Datasheet v2.1
A datasheet describes a cultural heritage dataset for the people who will reuse it: what it contains, how it was made, under which conditions it can be used, and what to watch out for. This profile follows the Datasheets for Digital Cultural Heritage Datasets template and exports to DCAT-AP.
Based on
- Datasheets for Digital Cultural Heritage Datasets (Zenodo): The datasheet template this profile implements, version 2 (2025-07-07).
| Schema | https://sidedoc.app/schemas/0.2.4/datasheet.json |
| Profile id | ds:v00003 |
| Version | 0.2.4 |
| Help file | help/docs.datasheet.json (shared with datasheet-v2) |
| Fields | 72 |
| Help | 71 of 72 fields have a short text, 72 an extended text, 57 a tooltip and 42 details (English). |
Types link to the type library. For what each fact means, see x-sidedoc keys and sheet columns.
Fields
Field reference
Title
| Id | ds:f00100 |
| Path | title |
| Type | Literal |
| Values | 1..1 (required, one value) |
| Multilingual | yes |
| Title field | yes: shown as the document title |
| DCAT-AP | dcterms:title |
Tooltip (level 1)
Provide a title for the dataset
Short help (level 2)
The title should be concise and should provide users with an idea about the scope and purpose of the dataset
Extended help (level 3)
The title should be concise and should provide users with an idea about scope and purpose of the dataset
Details (level 4)
Examples:
Delpher Historical Dutch Newspapers Corpus (1618–1995)Names the source, the content type and the period.KB Medieval Manuscripts: Page-level Images and TranscriptionsThe part after the colon says what the dataset contains.
Avoid:
Dataset 1Says nothing about content or scope.kranten_export_v3_final.csvA file name is not a title.
Notes:
Aim for 5 to 15 words. Put the most distinctive words first: the title is what people see in search results and catalogues.
Add the title in each language you describe the dataset in. Use the same title as on the dataset's landing page, so people recognise it.
Datasheet identifier
| Id | ds:f00120 |
| Path | datasheetIdentifier |
| Type | TypedValue |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| Identifier schemes | DOI, Handle, ARK, URN:NBN, PURL, w3id |
| DCAT-AP | adms:identifier (on the catalog record (dcat:CatalogRecord), not the dataset; values that do not fit go to dcterms:identifier) |
Tooltip (level 1)
Not written yet.
Short help (level 2)
A persistent identifier (PID) for this datasheet itself — the metadata record, as distinct from the dataset it describes.
Extended help (level 3)
Provide one or more persistent identifiers for the datasheet document. Choose a PID scheme (e.g. DOI, Handle, ARK) or enter a custom scheme, then the identifier value. This identifies the record, not the dataset — use the Dataset identifier for the dataset being described. In DCAT exports it becomes the identifier of a catalog record whose primary topic is the dataset.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Examples:
10.5281/zenodo.8123456A DOI, without the https://doi.org/ prefix.https://hdl.handle.net/10622/EXAMPLEA Handle.
Notes:
This identifies the datasheet (the description), not the dataset. If you do not have a separate PID for the datasheet, leave this empty and fill in Dataset identifier instead.
See also:
Description
| Id | ds:f00200 |
| Path | description |
| Type | RichText, with sub-fields |
| Values | 1..1 (required, one value) |
| Multilingual | yes |
| DCAT-AP | dcterms:description (Markdown text also goes to ddchd:descriptionMarkdown) |
| Sub-fields | Dataset identifier, Homepage, Publisher, Dataset curator, Other contributor, Point of contact, Paper or other reference, Keyword(s), Theme, Supported tasks or shared tasks, Citation information |
Tooltip (level 1)
Create description
Short help (level 2)
Provide core information about the dataset
Extended help (level 3)
Provide a brief descriptive overview of the dataset and information on the purpose for which it was created. Shortly describe its structure, additional files, the motivation for its creation, who created the dataset (and on which behalf) and who funded it. Bear in mind that this field will reliably be read.
Details (level 4)
Notes:
This group holds the core facts: homepage, publisher, curators, contact, keywords and theme. Fill in at least the publisher and a point of contact.
Dataset identifier
| Id | ds:f00250 |
| Path | description.datasetIdentifier |
| Type | TypedValue |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| Identifier schemes | DOI, Handle, ARK, URN:NBN, PURL, w3id |
| DCAT-AP | adms:identifier (values that do not fit go to dcterms:identifier) |
| Part of | Description |
Tooltip (level 1)
Not written yet.
Short help (level 2)
A persistent identifier (PID) for the dataset being described.
Extended help (level 3)
Provide one or more persistent identifiers for the described dataset. Choose a PID scheme (e.g. DOI, Handle, ARK) or a custom scheme and enter the value. Distinct from the datasheet's own identifier; exports as the dataset's identifier.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Examples:
10.17026/dans-exampleA DOI from a data repository.urn:nbn:nl:ui:13-exampleA URN:NBN, as used by Dutch repositories.
Avoid:
https://www.example.org/datasets/newspapersA web address can change. Put it under Homepage instead.
Notes:
Use a persistent identifier (PID) that keeps resolving when the website changes. If the dataset has none yet, ask your repository or data steward to mint one.
Homepage
| Id | ds:f00300 |
| Path | description.homePage |
| Type | Url |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | url |
| Validation | href: Starts with http:// or https:// ^https?:\/\/.+$As written in the sheet: {"href":"/^https?:\\/\\/.+$/"} |
| DCAT-AP | dcat:landingPage |
| Part of | Description |
Tooltip (level 1)
Add the URL of the dataset
Short help (level 2)
Add the URL of the homepage of the dataset here if available
Extended help (level 3)
Add the URL of the homepage of the dataset here if available
Details (level 4)
Examples:
https://www.delpher.nl/nl/krantenThe page where people learn about and find the dataset.
Notes:
Link to a human-readable landing page. A direct download link goes under Distribution, Download URL.
Publisher
| Id | ds:f00400 |
| Path | description.publisher |
| Type | Agent |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| Identifier schemes | identifier: ORCID, VIAF, ISNI, ROR, Wikidata, GND |
| DCAT-AP | dcterms:publisher |
| DCAT-AP | prov:qualifiedAttribution |
| Part of | Description |
Tooltip (level 1)
Name the publishing institution or individual
Short help (level 2)
Name the publishing institution or individual
Extended help (level 3)
Either the individual publisher or the responsible organisation should be provided - preferably the latter. Note that this is likely to be different from "curators" and "other contributors" to be filled in other sections. If you list more than one publisher, put the main one first: DCAT-AP allows a single publisher, so the export uses the first one as the publisher and records the others as additional publishers.
Details (level 4)
Examples:
KB, National Library of the NetherlandsAn organisation. Add its ROR identifier in the agent form.
Notes:
The publisher is the organisation (or person) that makes the dataset available and is responsible for it. This is often not the person who did the work: credit them as curator or contributor.
Dataset curator
| Id | ds:f00500 |
| Path | description.datasetCurator |
| Type | Agent |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| Free text | yes: free text can be given instead of the structured value |
| Identifier schemes | identifier: ORCID, VIAF, ISNI, ROR, Wikidata, GND |
| DCAT-AP | prov:qualifiedAttribution (values that do not fit go to ddchd:datasetCurator) |
| Part of | Description |
Tooltip (level 1)
Name the data curator(s)
Short help (level 2)
Name the dataset curator(s) and provide email addresses and affiliations
Extended help (level 3)
List the key people involved in collecting the dataset and their affiliation(s). Dataset curators are encouraged to insert some information about their positionality here, e.g. in the form of a short bio containing information on their formation, role and function within the CHI or project and the task(s) with which they were entrusted during the creation of the dataset.
Details (level 4)
Examples:
Jane de Vries, KB Research Department, jane.devries@example.orgName, affiliation and email, each in its own field of the agent form.
Notes:
Curators select, clean and maintain the data. Add an ORCID for people where you can, so the credit is unambiguous.
See also:
Other contributor
| Id | ds:f00600 |
| Path | description.otherContributor |
| Type | Agent |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| Free text | yes: free text can be given instead of the structured value |
| Identifier schemes | identifier: ORCID, VIAF, ISNI, ROR, Wikidata, GND |
| DCAT-AP | prov:qualifiedAttribution (values that do not fit go to ddchd:otherContributor) |
| Part of | Description |
Tooltip (level 1)
Name additional people who have contributed to the dataset
Short help (level 2)
Credit the additional people who have contributed to this dataset other than those who have curated and published it and annotators (as per the section below)
Extended help (level 3)
Other contributors may be annotators, cultural heritage practitioners involved in the creation of the dataset, student assistants, people employed in other institutions, curators of the physical collection.
Details (level 4)
Notes:
Use this for people and organisations that helped but are not curators or publisher: annotators, funders, digitisation partners, volunteers. Choose a role for each, so readers know what they did.
Point of contact
| Id | ds:f00700 |
| Path | description.pointOfContact |
| Type | Agent or Url |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| Identifier schemes | identifier: ORCID, VIAF, ISNI, ROR, Wikidata, GND |
| DCAT-AP | dcat:contactPoint (values that do not fit go to ddchd:pointOfContact) |
| Part of | Description |
Tooltip (level 1)
Name a contact person or provide a functional email address
Short help (level 2)
Provide at least one reliable person or organisation that can answer questions about the dataset or this datasheet. This may be a named contact, an organisational contact, an email address, an ORCID, or another resolvable URL
Extended help (level 3)
The point of contact identifies who readers can reach for questions, corrections, comments, or follow-up information about the dataset and/or its datasheet. Because datasets and datasheets may change over time, this contact should be reliable and suitable for future communication and capable to incorporate responses and comments into a new version of the datasheet. Prefer a simple contact record when possible, including a name, role, affiliation, email address, identifier, or homepage. When only a simple contact reference is available, a plain URL may be provided instead.
Details (level 4)
Examples:
Datasets team, KB, datasets@example.orgA shared mailbox stays valid when people change jobs.
Notes:
Prefer a function or team address over a personal one. Check that someone actually reads it.
Paper or other reference
| Id | ds:f00800 |
| Path | description.reference |
| Type | Reference |
| Values | 0..* (optional, repeatable) |
| Multilingual | yes |
| Free text | yes: free text can be given instead of the structured value |
| DCAT-AP | foaf:page (values that do not fit go to ddchd:reference) |
| Part of | Description |
Tooltip (level 1)
Insert papers or references relevant for this dataset
Short help (level 2)
If the dataset was introduced by a paper or there was a paper written describing it, add URL here
Extended help (level 3)
Insert the citation of a paper which uses this dataset; if the dataset re-uses (parts of) other datasets, insert references to them here (paper citations or persistent identifiers). Include DOIs or other persistent identifiers in citations.
Details (level 4)
Examples:
https://doi.org/10.1234/example-paperThe DOI link of the paper that introduced the dataset.
Notes:
Use the paper's DOI link where it has one. You can also paste a BibTeX entry: SideDoc fills in the reference fields from it.
Keyword(s)
| Id | ds:f00850 |
| Path | description.keywords |
| Type | Literal |
| Values | 0..* (optional, repeatable) |
| Multilingual | yes |
| DCAT-AP | dcat:keyword |
| Part of | Description |
Tooltip (level 1)
Not written yet.
Short help (level 2)
Free-text keywords describing the dataset.
Extended help (level 3)
Add terms that help users find the dataset. Keywords are free text; for controlled-vocabulary categories use Theme.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Examples:
historical newspapers(en only)OCR(en only)Dutch press history(en only)historische kranten(nl only)
Avoid:
newspapers, OCR, DutchOne keyword per value, not a comma-separated list.
Notes:
Keywords help people find the dataset. Think of the words someone would type in a search box. Use Theme for terms from a controlled vocabulary.
Theme
| Id | ds:f00875 |
| Path | description.theme |
| Type | Term |
| Values | 0..* (optional, repeatable) |
| Multilingual | yes |
| DCAT-AP | dcat:theme |
| Part of | Description |
Tooltip (level 1)
Not written yet.
Short help (level 2)
Controlled-vocabulary theme(s) categorising the dataset.
Extended help (level 3)
Select theme(s) from a controlled vocabulary. Themes with a URI export as dcat:theme; free-text values fall back to keywords.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Notes:
Pick one or more themes from the list. Themes put the dataset in a broad category that catalogues use for browsing, so a few well-chosen ones are better than many.
Supported tasks or shared tasks
| Id | ds:f00900 |
| Path | description.supportedTasks |
| Type | RichText, with sub-fields |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:supportedTasks (not mapped to a DCAT-AP property; written to an extension property) |
| Part of | Description |
| Sub-fields | AI category, Type of cultural heritage application, Cultural heritage application example |
Tooltip (level 1)
Add tasks supported by this dataset or shared tasks in which the dataset was created
Short help (level 2)
Give a brief description of the tasks identified for this dataset, metrics, and suggested models. For each task, also provide reference to any shared task the dataset is part of
Extended help (level 3)
Give a brief description of the tasks identified for this dataset, metrics, and suggested models. For each task, also provide reference to any shared task the dataset is part of.
Details (level 4)
Examples:
OCR post-correction: train models to correct OCR errors in historical Dutch print.A task, with what the data is used for.
Notes:
Add one entry per task. For each, say which AI category it falls under and give a concrete heritage application.
AI category
| Id | ds:f01000 |
| Path | description.supportedTasks.aiCategory |
| Type | Term or Literal |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | term_lookup |
| Vocabulary | AI4Culture AI Categories (https://sidedoc.app/vocabs/ai4culture-category)Enforcement: stores a Term (URI and label), only these schemes, own values allowed, custom labels allowed Binding: embedded in the schema |
| DCAT-AP | dcat:theme (values that do not fit go to dcat:keyword) |
| Part of | Supported tasks or shared tasks |
Tooltip (level 1)
Select an AI category
Short help (level 2)
Preferably use shared vocabularies like the AI4Culture one or the HuggingFace taxonomy of tasks; free text is also possible
Extended help (level 3)
Preferably use shared vocabularies like the AI4Culture one (https://ai4culture.eu/) or the HuggingFace taxonomy of tasks (https://huggingface.co/tasks). Alternatively use free text.
Details (level 4)
Not written yet.
Type of cultural heritage application
| Id | ds:f01100 |
| Path | description.supportedTasks.typeOfApplication |
| Type | Term or Literal |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | term_lookup |
| Vocabulary | AI4Culture Application Types (https://sidedoc.app/vocabs/ai4culture-application-type)TaDiRAH - Taxonomy of Digital Research Activities in the Humanities ( https://vocabs.dariah.eu/tadirah/)Enforcement: stores a Term (URI and label), only these schemes, own values allowed, custom labels allowed Binding: embedded in the schema |
| DCAT-AP | dcat:theme (values that do not fit go to dcat:keyword) |
| Part of | Supported tasks or shared tasks |
Tooltip (level 1)
Select a cultural heritage application
Short help (level 2)
Preferably use a shared vocabulary like TADIRAH or the AI4Culture categories; free text is also possible
Extended help (level 3)
Preferably use a shared vocabulary like TADIRAH (https://vocabs.dariah.eu/tadirah/en/) or the AI4Culture categories (https://ai4culture.eu/). Alternatively use free text.
Details (level 4)
Not written yet.
Cultural heritage application example
| Id | ds:f01200 |
| Path | description.supportedTasks.applicationExample |
| Type | RichText |
| Values | 0..* (optional, repeatable) |
| Multilingual | yes |
| DCAT-AP | ddchd:applicationExample (Markdown text also goes to ddchd:applicationExampleMarkdown) |
| Part of | Supported tasks or shared tasks |
Tooltip (level 1)
Provide an example of a specific cultural heritage application
Short help (level 2)
Specific instance of application
Extended help (level 3)
Provide an example of a specific cultural heritage application, preferably in the form of keywords.
Details (level 4)
Not written yet.
Citation information
| Id | ds:f02000 |
| Path | description.citation |
| Type | Reference |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| Free text | yes: free text can be given instead of the structured value |
| DCAT-AP | dcterms:bibliographicCitation (values that do not fit go to ddchd:citation) |
| Part of | Description |
Tooltip (level 1)
A suggested citation for the dataset
Short help (level 2)
Provide a well formatted reference for the dataset, preferably using BibTeX
Extended help (level 3)
Provide a well formatted reference for the dataset, preferably using BibTeX. If the dataset has a DOI or other kind of persistent identifier, it can be also included in this field. Different types can be employed such as article, book or misc (generic bibliograph entry).
Details (level 4)
Examples:
@misc{kb_delpher_2023, title = {Delpher Historical Dutch Newspapers Corpus}, author = {{KB, National Library of the Netherlands}}, year = {2023}, doi = {10.5281/zenodo.8123456} }Paste a BibTeX entry; SideDoc checks that it starts with @type{.
Distribution
| Id | ds:f01300 |
| Path | distribution |
| Type | Container (group of sub-fields) |
| Values | 1..* (required, repeatable) |
| Multilingual | no |
| Row caption | {distributionTitle} |
| DCAT-AP | dcat:distribution |
| Sub-fields | Distribution title, Distribution description, Data access URL, Download URL, License, License information, Access rights, File format, Package format, Byte size, Checksum |
Tooltip (level 1)
A specific representation of a dataset
Short help (level 2)
Provide answers to the following points: distribution information dissemination strategy, licences, DOIs, third party involvement and restrictions
Extended help (level 3)
This field represents a general availability of a dataset. A dataset might be available in multiple serializations that may differ in various ways including format and URL.
Details (level 4)
Notes:
A distribution is one way of getting the data: a download, an API, a set of files in a given format. Add a distribution for each, with its own licence and format.
Distribution title
| Id | ds:f01400 |
| Path | distribution.distributionTitle |
| Type | Literal |
| Values | 1..1 (required, one value) |
| Multilingual | yes |
| DCAT-AP | dcterms:title (on each dcat:distribution) |
| Part of | Distribution |
Tooltip (level 1)
A title describing the distribution
Short help (level 2)
A name given to the distribution
Extended help (level 3)
A name given to the distribution.
Details (level 4)
Examples:
Full text (ALTO XML), per yearOAI-PMH metadata endpoint
Distribution description
| Id | ds:f01500 |
| Path | distribution.distributionDescription |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | dcterms:description (on each dcat:distribution; values that do not fit go to ddchd:distributionDescription) |
| Part of | Distribution |
Tooltip (level 1)
A specific representation of a dataset
Short help (level 2)
Dataset creators should provide answers to the following points: distribution information dissemination strategy, licences, DOIs, third party involvement and restrictions
Extended help (level 3)
This field represents a general availability of a dataset. A dataset might be available in multiple serializations that may differ in various ways including format and URL.
Details (level 4)
Not written yet.
Data access URL
| Id | ds:f01600 |
| Path | distribution.dataAccessUrl |
| Type | Url |
| Values | 1..* (required, repeatable) |
| Multilingual | no |
| UI hint | url |
| Validation | href: Starts with http:// or https:// ^https?:\/\/.+$As written in the sheet: {"href":"/^https?:\\/\\/.+$/"} |
| DCAT-AP | dcat:accessURL (on each dcat:distribution) |
| Part of | Distribution |
Tooltip (level 1)
The URL where the dataset can be accessed
Short help (level 2)
Provide the URL where the dataset can be accessed
Extended help (level 3)
Indicate the URL where the dataset can be accessed, possibly on an online repository, e.g., if the dataset is hosted on GitHub or Zenodo.
Details (level 4)
Examples:
https://www.example.org/datasets/newspapers/accessA page that explains how to get access.
Notes:
Access URL is a page or service that gives access (possibly after login or a request). If there is a direct file link, add it under Download URL as well.
Download URL
| Id | ds:f01650 |
| Path | distribution.downloadUrl |
| Type | Url |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | url |
| Validation | href: Starts with http:// or https:// ^https?:\/\/.+$As written in the sheet: {"href":"/^https?:\\/\\/.+$/"} |
| DCAT-AP | dcat:downloadURL (on each dcat:distribution) |
| Part of | Distribution |
Tooltip (level 1)
Not written yet.
Short help (level 2)
A direct download URL for this distribution's data.
Extended help (level 3)
Link directly to the downloadable file(s) for this distribution (exports as dcat:downloadURL). For a landing/access page use the Data access URL instead.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Examples:
https://data.example.org/newspapers/1918.zip
Notes:
Only for direct links to the data file. Leave it empty when access needs a login or a request.
License
| Id | ds:f01700 |
| Path | distribution.license |
| Type | Term |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | term_lookup |
| Vocabulary | SPDX License List (https://spdx.org/licenses/)Creative Commons Licenses ( https://creativecommons.org/)rightsstatements.org ( http://rightsstatements.org/vocab/)Enforcement: stores a Term (URI and label), only these schemes, own values allowed, custom labels allowed Binding: embedded in the schema |
| DCAT-AP | dcterms:license (on each dcat:distribution; values that do not fit go to ddchd:license) |
| Part of | Distribution |
Tooltip (level 1)
The licence for the current data distribution
Short help (level 2)
Provide the licence(s) for the current data distribution and link to the licence(s) webpage(s) if available
Extended help (level 3)
Indicate the licence(s) for the current data distribution and link to the licence(s) webpage(s) if available. The list is filled from SPDX License List, Creative Commons and RightsStatements.org.
Details (level 4)
Examples:
CC0 1.0Public domain dedication: anyone may reuse without conditions.CC BY 4.0Reuse allowed with attribution.In Copyright - Educational Use PermittedA RightsStatements.org statement, for material that is not openly licensed.
Notes:
Pick the licence from the list where you can: a standard licence is machine-readable and lets catalogues filter on it. Explain special conditions under License information.
See also:
License information
| Id | ds:f01800 |
| Path | distribution.licenseInformation |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | dcterms:rights (on each dcat:distribution; values that do not fit go to ddchd:licenseInformation) |
| Part of | Distribution |
Tooltip (level 1)
The licence information for the current data distribution
Short help (level 2)
Provide a description of the licence(s) for the current data distribution
Extended help (level 3)
Indicate the licence(s) for the current data distribution and link to the licence(s) webpage(s) if available. Be aware that some cases may exhibit different licenses for (parts of) metadata and content. Identifying different levels of copyright and constraints may also be needed.
Details (level 4)
Examples:
Texts published before 1880 are public domain. Later volumes may contain copyrighted material and are available for research use only.
Access rights
| Id | ds:f01850 |
| Path | distribution.accessRights |
| Type | Term |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | term_lookup |
| Vocabulary | Access rights (EU Authority) (https://sidedoc.app/vocabs/access-right)Enforcement: stores a Term (URI and label), only these schemes, only terms from the list, no custom labels Binding: embedded in the schema |
| DCAT-AP | dcterms:accessRights (on each dcat:distribution) |
| Part of | Distribution |
Tooltip (level 1)
Not written yet.
Short help (level 2)
The access conditions for this distribution.
Extended help (level 3)
State who may access the distribution and under what conditions (e.g. public, restricted, embargoed). Use a controlled term where available.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Notes:
Public: anyone can access the data. Restricted: access under conditions, for example after registration or a request. Non-public: not available outside the organisation.
See also:
File format
| Id | ds:f01900 |
| Path | distribution.fileFormat |
| Type | Term |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | term_lookup |
| Vocabulary | Media Types (https://www.iana.org/assignments/media-types)Enforcement: stores a Term (URI and label), only these schemes, own values allowed, custom labels allowed Binding: embedded in the schema |
| DCAT-AP | dcat:mediaType (on each dcat:distribution; values that do not fit go to ddchd:fileFormat) |
| Part of | Distribution |
Tooltip (level 1)
The format and encoding in which the data is available
Short help (level 2)
Describe the format(s) and encoding in which the data is available, using a MIME type specification.
Extended help (level 3)
Indicate the format(s) and encoding in which the data is available, using a MIME type specification. Some examples include text/csv and text/xml.
Details (level 4)
Examples:
application/xmlALTO or other XML files.text/csvimage/tiff
Avoid:
XMLUse the media type, not just a name.
See also:
Package format
| Id | ds:f01910 |
| Path | distribution.packageFormat |
| Type | Term |
| Values | 0..1 (optional, one value) |
| Multilingual | no |
| UI hint | term_lookup |
| Vocabulary | Media Types (https://www.iana.org/assignments/media-types)Enforcement: stores a Term (URI and label), only these schemes, own values allowed, custom labels allowed Binding: embedded in the schema |
| DCAT-AP | dcat:packageFormat (on each dcat:distribution; values that do not fit go to ddchd:packageFormat) |
| Part of | Distribution |
Tooltip (level 1)
The format of the package the data is delivered in
Dutch: Het formaat van het pakket waarin de data wordt geleverd
French: Le format du paquet dans lequel les données sont fournies
German: Das Format des Pakets, in dem die Daten geliefert werden
Short help (level 2)
The package format (e.g. ZIP or TAR) as a MIME type.
Dutch: Het pakketformaat (bijv. ZIP of TAR) als MIME-type.
French: Le format du paquet (p. ex. ZIP ou TAR) sous forme de type MIME.
German: Das Paketformat (z. B. ZIP oder TAR) als MIME-Typ.
Extended help (level 3)
The format of the package the data is delivered in, such as a ZIP or TAR file. Use a MIME type, e.g. application/zip. Leave empty if the data is not packaged.
Dutch: Het formaat van het pakket waarin de data wordt geleverd, zoals een ZIP- of TAR-bestand. Gebruik een MIME-type, bijv. application/zip. Laat leeg als de data niet verpakt is.
French: Le format du paquet dans lequel les données sont fournies, par exemple un fichier ZIP ou TAR. Utilisez un type MIME, p. ex. application/zip. Laissez vide si les données ne sont pas empaquetées.
German: Das Format des Pakets, in dem die Daten geliefert werden, etwa eine ZIP- oder TAR-Datei. Verwenden Sie einen MIME-Typ, z. B. application/zip. Leer lassen, wenn die Daten nicht gepackt sind.
Details (level 4)
Not written yet.
Byte size
| Id | ds:f01920 |
| Path | distribution.byteSize |
| Type | Literal |
| Values | 0..1 (optional, one value) |
| Multilingual | no |
| Validation | text: A whole number, zero or more ^\d+$As written in the sheet: {"text":"/^\\d+$/"} |
| DCAT-AP | dcat:byteSize (on each dcat:distribution; values that do not fit go to ddchd:byteSize) |
| Part of | Distribution |
Tooltip (level 1)
The size of the distribution in bytes
Dutch: De grootte van de distributie in bytes
French: La taille de la distribution en octets
German: Die Größe der Distribution in Byte
Short help (level 2)
Size in bytes, digits only. Usually best left empty.
Dutch: Grootte in bytes, alleen cijfers. Meestal beter leeg laten.
French: Taille en octets, chiffres uniquement. Le plus souvent, mieux vaut laisser vide.
German: Größe in Byte, nur Ziffern. Meist besser leer lassen.
Extended help (level 3)
The size of the distribution in bytes, as a whole number (e.g. 1048576). Leave this empty unless you need it; it may be filled in automatically in a later version.
Dutch: De grootte van de distributie in bytes, als geheel getal (bijv. 1048576). Laat dit leeg tenzij je het nodig hebt; in een latere versie wordt het mogelijk automatisch ingevuld.
French: La taille de la distribution en octets, sous forme de nombre entier (p. ex. 1048576). Laissez ce champ vide sauf si vous en avez besoin ; il pourra être rempli automatiquement dans une version ultérieure.
German: Die Größe der Distribution in Byte als ganze Zahl (z. B. 1048576). Lassen Sie dieses Feld leer, sofern Sie es nicht benötigen; es wird möglicherweise in einer späteren Version automatisch ausgefüllt.
Details (level 4)
Not written yet.
Checksum
| Id | ds:f04800 |
| Path | distribution.checksums |
| Type | TypedValue |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| Free text | yes: free text can be given instead of the structured value |
| UI hint | checksum |
| Vocabulary | Checksum/Hash Algorithms (https://www.iana.org/assignments/hash-alg-registry)Enforcement: stores the plain code, only these schemes, only terms from the list, no custom labels Binding: embedded in the schema |
| DCAT-AP | spdx:checksum (on each dcat:distribution; values that do not fit go to ddchd:checksums) |
| Part of | Distribution |
Tooltip (level 1)
Provide checksums
Short help (level 2)
Provide checksums and specify the algorithm used
Extended help (level 3)
Provide checksums and specify the checksum algorithm used (e.g., MD5, SHA-256).
Details (level 4)
Examples:
SHA-256: 9f86d081884c7d659a2feaa0c55ad015a3bf4f1b2b0b822cd15d6c15b0f00a08
Composition
| Id | ds:f02100 |
| Path | composition |
| Type | RichText, with sub-fields |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:composition (not mapped to a DCAT-AP property; written to an extension property) |
| Sub-fields | Temporal coverage, Spatial coverage, Dataset category, Media category, Object type, Dataset structure |
Tooltip (level 1)
Not written yet.
Short help (level 2)
Not written yet.
Extended help (level 3)
Provide dataset consumers with the information about the contents of the dataset, which are expected to be relevant for their chosen tasks: the kind of objects in the dataset, the structure and organisation of the dataset, the language in which it is provided and some elementary statistics.
Details (level 4)
Not written yet.
Temporal coverage
| Id | ds:f02110 |
| Path | composition.temporalCoverage |
| Type | Date or Term |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | date_picker |
| DCAT-AP | dcterms:temporal |
| Part of | Composition |
Tooltip (level 1)
Not written yet.
Short help (level 2)
The time period the dataset's content covers — a single date, a date range, or a named period.
Extended help (level 3)
Give the temporal extent of the content. Enter a single date, or switch to an interval (from/to). Dates accept variable precision (YYYY, YYYY-MM, or YYYY-MM-DD, date only). For a named era, use the term option.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Examples:
1618/1995A date range: start and end.1914-07-28/1918-11-11Dutch Golden AgeA named period, when exact dates do not apply.
Spatial coverage
| Id | ds:f02120 |
| Path | composition.spatialCoverage |
| Type | Literal or Term |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| DCAT-AP | dcterms:spatial |
| Part of | Composition |
Tooltip (level 1)
Not written yet.
Short help (level 2)
The geographic area the dataset's content covers.
Extended help (level 3)
Give the spatial extent — either free text (a place name) or a controlled term from a geographic vocabulary, which exports as a location IRI.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Examples:
NetherlandsDutch East Indies
Notes:
The area the content is about, not where the data is stored.
Dataset category
| Id | ds:f02200 |
| Path | composition.datasetCategory |
| Type | Enum |
| Values | 0..1 (optional, one value) |
| Multilingual | no |
| UI hint | enum_lookup |
| Vocabulary | Dataset Category (https://sidedoc.app/vocabs/dataset-category)Enforcement: stores a Term (URI and label), only these schemes, only terms from the list, no custom labels Binding: embedded in the schema |
| DCAT-AP | dcterms:type (values that do not fit go to ddchd:datasetCategory) |
| Part of | Composition |
Tooltip (level 1)
A main category of the resource.
Short help (level 2)
Select the appropriate answer: content and/or metadata.
Extended help (level 3)
A main category of the resource. Select the appropriate answer: content and/or metadata.
Details (level 4)
Notes:
Content: the data itself, such as texts, images or recordings. Metadata: descriptions of objects, such as catalogue records. Choose both when the dataset has both.
Media category
| Id | ds:f02300 |
| Path | composition.mediaCategory |
| Type | Term or Literal |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | term_lookup |
| DCAT-AP | dcterms:type (values that do not fit go to ddchd:mediaCategory) |
| Part of | Composition |
Tooltip (level 1)
Describe the category(ies) of media (modality) associated with the dataset.
Short help (level 2)
Describe the category(ies) of media (modality) associated with the dataset.
Extended help (level 3)
Describe the category(ies) of media (modality) associated with the dataset. For example using the list in the Europeana Data Model, an OCR dataset (including fulltext and images of documents) is associated with 'image' and 'text', a sound archive is associated with 'audio', 3D models of historical monuments are associated with '3D model', a dataset that contains only descriptive metadata is associated with 'text'.
Details (level 4)
Examples:
TextImage
Object type
| Id | ds:f02400 |
| Path | composition.objectType |
| Type | Term or Literal |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | term_lookup |
| Vocabulary | RDA Content Type (http://rdaregistry.info/termList/RDAContentType/)schema.org Classes ( https://schema.org/)Enforcement: stores a Term (URI and label), only these schemes, own values allowed, custom labels allowed Binding: embedded in the schema |
| DCAT-AP | dcterms:type (values that do not fit go to ddchd:objectType) |
| Part of | Composition |
Tooltip (level 1)
Indicate one or more specialised type(s) of CH objects covered by the dataset
Short help (level 2)
Indicate one or more specialised type(s) of CH objects covered by the dataset
Extended help (level 3)
Indicate one or more specialised type(s) of CH objects covered by the dataset, like 'newspaper', 'painting', 'sound recording', 'historical monuments', 'web pages'. Using keywords or concepts from a relevant standard vocabulary like AAT or RDA Content Type here will help make the dataset better findable.
Details (level 4)
Examples:
newspapersFrom the Art & Architecture Thesaurus (AAT).manuscripts (documents)
See also:
Dataset structure
| Id | ds:f02500 |
| Path | composition.datasetStructure |
| Type | RichText, with sub-fields |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:datasetStructure (not mapped to a DCAT-AP property; written to an extension property) |
| Part of | Composition |
| Sub-fields | Data instances, Data fields, Compliant with standard, Data splits, Language, Descriptive statistics |
Tooltip (level 1)
Provide a systematic description of the dataset, including schemas and hierarchical organisation.
Short help (level 2)
Provide a systematic description of the (conceptual and/or technical) dataset, including schemas and hierarchical organisation.
Extended help (level 3)
Provide a systematic description of the (conceptual and/or technical) dataset, including schemas and hierarchical organisation. It covers a selection of fields including data instances, data splits, language and statistics.
Details (level 4)
Not written yet.
Data instances
| Id | ds:f02600 |
| Path | composition.datasetStructure.dataInstances |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:dataInstances (Markdown text also goes to ddchd:dataInstancesMarkdown) |
| Part of | Dataset structure |
Tooltip (level 1)
Provide an example (possibly a data snippet) and brief description of a typical instance.
Short help (level 2)
Provide an example (possibly a data snippet) and brief description of a typical instance in the dataset.
Extended help (level 3)
Provide an example (possibly a data snippet) and brief description of a typical instance in the dataset. What metadata are available for the dataset items? Alternatively, it is possible to include the URL of an image, audio content or 3D model (possibly a snapshot of it).
Details (level 4)
Examples:
{ "id": "ddd:010013335:mpeg21:a0001", "title": "Binnenland", "date": "1918-11-12", "text": "…" }A short snippet of one record, with a sentence on what it shows.
Data fields
| Id | ds:f02700 |
| Path | composition.datasetStructure.dataFields |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:dataFields (Markdown text also goes to ddchd:dataFieldsMarkdown) |
| Part of | Dataset structure |
Tooltip (level 1)
List and describe the fields present in the dataset.
Short help (level 2)
List and describe the fields present in the dataset. Mention their data type (number, string, date, coordinates) and whether they are used as input or output in any of the tasks the dataset currently supports
Extended help (level 3)
List and describe the fields present in the dataset. Mention their data type (number, string, date, coordinates), and whether they are used as input or output in any of the tasks the dataset currently supports. If the data has e.g. span indices, describe their attributes, such as whether they are at the character or word level, whether they are contiguous or not, etc.
Details (level 4)
Examples:
date (string, ISO 8601): publication date of the issueocr_confidence (number, 0 to 1): average OCR word confidence for the page
Compliant with standard
| Id | ds:f02800 |
| Path | composition.datasetStructure.compliantWithStandard |
| Type | Term or Literal |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | term_lookup |
| DCAT-AP | dcterms:conformsTo (values that do not fit go to ddchd:compliantWithStandard) |
| Part of | Dataset structure |
Tooltip (level 1)
A standard data model or schema and/or established data practice, for example W3C Web Annotation.
Short help (level 2)
A standard data model or schema and/or established data practice, for example W3C Web Annotation.
Extended help (level 3)
A standard data model or schema and/or established data practice, for example W3C Web Annotation, that the dataset conforms to. It is appropriate to use a web identifier or link to the standard or its documentation, like http://www.cidoc-crm.org/cidoc-crm/. It can be version-specific, like http://id.loc.gov/ontologies/bibframe-2-5-0/.
Details (level 4)
Examples:
ALTO 2.0IIIF Presentation API 3.0W3C Web Annotation Data Model
Data splits
| Id | ds:f02900 |
| Path | composition.datasetStructure.dataSplits |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:dataSplits (Markdown text also goes to ddchd:dataSplitsMarkdown) |
| Part of | Dataset structure |
Tooltip (level 1)
Describe and name machine learning splits in the dataset if there are any.
Short help (level 2)
Describe and name machine learning splits in the dataset if there are any. Describe any criteria for splitting the data, if used.
Extended help (level 3)
Describe and name machine learning splits in the dataset if there are any. Describe any criteria for splitting the data, if used. If there are differences between the splits (e.g. if the training annotations are machine-generated and the development and test ones are created by humans), describe them here. Provide the sizes of each split.
Details (level 4)
Not written yet.
Language
| Id | ds:f03000 |
| Path | composition.datasetStructure.language |
| Type | Term or Literal |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | term_lookup |
| Vocabulary | ISO 639-2 Language Codes (http://id.loc.gov/vocabulary/iso639-2)Enforcement: stores the plain code, only these schemes, only terms from the list, no custom labels Binding: embedded in the schema |
| DCAT-AP | dcterms:language (values that do not fit go to ddchd:language) |
| Part of | Dataset structure |
Tooltip (level 1)
Explicitly mention the languages present in the dataset
Short help (level 2)
Explicitly mention the languages present in the dataset
Extended help (level 3)
Explicitly mention the languages present in the dataset (possibly in broad terms, e.g. translations between several pairs of European languages). Describe relevant details about specifics of the language using e.g. ISO 639-2 codes.
Details (level 4)
Examples:
dutDutch, as an ISO 639-2 code.fryWestern Frisian.
Notes:
List the languages of the content, not of this description. Add each language as a separate value.
Descriptive statistics
| Id | ds:f03100 |
| Path | composition.datasetStructure.descriptiveStatistics |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:descriptiveStatistics (Markdown text also goes to ddchd:descriptiveStatisticsMarkdown) |
| Part of | Dataset structure |
Tooltip (level 1)
Provide valuable information about the richness of the dataset.
Short help (level 2)
Provide valuable information about the richness of the dataset.
Extended help (level 3)
The information presented in this section should be chosen on the basis of whether it tells something valuable about the richness of the dataset. How large is the dataset, both in cardinality and in disk storage? What are its key descriptive statistics per field? Note that statistics that reflect the richness of data are likely to be domain-specific, i.e. for musical datasets, you might use average recording length, distribution by genre, sampling rate, number of tracks, etc.
Details (level 4)
Examples:
11.2 million pages from 1.9 million issues; 8.4 billion OCR words; 70% of pages from 1850–1940.
Data collection process
| Id | ds:f03200 |
| Path | dataCollectionProcess |
| Type | RichText, with sub-fields |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | dcterms:provenance (Markdown text also goes to ddchd:dataCollectionProcessMarkdown) |
| Sub-fields | Curation rationale, Source data, Initial data collection, Digitisation pipeline, Preprocessing and cleaning, Annotations, Annotator, Crowd labor, Data provenance, Use of Linked Open Data, Version information, Maintenance plan |
Tooltip (level 1)
Document data collection and curation history
Short help (level 2)
Describe how the dataset was created and maintained
Extended help (level 3)
This section documents the processes, decisions, people, tools and sources that shaped the dataset, including selection, source data, preprocessing, annotations, provenance, linked data, versions and maintenance.
Details (level 4)
Not written yet.
Curation rationale
| Id | ds:f03300 |
| Path | dataCollectionProcess.curationRationale |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | dcterms:provenance (values that do not fit go to ddchd:curationRationale; Markdown text also goes to ddchd:curationRationaleMarkdown) |
| Part of | Data collection process |
Tooltip (level 1)
Explain the curation rationale
Short help (level 2)
Explain selection criteria and curatorial decisions
Extended help (level 3)
Describe the rationale for selecting, excluding, sampling, transforming or grouping materials, including institutional, research, legal, technical or practical motivations.
Details (level 4)
Examples:
All newspapers held by the KB that were digitised in the national programme 2006–2018, excluding titles still in copyright for which no permission was given.
Source data
| Id | ds:f03400 |
| Path | dataCollectionProcess.sourceData |
| Type | Container (group of sub-fields) |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | url |
| Row caption | {title} |
| DCAT-AP | ddchd:sourceData |
| Part of | Data collection process |
| Sub-fields | Title, Link, Description, Source data producer |
Tooltip (level 1)
Describe the source data
Short help (level 2)
Identify the original source data and holding context
Extended help (level 3)
Describe the source collections, systems, catalogues, institutions, creators, rights, access conditions and any relevant historical or cultural context from which the dataset was derived.
Details (level 4)
Not written yet.
Title
| Id | ds:f03410 |
| Path | dataCollectionProcess.sourceData.title |
| Type | Literal |
| Values | 1..1 (required, one value) |
| Multilingual | yes |
| Part of | Source data |
Tooltip (level 1)
Not written yet.
Short help (level 2)
A short title for this source dataset or collection.
Extended help (level 3)
Name the source the data was derived from — a collection, corpus, archive, or upstream dataset — so users can recognise it.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Not written yet.
Link
| Id | ds:f03420 |
| Path | dataCollectionProcess.sourceData.link |
| Type | Url |
| Values | 0..1 (optional, one value) |
| Multilingual | no |
| Part of | Source data |
Tooltip (level 1)
Not written yet.
Short help (level 2)
A URL to the source dataset or collection.
Extended help (level 3)
Link to the source's landing page or record. Use a stable, resolvable URL where possible; in DCAT exports this becomes the source dataset's identifier and landing page.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Not written yet.
Description
| Id | ds:f03430 |
| Path | dataCollectionProcess.sourceData.description |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| Part of | Source data |
Tooltip (level 1)
Not written yet.
Short help (level 2)
A description of the source data and its relationship to this dataset.
Extended help (level 3)
Describe what the source contains and how it was used to produce this dataset — provenance, selection, and scope.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Not written yet.
Source data producer
| Id | ds:f03600 |
| Path | dataCollectionProcess.sourceData.sourceDataProducer |
| Type | Agent |
| Values | 0..* (optional, repeatable) |
| Multilingual | yes |
| Free text | yes: free text can be given instead of the structured value |
| Identifier schemes | identifier: ORCID, VIAF, ISNI, ROR, Wikidata, GND |
| Part of | Source data |
Tooltip (level 1)
Identify the data producers
Short help (level 2)
Identify the data producers (human or machine-generated)
Extended help (level 3)
State whether the source data or source objects were produced by humans (such as the writers of books) or machine generated and provide a description for them. If available, include self-reported demographic or identity information for the creators of the source data or objects, but avoid inferring this information. Instead, state that this information is unknown.
Details (level 4)
Not written yet.
Initial data collection
| Id | ds:f03500 |
| Path | dataCollectionProcess.initialDataCollection |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:initialDataCollection (Markdown text also goes to ddchd:initialDataCollectionMarkdown) |
| Part of | Data collection process |
Tooltip (level 1)
Describe the data collection process
Short help (level 2)
Describe any criteria for data selection, filtering or harvesting
Extended help (level 3)
Describe the data collection process. Describe any criteria for data selection, filtering or harvesting (in particular, list any keywords or search terms used). If possible and relevant, include runtime information for the collection process. If data was collected from other pre-existing datasets, link to the source here.
Details (level 4)
Not written yet.
Digitisation pipeline
| Id | ds:f03700 |
| Path | dataCollectionProcess.digitisationPipeline |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | dcterms:provenance (values that do not fit go to ddchd:digitisationPipeline; Markdown text also goes to ddchd:digitisationPipelineMarkdown) |
| Part of | Data collection process |
Tooltip (level 1)
Describe the digitisation process
Short help (level 2)
Describe the digitisation process, and any selection effects
Extended help (level 3)
Provide information on the digitisation process relevant for reuse, such as the motivation for digitisation, image capture equipment, OCR/HTR software and parameters, and quality metrics. If applicable, describe any selection introduced by digitisation, including selection criteria, coverage metrics, and the extent to which the digitised material represents the original collection. State if no such selection occurred.
Details (level 4)
Examples:
Microfilm scanned at 400 dpi greyscale; OCR with ABBYY FineReader 9; no manual correction.
Notes:
Mention what was digitised from (originals or microfilm), the scan settings, the OCR or HTR software and version, and any quality checks. Selection effects matter: say what was left out and why.
Preprocessing and cleaning
| Id | ds:f03800 |
| Path | dataCollectionProcess.dataPreProcessing |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | dcterms:provenance (values that do not fit go to ddchd:dataPreProcessing; Markdown text also goes to ddchd:dataPreProcessingMarkdown) |
| Part of | Data collection process |
Tooltip (level 1)
Describe preprocessing and cleaning
Short help (level 2)
Describe transformations applied before publication or use
Extended help (level 3)
Document any digitisation, OCR/HTR, transcription, conversion, normalisation, filtering, deduplication, enrichment, quality control or cleaning steps.
Details (level 4)
Not written yet.
Annotations
| Id | ds:f03900 |
| Path | dataCollectionProcess.annotations |
| Type | Container (group of sub-fields) |
| Values | 0..1 (optional, one value) |
| Multilingual | no |
| Part of | Data collection process |
| Sub-fields | Annotation process |
Tooltip (level 1)
Describe annotations
Short help (level 2)
Document labels, annotations and enrichments
Extended help (level 3)
Describe annotations or enrichments added to the dataset, including annotation tasks, labels, guidelines, tools, quality assurance and agreement measures where available.
Details (level 4)
Not written yet.
Annotation process
| Id | ds:f04000 |
| Path | dataCollectionProcess.annotations.annotationProcess |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | dcterms:provenance (values that do not fit go to ddchd:annotationProcess; Markdown text also goes to ddchd:annotationProcessMarkdown) |
| Part of | Annotations |
Tooltip (level 1)
Describe the annotation process
Short help (level 2)
Describe the annotation process, guidelines, coverage, validation, and agreement metrics
Extended help (level 3)
If applicable, describe the annotation process, tools used, annotation guidelines, data coverage, aggregation methods, limitations, validation procedures, and any available inter-annotator agreement statistics.
Details (level 4)
Not written yet.
Annotator
| Id | ds:f04100 |
| Path | dataCollectionProcess.annotator |
| Type | Agent |
| Values | 0..* (optional, repeatable) |
| Multilingual | yes |
| Free text | yes: free text can be given instead of the structured value |
| Identifier schemes | identifier: ORCID, VIAF, ISNI, ROR, Wikidata, GND |
| DCAT-AP | prov:qualifiedAttribution (values that do not fit go to ddchd:annotators) |
| Part of | Data collection process |
Tooltip (level 1)
Describe annotators
Short help (level 2)
Identify who created annotations
Extended help (level 3)
If annotations were collected for the source data (such as class labels), state whether the annotations were produced by humans or machine generated. Describe the people or systems who originally created the annotations and their selection criteria, if applicable. If available, include self-reported demographic or identity information for the annotators, but avoid inferring this information. Instead, state that this information is unknown.
Details (level 4)
Not written yet.
Crowd labor
| Id | ds:f04200 |
| Path | dataCollectionProcess.crowdLabor |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | dcterms:provenance (values that do not fit go to ddchd:crowdLabor; Markdown text also goes to ddchd:crowdLaborMarkdown) |
| Part of | Data collection process |
Tooltip (level 1)
Describe crowd labour if used
Short help (level 2)
Document crowd, volunteer or platform labour
Extended help (level 3)
Explain whether crowd or volunteer labour was used, how participants were recruited and instructed, compensation or recognition, consent, safeguards and quality control.
Details (level 4)
Not written yet.
Data provenance
| Id | ds:f04300 |
| Path | dataCollectionProcess.dataProvenance |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | dcterms:provenance (values that do not fit go to ddchd:dataProvenance; Markdown text also goes to ddchd:dataProvenanceMarkdown) |
| Part of | Data collection process |
Tooltip (level 1)
Describe data provenance
Short help (level 2)
Document any provenance considerations that may have influenced data collection
Extended help (level 3)
If applicable, provide additional information on the aspects of provenance that have impacted the data collection, for example ownership, licences and other obligations associated with sources of data, such as rights attached to a physical collection.
Details (level 4)
Not written yet.
Use of Linked Open Data
| Id | ds:f04400 |
| Path | dataCollectionProcess.linkedOpenDataUse |
| Type | Term or Literal |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| UI hint | term_lookup |
| Vocabulary | Common Linked Open Data Namespaces (https://sidedoc.app/vocabs/common-namespaces)Enforcement: stores a Term (URI and label), other schemes allowed, own values allowed, custom labels allowed Binding: embedded in the schema |
| DCAT-AP | dcat:qualifiedRelation (values that do not fit go to ddchd:linkedOpenDataUse) |
| Part of | Data collection process |
Tooltip (level 1)
Describe linked open data and vocabularies
Short help (level 2)
If used, document LOD, identifiers, controlled vocabularies and ontologies
Extended help (level 3)
Describe linked open data schemes, controlled vocabularies, ontologies and taxonomies if used during the establishment of the dataset. It is appropriate to use a web identifier or link to the vocabulary (possibly version-specific).
Details (level 4)
Not written yet.
Version information
| Id | ds:f04500 |
| Path | dataCollectionProcess.versionInformation |
| Type | RichText, with sub-fields |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | adms:versionNotes (Markdown text also goes to ddchd:versionInformationMarkdown) |
| Part of | Data collection process |
| Sub-fields | Release date, Date of modification |
Tooltip (level 1)
Describe version information
Short help (level 2)
Record version, release date and/or change information
Extended help (level 3)
Provide information on the dataset version. Alternatively, provide a release date and a date when the dataset was last updated as well as information on whether there was a previous version of the dataset published before. If relevant, describe which versioning policy drives growing or changing datasets. Indicate if the status of previous versions has become deprecated.
Details (level 4)
Notes:
Fill in release date for every published version. Add a modification date when you change the data without a new release.
Release date
| Id | ds:f04600 |
| Path | dataCollectionProcess.versionInformation.releaseDate |
| Type | DateTime |
| Values | 0..1 (optional, one value) |
| Multilingual | no |
| Free text | yes: free text can be given instead of the structured value |
| UI hint | date_picker |
| Date mode | single |
| DCAT-AP | dcterms:issued (values that do not fit go to ddchd:releaseDate) |
| Part of | Version information |
Tooltip (level 1)
Record the timestamp of this version
Short help (level 2)
Specify the timestamp associated with this version
Extended help (level 3)
If applicable, fill the timestamp of the date of release of this version.
Details (level 4)
Examples:
2023-06-01
Date of modification
| Id | ds:f04700 |
| Path | dataCollectionProcess.versionInformation.modificationDate |
| Type | DateTime |
| Values | 0..1 (optional, one value) |
| Multilingual | no |
| Free text | yes: free text can be given instead of the structured value |
| UI hint | date_picker |
| Date mode | single |
| DCAT-AP | dcterms:modified (on the catalog record (dcat:CatalogRecord), not the dataset) |
| DCAT-AP | dcterms:modified (values that do not fit go to ddchd:modificationDate) |
| Part of | Version information |
Tooltip (level 1)
Record the timestamp of the latest modification
Short help (level 2)
Specify the timestamp associated with the date of modification of this version
Extended help (level 3)
If applicable, fill the timestamp of the date of (minor) modification of this version.
Details (level 4)
Not written yet.
Maintenance plan
| Id | ds:f04900 |
| Path | dataCollectionProcess.maintenancePlan |
| Type | RichText, with sub-fields |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:maintenancePlan (not mapped to a DCAT-AP property; written to an extension property) |
| Part of | Data collection process |
| Sub-fields | Maintenance level, Update periodicity |
Tooltip (level 1)
Describe the maintenance plan
Short help (level 2)
Provide details on the dataset's maintenance plan
Extended help (level 3)
Describe who maintains the dataset, how corrections or issues are handled, maintenance level, update periodicity, planned updates and end-of-support expectations.
Details (level 4)
Not written yet.
Maintenance level
| Id | ds:f05000 |
| Path | dataCollectionProcess.maintenancePlan.maintenanceLevel |
| Type | Term |
| Values | 0..1 (optional, one value) |
| Multilingual | no |
| UI hint | term_lookup |
| Vocabulary | Maintenance Level (https://sidedoc.app/vocabs/maintenance-level)Enforcement: stores a Term (URI and label), only these schemes, only terms from the list, no custom labels Binding: embedded in the schema |
| DCAT-AP | ddchd:maintenanceLevel |
| Part of | Maintenance plan |
Tooltip (level 1)
State the level of maintenance commitment
Short help (level 2)
State the level of commitment: actively maintained, limited maintenance, no maintenance
Extended help (level 3)
Describe the level of commitment for the maintenance of the dataset. Three suggestions: a) Actively Maintained - This dataset will be actively maintained, including but not limited to updates to the data. b) Limited Maintenance - The data will not be updated, but any technical issues will be addressed. c) No maintenance.
Details (level 4)
Notes:
Be honest: 'no maintenance' is useful information. It tells users not to expect corrections.
Update periodicity
| Id | ds:f05100 |
| Path | dataCollectionProcess.maintenancePlan.updatePeriodicity |
| Type | Term |
| Values | 0..1 (optional, one value) |
| Multilingual | no |
| UI hint | term_lookup |
| Vocabulary | EU Frequency Vocabulary (http://publications.europa.eu/resource/authority/frequency)Enforcement: stores a Term (URI and label), only these schemes, only terms from the list, custom labels allowed Binding: embedded in the schema |
| DCAT-AP | dcterms:accrualPeriodicity |
| Part of | Maintenance plan |
Tooltip (level 1)
The update frequency
Short help (level 2)
Specify the expected update frequency
Extended help (level 3)
If applicable, what is the expected update frequency of the dataset?
Details (level 4)
Examples:
AnnualIrregular
Examples and considerations for using the data
| Id | ds:f05200 |
| Path | examplesAndUsageConsiderations |
| Type | RichText, with sub-fields |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:usageConsiderations (not mapped to a DCAT-AP property; written to an extension property) |
| Sub-fields | Ethical considerations, Example of dataset reuse, Non-ethical limitations, Unanticipated uses |
Tooltip (level 1)
Reflections about ethical and non-ethical conditions of use and cases of reuse
Short help (level 2)
Define the intended scope and provide usage guidelines for the dataset
Extended help (level 3)
This section should outline the known and potential tasks or research areas that could match with this dataset, hinting at the original ones this dataset was designed for. Outline clear guidelines on how the data should be used to support its integrity and meaningful use, naming reuse cases (if known or present), explicitly mentioning tasks for which the dataset is not suitable.
Details (level 4)
Not written yet.
Ethical considerations
| Id | ds:f05300 |
| Path | examplesAndUsageConsiderations.ethicalConsiderations |
| Type | Container (group of sub-fields) |
| Values | 0..1 (optional, one value) |
| Multilingual | no |
| DCAT-AP | ddchd:ethicalConsiderations (not mapped to a DCAT-AP property; written to an extension property) |
| Part of | Examples and considerations for using the data |
| Sub-fields | Personal or other sensitive information, Discussion of biases, Potential societal impact of using the dataset |
Tooltip (level 1)
Information about sensitivity, privacy, biases and their potential impact
Short help (level 2)
Provide information about eventual data sensitivities, identifying emergent ethical issues, discussing eventual bias or modelling occurrences, sketching broad societal implications
Extended help (level 3)
Provide a narrative account of sensitive elements within the dataset. Identify who or what is depicted, addressing privacy compliance (GDPR), whether individuals are identifiable, and if the data pertains to difficult or contested histories. Discuss known biases in the collection (e.g., representation gaps) and any mitigation or modelling steps taken. Evaluate the potential societal impact, such as empowering under-represented groups or the risk of reinforcing harmful stereotypes. State whether impacted communities were consulted during curation or if they should be engaged in reuse instances.
Details (level 4)
Not written yet.
Personal or other sensitive information
| Id | ds:f05400 |
| Path | examplesAndUsageConsiderations.ethicalConsiderations.personalAndSensitiveInformation |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:personalAndSensitiveInformation |
| Part of | Ethical considerations |
Tooltip (level 1)
Not written yet.
Short help (level 2)
Whether the dataset contains personal or otherwise sensitive information.
Extended help (level 3)
Note any personal data, special-category data, or otherwise sensitive content, and how it is handled — anonymisation, consent, or access restrictions.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Examples:
Yes. Family announcements and court reports contain names of private persons, some of whom may still be alive.
Discussion of biases
| Id | ds:f05500 |
| Path | examplesAndUsageConsiderations.ethicalConsiderations.discussionOfBiases |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:biasDiscussion (not mapped to a DCAT-AP property; written to an extension property) |
| Part of | Ethical considerations |
Tooltip (level 1)
Not written yet.
Short help (level 2)
Known biases in the dataset and their possible effects.
Extended help (level 3)
Describe biases in collection, sampling, labelling, or representation that could affect analyses or downstream models, so reusers can account for them.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Examples:
Coverage favours national and urban newspapers; regional and colonial titles are under-represented. OCR quality is lower for pre-1850 print, so text-based analyses are skewed to later years.
Potential societal impact of using the dataset
| Id | ds:f05600 |
| Path | examplesAndUsageConsiderations.ethicalConsiderations.societalImpact |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:societalImpact (not mapped to a DCAT-AP property; written to an extension property) |
| Part of | Ethical considerations |
Tooltip (level 1)
Not written yet.
Short help (level 2)
Potential societal impact of using the dataset.
Extended help (level 3)
Reflect on possible positive and negative societal effects of using the dataset, including risks of misuse.
⚠️ AI-generated placeholder — to be replaced with the official Datasheet v2.1 template content.
Details (level 4)
Not written yet.
Example of dataset reuse
| Id | ds:f05700 |
| Path | examplesAndUsageConsiderations.reuseExample |
| Type | QualifiedRelation |
| Values | 0..* (optional, repeatable) |
| Multilingual | no |
| DCAT-AP | dcat:qualifiedRelation (values that do not fit go to ddchd:reuseExample) |
| Part of | Examples and considerations for using the data |
Tooltip (level 1)
Links to known derivative works featuring this dataset.
Short help (level 2)
Provide links to known publications, models trained on this dataset, or research publications that offer examples of reuse of this resource
Extended help (level 3)
List URLs or citations for existing reuses of this data. This includes derivative datasets, AI/ML models trained on this information, or research publications (including Jupyter notebooks) that demonstrate how the data can be successfully reused to produce novel outputs.
Details (level 4)
Examples:
https://doi.org/10.1234/example-reuse-studyA publication that used the dataset.
Non-ethical limitations
| Id | ds:f05800 |
| Path | examplesAndUsageConsiderations.nonEthicalLimitations |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:nonEthicalLimitations (not mapped to a DCAT-AP property; written to an extension property) |
| Part of | Examples and considerations for using the data |
Tooltip (level 1)
Technical and structural constraints requiring attention.
Short help (level 2)
Describe the limitations related to non-ethical aspects, such as gaps in data collection, annotation inconsistencies or additional factors impacting on the interpretation of the dataset
Extended help (level 3)
Detail limitations that are not ethical in nature but that may impact on data modelling if left unacknowledged. This includes the eventual presence of annotation artifacts (e.g. errors or irregularities introduced during labeling), gaps in the collection process, technical inconsistencies, or any other factor that may impact modelling and interpretation of the data.
Details (level 4)
Not written yet.
Unanticipated uses
| Id | ds:f05900 |
| Path | examplesAndUsageConsiderations.unanticipatedUses |
| Type | RichText |
| Values | 0..1 (optional, one value) |
| Multilingual | yes |
| DCAT-AP | ddchd:unanticipatedUses (not mapped to a DCAT-AP property; written to an extension property) |
| Part of | Examples and considerations for using the data |
Tooltip (level 1)
Unintended applications and edge cases
Short help (level 2)
Document or sketch instances of use and reuse not envisioned as part of the original dataset design
Extended help (level 3)
Record cases where the dataset was (or could) be used in contexts outside its original design, complementing the examples provided in Example of Dataset Reuse. Note whether these 'off-script' uses provided or may provide new insights or if they surfaced unexpected risks that future users should be warned about.
Details (level 4)
Not written yet.