish guidelines and best practices for handling data set series. But I think it is an interesting discussion – there might be different interpretations across EU of what is a data set and what is a data set series (based on ISO TC211 definitions).
 
See you on Monday,
 
Lars

#10 Updated by Michael Lutz almost 3 years ago

Comment from Michal Med (CZ) through the ETS Github repository:

I open this issue to keep track of how it goes. There have been some troubles with definition of dataset and series recently and all we know that it is a problem that has to be solved on the level of MIG. I'm creating this issue to raise its importance.


Case of the Czech Republic -- Series contains all data for the Czech Republic for one schema. Datasets are divided by regions, municipalities or other areas. The files are generated for datasets. It is really important approach towards ATOM, where it really depends on the level in order to be able to filter data based on feed or subfeed information. It is not possible to filter data of subsets, because they are portrayed as entries in ATOM, and only feeds or subfeeds may be returned. This problem needs to find solution on a MIG ground, one way or another, but current situation is ambiguos.

#11 Updated by Christina Wasström almost 3 years ago

  • Are there any guidelines or common practices for the use of data set series in your countries?
    • No
  • For what purpose(s) are they used (e.g. time series, series of map sheets, collections of INSPIRE data sets from different regions in your country, …)
    • We have very few data set series, but we have some examples;
      • Aerial and satellite imagery

      • Orthophoto

      • <!-- -->

        Series of topographic map sheets

  • Have you encountered any implementation issues related to data set series?
    • No

Also available in: Atom PDF