Request for registering Prefixes at the semantic.farm

[HendrikBorgelt]

Dear All,

As part of a DCAT-AP extension (specifically DCAT-AP+) I would like to inquire whether it would be possible to register the prefix specific to DCAT-AP with the Semantic Farm.

The Semantic Farm is a website similar to an ontology lookup service (https://www.ebi.ac.uk/ols4/), but it aims to prevent prefix conflicts between semantic artifacts. It manages established prefixes (e.g. DCAT) by registering them there and providing tools designed to prevent arbitrary prefixes for the same namespace.

While checking whether we are using consistent prefixes, we noticed that the prefix used for DCAT-AP (http://data.europa.eu/r5r/ see nfdi-de/dcat-ap-plus#73) has not yet been registered. I would be happy to take on the task of registering these with Semantic Farm, but I wanted to check first whether this is desired.

Since a contact person must also be specified during prefix registration, I would appreciate it if I could enter the contact information of an active DCAT-AP maintainer here.

Template for registering a new prefix

[cthoyt]

Semantic Farm is very proactive and we actually already added a prefix for DCAT-AP based on the NFDI use case. It's here: https://semantic.farm/registry/dcatap. Any feedback on how to improve it (e.g., if a different person should be the primary, secondary contact, or if you want to update the description) then that would be much appreciated!

[bertvannuffelen]

@cthoyt I downloaded the json structure of the link and updated the information.

There are some challenges in the record:

We make a distinction between the r5r vocabulary (http://data.europa.eu/r5r) supporting the DCAT-AP application profile (https://semiceu.github.io/DCAT-AP/releases/3.0.1/).

These are two independent artifacts with a connected life cycle. If there is a need in DCAT-AP to introduce a new term that does not exists in DCAT (or is adviced by DCAT) then we use the r5r vocabulary to create one. That means this vocabulary can support multiple versions of the application profile DCAT-AP. And even the introduced terms could be (if fit for purpose of-course) be used beyond the DCAT context.

That has consequences in the updated jsons I added:

• a vocabulary has a namespace and thus a prefix (*)
• an application profile has no namespace (or a versioned one and thus not genericly reusable) and thus no prefix.

One question I have is how these records are kept up to date. Will you ensure that each time there is a new version a new entry is registered?

The json for the vocabulary

{"prefix":"dcatap",
"name":"DCAT-AP vocabulary, the r5r namespace",
"description":"This vocabulary is supporting DCAT-AP, the DCAT application profile for European data portals The terms defined in this namespace are defined as broad as possible for wider reuse but are scoped to the needs of DCAT-AP. All terms in this vocabulary have therefore a strong connection with DCAT-AP.",
"uri_format":"http://data.europa.eu/r5r/$1",
"providers":[],
"homepage":"https://semiceu.github.io/DCAT-AP/r5r/releases/3.0.0/",
"repository":"https://github.com/SEMICeu/DCAT-AP",
"contact":{"name":"George Lobo","email":"https://interoperable-europe.ec.europa.eu/collection/semic-support-centre/helpdesk-faq","github":""},
"example":"applicableLegislation",
"example_extras":[],
"deprecated":false,
"mappings":{},
"synonyms":["dcat_ap"],
"keywords":["data service","dataset","dcat", "metadata","distribution"],
"publications":[],
"appears_in":[],
"depends_on":[],
"namespace_in_lui":false,
"contributor":{
           "name":"Charles Tapley Hoyt",
           "orcid":"0000-0003-4423-4370",
           "email":"cthoyt@gmail.com","github":"cthoyt"},
"preferred_prefix":"dcatap",
"github_request_issue":1869}

{"prefix":"",
"name":"DCAT AP",
"description":"DCAT-AP is a DCAT profile for sharing information about Catalogues containing Datasets and Data Services descriptions in Europe, under maintenance by the SEMIC action, Interoperable Europe.\n\nThis Application Profile provides a minimal common basis within Europe to share Datasets and Data Services cross-border and cross-domain. [from homepage]",
"uri_format":"",
"providers":[],
"homepage":"https://semiceu.github.io/DCAT-AP/releases/3.0.1/",
"repository":"https://github.com/SEMICeu/DCAT-AP",
"contact":{"name":"George Lobo","email":"https://interoperable-europe.ec.europa.eu/collection/semic-support-centre/helpdesk-faq","github":""},
"example":"",
"example_extras":[],
"deprecated":false,
"mappings":{},
"synonyms":["dcat_ap"],
"keywords":["data service","dataset","dcat", "metadata","distribution"],
"publications":[],
"appears_in":[],
"depends_on":[],
"namespace_in_lui":false,
"contributor":{
           "name":"Charles Tapley Hoyt",
           "orcid":"0000-0003-4423-4370",
           "email":"cthoyt@gmail.com","github":"cthoyt"},
"preferred_prefix":"dcatap",
"github_request_issue":1869}

(*) as a personal note, for me the prefix is an local abbreviation to support human readibility. I know this is in the community often treated as an alternative identifier, but I prefer to stick to the absolute URIs as they are entirely unambiguosly and are not manipulated by some program. This makes the specification more implementation technical neutral.