Metadata Model (v2)

This is the metadata model that the DPE serves. It describes DaSCH research projects and their context as a hierarchy of entities. The model originated in the now-retired dsp-meta repository, where it was designed as the successor to the earlier v1 model. The metadata has since been migrated to this repository and converted to v2; the model is implemented as Rust types in the dpe-core crate (modules/dpe/core/src/) and the data lives under modules/dpe/server/data/.

The enhancements over v1 are designed to better accommodate the inherent complexity of humanities projects, while still supporting simpler project structures. The two main additions are:

• a hierarchical level above the research project — the project cluster — which represents overarching initiatives that span multiple projects over long periods;
• collections, which allow more precise referencing and grouping of parts of the data, including cross-project and nested groupings. Collections replace the v1 dataset concept.

Licensing

All metadata is considered public domain. By signing the deposit agreement, projects consent to that. This is unlike the domain metadata which is part of the project's data and can be licensed as the project wishes.

Whenever metadata is served to a client, it is served with legal information. Legal information on metadata, as everywhere else, consists of the license, the copyright holder and the authorship. For metadata the license is always "public domain", the copyright holder is always "DaSCH" and the authorship is always the project and DaSCH.

Metadata is always publicly available, even if the corresponding project, collection or record is not. This ensures the metadata stays findable and reusable even if the data itself is not. The only exception is the status "embargoed", during which the metadata is only available on the project level.

Model Overview

The metadata model is a hierarchical structure of metadata elements.

flowchart TD
    projectCluster[Project Cluster]
    project[Research Project]
    record[Record]
    collection[Collection]

    projectCluster -->|0-n| project
    projectCluster -->|0-n| collection

    project -->|0-n| collection
    project -->|1-n| record
    collection --> |0-n| record
    collection --> |0-n| collection

• A Project Cluster collects research projects (or nested project clusters). It is typically institutional in nature, not directly tied to a specific funding grant, and may be long-lived. Examples are EKWS/CAS, BEOL or LHTT.
• A Research Project is the main entity of the model. It corresponds to a project in the DSP. It is typically tied to a specific funding grant and hence has a limited lifetime of ~3–5 years; multiple funding rounds and a longer lifetime are possible. A research project is part of 0–n project clusters and contains both collections and records. All records in the project are listed in the project's records array, regardless of collection membership.
• A Collection is a flexible grouping of records that can span multiple projects or be nested within other collections. Collections enable cross-project organization and support subsetting and specialized access patterns. They may contain both individual records and nested collections.
• A Record is a single entry within a project — the smallest unit that can meaningfully have an identifier. It maps to a knora-base:Resource (DSP-API) or an Asset (SIPI/Ingest) in the DSP. For DSP resources, the metadata of the record is the existence of the resource itself plus information such as label, access rights and provenance; the core data are the values on that resource. For assets, the metadata is the existence of the asset and its access rights; the core data is the binary content. A record is part of exactly 1 research project and may be part of 0–n collections.

Additionally, Person and Organization are entities independent of the project hierarchy, related to various entities within it (e.g. as contacts, contributors or funders).

Entity Types

Project Cluster

• id: A unique internal identifier; not exposed to the user and not persistent.
• pid: A unique persistent identifier (currently an ARK URL).
• name: The name of the project cluster.
• projects: Identifiers of the projects in the cluster.
• projectClusters: Identifiers of nested project clusters.
• description: The description of the cluster.
• url: The URL to the web presence of the cluster.
• howToCite: How to cite the cluster. If not provided, the standard form <name> (<year>). [Project Cluster]. DaSCH. <ARK> is used.
• alternativeNames: Alternative names of the cluster.
• contactPoint: Persons or organizations responsible for the cluster.
• documentationMaterial: URLs pointing to documentation material.

Most fields are optional, to keep the entity flexible. There is no difference in cardinality between the archival and in-progress stages.

Project

• id: A unique internal identifier; not exposed to the user and not persistent.
• pid: A unique persistent identifier (currently an ARK URL).
• shortcode: The project's DSP shortcode, internal only. Four hexadecimal characters, upper case.
• officialName: The official name of the project.
• status: The status of the project — either "Ongoing" or "Finished".
• name: The name of the project.
• shortDescription: A short teaser. Maximum length: 200 characters.
• description: The full description of the project.
• startDate: The start date of the project.
• endDate: The end date of the project.
• dataPublicationYear: The year the data is published — normally the year the project finishes and the data moves to the archive. Under embargo, the year the embargo is lifted. Projects published while in the VRE may set a specific year.
• url: The URL(s) to the web presence of the project. The first should point to where the data is available; the second, optional, may point to the project website.
• howToCite: How to cite the project. If not provided, the standard form <contributors> (<year>). <project name> [Database]. DaSCH. <ARK> is used.
• accessRights: The access rights of the project (see Access Rights). Defines to what extent the project data is accessible in the DPE. If the project is embargoed, the metadata is only available on the project level.
• legalInfo: Legal information about the project. Calculated from records; cannot be specified explicitly on the project.
• dataManagementPlan: A data management plan (string or URL); use "not accessible" if not available.
• typeOfData: The type(s) of data — "XML", "Text", "Image", "Video", "Audio". Computed from the records where available and optionally added manually.
• dataLanguage: Languages contained in the project. Computed from the records where available and optionally added manually.
• collections: Collection identifiers that optionally group project data.
• records: Identifiers of all records that make up the project data. This is the canonical list of all records in the project.
• keywords: Keywords describing the project.
• disciplines: Disciplines the project relates to.
• temporalCoverage: Epochs or time periods the project relates to.
• spatialCoverage: References to spatial entities (places, regions, …).
• attributions: Roles people/organizations have in the project. Entered manually, since there may be people without authorship (reviewers, organizers, …).
• abstract: An abstract of the project.
• contactPoint: Persons or organizations responsible for the project.
• publications: Publications related to the project.
• funding: Either a string ("No funding") or a list of grants.
• alternativeNames: Alternative names of the project.
• documentationMaterial: URLs pointing to documentation material.
• provenance: The history of the project, if applicable.
• additionalMaterial: Additional URLs related to the project.

Collection

• id: A unique internal identifier; not exposed to the user and not persistent.
• pid: A unique persistent identifier (currently an ARK URL).
• name: The name of the collection.
• accessRights: The access rights of the collection (see Access Rights).
• legalInfo: Legal information about the collection. Calculated from records/sub-collections; may be added manually.
• howToCite: How to cite the collection. If not provided, the standard form <contributors> (<year>). <collection name> [Collection]. DaSCH. <ARK> is used.
• description: The description of the collection.
• typeOfData: The type(s) of data — "XML", "Text", "Image", "Video", "Audio".
• dateCreated: When the collection was created.
• dateModified: When the collection was last modified.
• records: Identifiers of the records in the collection.
• collections: Identifiers of nested collections.
• languages: Languages contained in the collection.
• additionalMaterial: Additional URLs related to the collection.
• provenance: The history of the collection, if applicable.
• keywords: Keywords for search purposes.
• documentationMaterial: URLs pointing to documentation material.

Record

• id: A unique identifier for the record.
• pid: A unique persistent identifier (an ARK URL).
• label: The label of the record. For assets, this may be the original file name. For IIIF URLs, it is useful for the case when the URL is no longer available. In the long run IIIF Manifests, rather than image URLs, would let labels be extracted from there.
• accessRights: The access rights of the record. Defines to what extent the record data is accessible in the DPE.
• legalInfo: Legal information about the record.
• howToCite: How to cite the record. If not provided, the standard form <label> (<creation year>). [Data Record]. DaSCH. <ARK> is used.
• publisher: The publisher of the record. Literal "DaSCH"; required for OpenAIRE compliance.
• source: The provenance of the record. Recommended for OpenAIRE: use only if the record is a digitization of a non-digital source, in which case it should identify the original source.
• description: The description of the record. If the project does not want descriptions to be public domain and always open, it must not use this property but instead create a custom property.
• dateCreated, dateModified: Creation and modification dates.
• datePublished: When the record was made publicly available — normally when it moved to the archive, or when an embargo is lifted.
• typeOfData: The type of data — "XML", "Text", "Image", "Video", "Audio".
• size: The size of the record (OpenAIRE Size).
• keywords: Keywords for search purposes.

Person

Cardinality is the same for both stages.

• id: A unique internal identifier; not exposed to the user and not persistent.
• pid: A unique persistent identifier (currently an ARK URL).
• sameAs: References to external authority files (ORCID, VIAF, GND, …).
• givenNames: The given names of the person.
• familyNames: The family names of the person.
• honoraryPrefix: Honorary prefixes, e.g. "Prof. Dr.".
• honorarySuffix: Honorary suffixes, e.g. "PhD", "MA".
• affiliations: Organizations the person is affiliated with.
• email: The email address of the person.
• address: The postal address of the person — the address at their organization, not a personal address.

Organization

Cardinality is the same for both stages.

• id: A unique internal identifier; not exposed to the user and not persistent.
• pid: A unique persistent identifier (currently an ARK URL).
• sameAs: References to external authority files (e.g. ROR).
• name: The name of the organization.
• url: The URL of the organization.
• address: The address of the organization.
• email: The email address of the organization.
• alternativeName: Alternative names of the organization.

Value Types

String with Language Tag (lang_string)

An object with ISO language codes as keys and strings as values:

{
  "en": "Lorem ipsum in English.",
  "de": "Lorem ipsum auf Deutsch."
}

A single lang_string value can hold multiple translations.

Authority File Reference

An object representing a reference to an external authority file.

• type: The type of the reference — e.g. 'Geonames', 'Pleiades', 'Skos', 'Periodo', 'Chronontology', 'GND', 'VIAF', 'Grid', 'ORCID', 'Creative Commons', 'COAR'. Used to determine the semantics of the URL. The implementation also uses 'URL' for plain links.
• url: The URL itself.
• text: A human-readable text for display.

PID

A persistent identifier — may be an ARK or a DOI. Used e.g. on publications.

Publication

• text: The text of the publication.
• pid: A URL to the publication, e.g. a DOI, if available.

Address

Grant

Legal Info

License

Attribution

Modelled according to the OpenAIRE guidelines.

Access Rights

• accessRights: One of "Full Open Access", "Open Access with Restrictions", "Embargoed Access", "Metadata only Access".
• embargoDate: The date when the embargo ends.

Internal ID

An internal ID (internal_id) is a unique identifier for an entity within the system. It is not intentionally exposed to the user and is presented as a string.

OpenAIRE Mapping

The model includes a mapping to the OpenAIRE Guidelines for Data Archives, which are based on the DataCite Metadata Schema. Currently only projects are exposed as OpenAIRE datasets. For the endpoint that serves this, see the OAI-PMH Endpoint page.

The OpenAIRE Guidelines specify 18 fields, with cardinalities Mandatory (M), Recommended (R), Mandatory if Applicable (MA) and Optional (O):

1. Identifier (M)
2. Creator (M)
3. Title (M)
4. Publisher (M)
5. PublicationYear (M)
6. Subject (R)
7. Contributor (MA/O)
8. Date (M)
9. Language (R)
10. ResourceType (R)
11. AlternateIdentifier (O)
12. RelatedIdentifier (MA)
13. Size (O)
14. Format (O)
15. Version (O)
16. Rights (MA)
17. Description (MA)
18. GeoLocation (O)

Project → OpenAIRE Dataset Mapping

Open questions in the design include which attribution roles map to Creator vs. Contributor, how PublicationYear should be derived per project, and whether collections should also be exposed as OpenAIRE datasets.

Examples

The following examples show the conceptual JSON shape. Real data files live under modules/dpe/server/data/ and may differ in the ways noted in the As implemented callouts above (e.g. the legacy url array form).

Project Cluster

{
  "id": "cluster-0001",
  "pid": "https://ark.dasch.swiss/ark:/72163/1/cluster-0001",
  "name": "Project Cluster Name",
  "projects": ["project-0001", "project-0002"],
  "projectClusters": ["cluster-0002"],
  "description": {
    "en": "Project Cluster Description",
    "de": "Projektcluster Beschreibung"
  },
  "url": "https://example.com/project-cluster",
  "howToCite": "Project Cluster Name (2025). [Project Cluster]. DaSCH. https://ark.dasch.swiss/ark:/72163/1/cluster-0001",
  "alternativeNames": [{ "en": "Alternative Name", "de": "Alternativer Name" }],
  "contactPoint": ["person-0001", "organization-0001"],
  "documentationMaterial": ["https://example.com/documentation"]
}

Project

{
  "id": "project-0001",
  "pid": "https://ark.dasch.swiss/ark:/72163/1/project-0001",
  "shortcode": "1234",
  "officialName": "Project Official Name",
  "status": "Ongoing",
  "name": "Project Name",
  "shortDescription": "Short description of the project.",
  "description": { "en": "Project Description", "de": "Projektbeschreibung" },
  "startDate": "2023-01-01",
  "endDate": "2028-01-01",
  "url": { "type": "URL", "url": "https://data.dasch.swiss/projects/project-0001" },
  "secondaryUrl": { "type": "URL", "url": "https://example.com/project-website" },
  "howToCite": "Project Name (2025). [Project]. DaSCH. https://ark.dasch.swiss/ark:/72163/1/project-0001",
  "accessRights": { "accessRights": "Full Open Access" },
  "legalInfo": [
    {
      "license": {
        "licenseIdentifier": "CC BY 4.0",
        "licenseDate": "2023-01-01",
        "licenseURI": "https://creativecommons.org/licenses/by/4.0/"
      },
      "copyrightHolder": "DaSCH",
      "authorship": ["Project XYZ"]
    }
  ],
  "dataManagementPlan": "https://example.com/dmp",
  "collections": ["collection-0001"],
  "records": ["record-0001", "record-0002"],
  "keywords": [{ "en": "Keyword 1", "de": "Stichwort 1" }],
  "disciplines": [{ "en": "Discipline 1", "de": "Disziplin 1" }],
  "temporalCoverage": [{ "en": "2006-2016", "de": "2006-2016" }],
  "spatialCoverage": [
    { "type": "Geonames", "url": "https://www.geonames.org/2658434/", "text": "Switzerland" }
  ],
  "attributions": [
    { "contributor": "person-0001", "contributorType": ["Project leader", "Data curator"] }
  ],
  "abstract": { "en": "Project Abstract", "de": "Projektzusammenfassung" },
  "contactPoint": ["person-0001", "organization-0001"],
  "publications": [{ "text": "Publication Title", "pid": { "url": "https://doi.org/10.1234/5678" } }],
  "funding": [
    { "funders": ["organization-0001"], "number": "123456", "name": "Grant Name", "url": "https://example.com/grant" }
  ],
  "alternativeNames": [{ "en": "Alternative Name", "de": "Alternativer Name" }]
}

Person

{
  "id": "person-0001",
  "givenNames": ["Jane"],
  "familyNames": ["Doe"],
  "jobTitles": ["Senior lecturer"],
  "affiliations": ["organization-0001"],
  "sameAs": [{ "type": "ORCID", "url": "https://orcid.org/0000-0000-0000-0000" }],
  "email": "jane.doe@example.org"
}

Organization

{
  "id": "organization-0001",
  "name": "Université de Lausanne",
  "url": "https://www.unil.ch/",
  "address": {
    "street": "Unicentre",
    "postalCode": "1015",
    "locality": "Lausanne",
    "country": "Switzerland"
  }
}