Jump to: navigation, search

Ritmare metadata

Notes

Identifiers of coupled resources

We have three relations of this kind: those linking datasets in a time series, those identifying datasets in a series (as intended by INSPIRE), and finally those linking services to coupled resources and vice versa.

  • Elements "identificatore del file" (fileIdentifier) e "id file precedente" (parentIdentifier) allow to link one another individual dataset in a time series.
  • Elements "Identificatore" (identifier) e "Id livello superiore" (series) link datasets in a series (as intended by INSPIRE) and the series itself.

Perhaps it is possible to express the behaviour enabled by these four identifiers with just two of them. As an example, an identifier of type iPA:cod-Ente:yyyymmdd:hhmmss could be extended to include additional information, such as:

iPA:cod-Ente:id-dataset:yyyymmdd:hhmmss

so that individual datasets still have unique identifiers but time series "id-dataset" can be singled out. Or, even better:

iPA:cod-Ente:datasets:id-dataset:yyyymmdd:hhmmss

iPA:cod-Ente:series:id-series:tile-number

so that both time series and series can be distinguished.

  • Elements "Risorsa accoppiata" and "Risorsa on-line", respectively, relate a service to datasets that can be applied as input and a dataset to services that can process it.

Element "Risorsa on-line" may have further purposes.

URIs for resources in RITMARE

The proposal is to use a unique, automatically generated identifier to express "identificatore del file" and "Identificatore" and encoding in the URN scheme the semantics of "id file precedente" and "Id livello superiore".

A component of the INSPIRE "identifier" data item is the attribute "codeSpace" (not mentioned in the RNDT guidelines) and it could be statically set for RITMARE (e.g., "http://ritmare.it/").

The URN scheme could be something like:

iPA:cod-Ente:[datasets|series|timeseries|services|sensors]:id-resource:yyyymmdd:hhmmss

The URI of the metadata record in RDF format could then be:

codeSpace + resource URN

Ordering a resource according to the yyyymmdd:hhmmss part of the URI allows to distinguish among elements in series and time series (see if it is necessary to add the millisecond part of the timestamp).

INSPIRE RNDT SensorML ODP RDF mapping Notes
Resource type [1] Livello gerarchico [1] dataset / datasetType xmlns:ecodp="http://ec.europa.eu/open-data/ontologies/ec-odp#"

<dataset URI> ecodp:datasetType <datasetTypeURI>

1) INSPIRE mandates "dataset", "series", or "service"; RNDT mandates "dataset" or "serie" but the codelist also contains "servizio" ("service") and "sezione" ("tile"); 2) the corresponding NAL codelist has nothing to do with INSPIRE resource types
Identification
Resource title [1] Titolo [1] dataset / title <dataset URI> dct:title "title"
Resource abstract [1] Descrizione [1] dataset / description <dataset URI> dct:description "description"
Unique resource identifier [1...*] Identificatore [1] dataset / URI <dataset URI> rdf:type dcat:Dataset 1) not mandatory for services, 2) <dataset URI> combines the RITMARE codespace (e.g. http://ritmare.cnr.it/) and the URN associated with the resource (see above for the URN format)
Resource Type [1](*) Livello gerarchico [1](*) dataset / type <dataset URI> ecodp:datasetType <datasetTypeURI> compare INSPIRE codelist with http://ec.europa.eu/open-data/kos/dataset-type/Taxonomy http://ec.europa.eu/open-data/kos/dataset-type/Thesaurus
Resource locator(*) dataset / distribution / URL <distribution URI> dcat:accessURL "URL" a "distribution" entity shall be created for each resource
Coupled resource ? ? find suitable ODP metadata field to express this
Id livello superiore [1](*)
Resource language [0...n] Lingua [1](*) dataset / language <dataset URI> dct:language "<NAL code>" NAL codes are different from those used in INSPIRE metadata, requires mapping between the two SKOS thesauri
Classification of spatial data and services
Topic category [1...n](*) Categoria tematica [1...n](*) dataset / keyword <dataset URI> ecodp:keyword "keyword" 1) use the originating thesaurus to distinguish between Topic category, Spatial data service type, and keywords; 2)use prefLabel (in the language associated with metadata) for a text representation; 3) need to store a URI
Spatial data service type dataset / keyword <dataset URI> ecodp:keyword "keyword" 1) use the originating thesaurus to distinguish between Topic category, Spatial data service type, and keywords; 2)use prefLabel (in the language associated with metadata) for a text representation; 3) need to store a URI
Formato di presentazione [1...n](*) <dataset URI> dct:format <CI_PresentationFormCode> creare thesaurus da 3.4.3.2 - all. 2 DM
Altri dettagli [0...n]
Informazioni supplementari [0...1]
Set di caratteri [0...n]
Tipo di rappresentazione spaziale [1...n](*) aggiungere codelist
Keyword
Keyword value Parole chiave dataset / keyword <dataset URI> ecodp:keyword "keyword" 1) use the originating thesaurus to distinguish between Topic category, Spatial data service type, and keywords; 2) use prefLabel (in the language associated with metadata) for a text representation; 3) need to store a URI
Originating controlled vocabulary 1) implied by keyword value; 2) use prefLabel (in the language associated with metadata) for a text representation, use the thesaurus publication date and type
Geographic location
Geographic bounding box [1] Localizzazione geografica [1] dataset / geographical coverage <dataset URI> dct:spatial <spatial URI> may not be sufficient to express a bounding box
Estensione verticale [0...1]
Temporal reference
Temporal extent Estensione temporale dataset / temporal coverage / coverage from

dataset / temporal coverage / coverage to

<dataset URI> dct:temporal <period>

<period> ecodp:periodStart <datetime>

<period> ecodp:periodEnd <datetime>

Not mandatory but at least one temporal reference is required
Date of publication Data dataset / issued <dataset URI> rndt:dateStruct <_1>

<_1> rndt:dateValue "yyyymmdd"

<_1> rndt:dateType <date type URI>

Not mandatory but at least one temporal reference is required
Date of last revision Data dataset / last modification date <dataset URI> dct:modified <datetime> Not mandatory but at least one temporal reference is required
Date of creation Data ? ? Not mandatory but at least one temporal reference is required
Conformity
Degree ? ?
Specification ? ?
Quality and validity
Lineage [1] for data, [0...n] for services Genealogia [1](*)
Spatial resolution [0...n] Risoluzione spaziale [0...n]
Livello di qualità [1](*)
Accuratezza posizionale [1](*)
Constraints related to access and use
Limitations on public access [1...n](*) Vincoli di accesso [1...n](*) dataset / licence <dataset URI> dct:license <license URI> 1) add ISO codelist to thesauri; 2) mismatch with NAL codes
Conditions applying to access and use [1...n](*) Limitazione d’uso [1](*) ? ? RITMARE data policy?
Vincoli di fruibilità [1...n](*)
Altri vincoli [0...n] come per Vincoli di accesso
Vincoli di sicurezza [1] come per Vincoli di accesso
Responsible organisation
Responsible party [1](*) Punto di contatto [1...n](*) <dataset URI> dct:publisher <publisher URI> <dataset URI> ecodp:contactPoint <contactPointURI> 1) verify http://publications.europa.eu/resource/authority/corporate-body/INFSO; 2) the ODP metadata format requires this value to be a resource: using the FOAF entity describing a RITMARE individual provides the data items required by INSPIRE (organisation name, E-mail); 3) need to add website or tel no. to the info associated with organisations
Responsabile [1..n](*) assume same as Point of contact with role changed from "punto di contatto" to "proprietario"?
Responsible party role ? ?
Metadata on metadata
V Identificatore del file [1](*) 1) URN of format iPA:cod-Ente:[datasets, series, timeseries, services, sensors]:id-resource:yyyymmdd:hhmmss. 2) why can't it be equal to field "identificatore"? 3) Can be derived from the URI, no need for an RDF property specifying this item.
Metadata language [1] V Lingua dei metadati [1] catalog record / language <record URI> dct:language <language URI> added to catalog record to distinguish it from dataset language
V Id file precedente [1] <record URI> rndt:precedingFile <record URI> for the first version of a dataset (or for a plain dataset), it is same as Identificatore del file
Metadata point of contact [1...n] V Responsabile dei metadati [1...n] catalogue record / publisher <record URI> dct:publisher _05

_05 rndt:person <FOAF record>

_05 rndt:personRole <CI_FixedRoleCode>

1) institution, website, and email from the FOAF record; 2) role if fixed "punto di contatto"
Metadata date [1] V Data dei metadati [1] dataset / catalog record / issued <record URI> dct:issued <datetime>
V Nome dello Standard [1] fixed "DM - Regole tecniche RNDT"
V Versione dello Standard [1] fixed "10 novembre 2011"

(*) = mandatory item