NERDm: the NIST Extensible Resource Data model.

Human-readable, descriptive name of the resource

• Acronyms should be avoided

a string indicating the version of the release of this resource

a list describing the release of different versions of this resource

a listing of other existing resources that are deprecated by this resource

another existing resource that should be considered a replacement for this resource

• This typically refers to a newer version of this resource
• In the PDR system, this value is automatically generated given the overall state of the PDR's catalog

Human-readable description (e.g., an abstract) of the resource

• Each list element should be considered as a separate paragraph.

• Each element in the array should be considered a separate paragraph

Tags (or keywords) help users discover your dataset; please include terms that would be used by technical and non-technical users.

• Each value in the list represents a different keyword. Keywords can be made up of multiple words

• Surround each keyword with quotes. Separate keywords with commas. Avoid duplicate keywords in the same record.

A tag or label taken from a pre-defined vocabulary for a thing or concept that this resource addresses or speaks to

Most recent date on which the dataset was changed, updated or modified.

• The value is usually a single formated date, like 2017-04-21, or a data and time, like 2019-02-23T14:31:10.3
• A special format is used to indicate the resource is regularly update; e.g., R/P1W (for once a week); see ISO8601DateRange for details.

• Dates should be ISO 8601 of highest resolution. In other words, as much of YYYY-MM-DDThh:mm:ss.sTZD as is relevant to this dataset. If there is a need to reflect that the dataset is continually updated, ISO 8601 formatting can account for this with repeating intervals. For instance, R/P1D for daily, R/P2W for every two weeks, and R/PT5M for every five minutes.

Date of formal issuance of the resource

• The value is usually a single formated date, like 2017-04-21

The publishing entity and optionally their parent organization(s).

• This is a container for a publisher object which groups together the fields: name and subOrganization. The subOrganization field can also contain a publisher object which allows one to describe an organization's hierarchy. Where greater specificity is desired, include as many levels of publisher as is useful, in ascending order, using the below format.

Contact information for getting more information about this resource

• This should include at least a name and an email address
• The information can reflect either a person or a group (such as a help desk)

The degree to which this dataset is publicly-available (or is restricted in some way)

public

Data asset is or could be made publicly available to all without restrictions

restricted public

Data asset is available under certain use restrictions

non-public

Data asset is not available to members of the public

A pointer to the primary license or non-license (i.e. Public Domain) statement with which the dataset or API has been published

• Software and data developed primarily by federal employees must be considered in the public domain; software primarily developed by contract can be assigned a license, including an open source license.
• By default, NIST-produced data and software should point to http://www.nist.gov/data/license.cfm
• See Open Licenses (https://project-open-data.cio.gov/open-licenses/) for more information.

information regarding access or restrictions based on privacy, security, or other policies

• This should also provide an explanation for the selected "accessLevel" including instructions for how to access a restricted file, if applicable, or explanation for why a "non-public" or "restricted public" data assetis not "public," if applicable.
• Text must be 255 or fewer characters.

a listing of the various component resources, tools, and distributions of this resource

• This list often includes the list of data files (represented as DataFile components) that are available to download.
• Data files can be organized into a hierarchy of subcollections (like file folders). When this is the case, this list will also include Subcollection components. Both DataFile and Subcollection components included a filepath that indicates where it is located in the hierarchy.

• Records for the resources referenced in this list should specify this resource in its isPartOf field.
• The @type property will indicate which type of component it is. The first value in the @type list should be considered the primary or most important classification.
• If an item's @type includes dcat:Distribution, the item should be convertable to a distribution in the POD schema.
• The order should be considered meaningful (at least for components of the same type). The meaning or intention behind the order can depend on the type; however, generally, display of the components of a common type should preserve the order. For clarity then, it is recommended that items of the same primary type should be grouped together in the list.

URI used to identify a standardized specification the resource conforms to

• This identifier might refer to a file or dataset format, to database schema, or to a web API.

The collection of which the dataset is a subset

• The resource being referenced normally is one that is also described by a NERDm record which is accessible via its identifier (given in the proxyFor property).

The primary language used in the dataset

a URL to a human-friendly web page for accessing the resource.

Other documents describing or relating to this resource

Main thematic category of the dataset.

• Values are taken from the Taxonomy of NIST Research Areas.

• Could include ISO Topic Categories (http://www.isotopicmaps.org/)

The NIST-assigned identifier for the resource

• This identifier takes a the form of a NIST-issued ARK identifier (and thus starts with ark:/88434/)
• If the resource has been assigned a DOI, it will appear in the doi

• It is expected that this field will contain a type of identifier that is of a uniform type across all resource descriptions in the system
• This identifier should, by default, resolve to the metadata record (or some rendering of it) rather than to the resource itself (e.g. via its landing page).

The Digital Object Identifier (DOI)

• The DOI value should not have the resolver URI base, but should have the form 'doi:NNNNN/XXXXXX'
• It is expected that this will primarily be Datacite DOIs

the NIST EDI identifier assigned to the resource

an abbreviated form of the resource's title

• this can be used as a label for a compact display or text for a link to this resource

the linked-data class types for this resource

• This property is a list because it is possible for a resource to be catorgorized as multiple types simultaneously
• The first value in the list will be the most specific resource type that applies to the resource and should be considered, therefore, its primary type.
• Often, the values refer to Named Types defined explicitly in the NERDm schema (e.g. DataPublication); in this case, the resource metadata can include the extended metadata properties for that type.
• Some values are just semantic labels that do not imply the existance of additional extended metadata.
• These values are intended to be compliant with use of @type property in the JSON-LD format.

• The value must always be given as an array, even when only one type is provided.
• Multiple values indicate that the Resource can be considered of multiple types
• If this resource is not to be considered a subtype of the nrd:Resource, its value should be 'nrd:Resource'.

Frequency with which dataset is published.

an identifier provided by the OMB Circular A-11, Appendix C that identifies the originating federal agency

• OMB Circular A-11, Appendix C is available via http://www.whitehouse.gov/sites/default/files/omb/assets/a11_current_year/app_c.pdf
• A machine-readable listing of the defined codes is available via https://project-open-data.cio.gov/data/omb_bureau_codes.csv
• Codes have the format of 015:01

an identifier provided by the Federal Program Inventory that identifies the primary program related to this data asset

• A machine-readable listing of the defined codes is available via https://www.performance.gov/sites/default/files/files/FederalProgramInventory_FY13_MachineReadable_091613.xls
• Codes have the format of 015:001

Whether the dataset meets the agency's Information Quality Guidelines

The IT Unique Investment Identifier (UII) that is associated with this dataset

The URL to the System of Records Notice related to this dataset if is so designated under the Privacy Act of 1974

a secondary or sub-title for the resource

other (unofficial) titles that this resource is sometimes known as

the ordered list of authors of this data publication

• Authors should generally be assumed to be considered creators of the data; where this is is not true or insufficient, the contributors property can be used ot add or clarify who contributed to data creation.

a recommended formatting of a citation to this data publication

a relative identifier for the component.

• Though not required by this schema, providing an identifier for a component is critical for merging information from different sources.
• A relative identifier requires that @base be set to the Resource identifier in the @context.

the types of components that this component can be classified as

• This property is a list because it is possible for a component to be catorgorized as multiple types simultaneously
• An included value of nrd:Hidden indicates that the component is part of the NIST-internal management of the resource and is not expected to be useful to external users
• These values are intended to be compliant with use of @type property in the JSON-LD format.

• the first value should be considered its primary type. This is usually a subtype of Component.

a descriptive title for the component

a description of the nature and contents of the component, including the role it plays as part of the resource

A tag or label taken from a pre-defined vocabulary for a thing or concept that this component addresses or speaks to

URI used to identify a standardized specification the component conforms to

a relative identifier for this component

• @id should be the ID for the included resource as a component of this resource; use the required proxyFor property to give the native ID of the included resource.

the linked-data class types for the included resource

• The @type should contain values that are subclasses of Component; resourceType is provided to list the Resource types for the included resource.
• Just like a Resource:@type, the value must always be given as an array, even when only one type is provided.
• Multiple values indicate that the Resource can be considered of multiple types
• If this resource is not to be considered a subtype of the nrd:Resource, its value should be 'nrd:Resource'.

a name for the data file reflecting its hierarchical location in the data source collection

• Forward slashes delimit the hierarchical names in the path
• If there are no slashes in this value, this file is assumed to be at the top of the file hierarchy
• The base name of this value (i.e. the last field in the path) can be used as the default filename to give to the file if downloaded.
• The component title may have the same value.

URL providing direct access to a downloadable file of a dataset

The machine-readable file format (IANA Media Type or MIME Type) of the file’s downloadURL

A human-readable description of the file format of a distribution

a checksum for the file

the size of the file in bytes

URL to the data dictionary for the distribution found at the downloadURL

The machine-readable file format (IANA Media Type or MIME Type) of the file’s describedBy URL

the algorithm used to produce the checksum hash

A flag, if True, indicating the the hash value contained in this ChecksumFile is confirmed to be correct for its associated data file.

a name for the data file reflecting its hierarchical location in the data source collection

• Forward slashes delimit the hierarchical names in the path
• If there are no slashes in this value, this file is assumed to be at the top of the file hierarchy
• The base name of this value (i.e. the last field in the path) can be used as the default filename to give to the file if downloaded.
• The component title may have the same value.

a listing of resource components that are directly part of this subcollection

• Each item is a URI identifier (possibly abbreviated)

The identifier for the parent collection that contains this subcollection

the URL for accessing this indirect access to the resource

A human-readable description of the file format of a distribution

the base URL for accessing the API

URL to a formal or informal description of the API

• Use describedByType to help distinguish between formal and informal (i.e. human readable) descriptions: no describedByType value should indicate a human-readable document.

• Use describedByType to help distinguish between formal and informal (i.e. human readable) descriptions.

The machine-readable file format (IANA Media Type or MIME Type) of the API's describedBy URL

• If no value is given for this property, it can be assumed that the document given by the describedBy URL is human-readable documentation for the API.

an identifier for the reference

• The most well-known PID used to identify the resource (see also notes for proxyFor
• This identifier can be a relative URL which can be used in special cases (See the notes for BibliographicReference:@type for when this ID should not refer to the referenced work).
• Though not required by this schema, providing an identifier for a component is critical for merging information from different sources.
• A relative identifier requires that @base be set to the Resource identifier in the @context.

• The single string value option is provided for backward compatibility (and may be removed in future versions; see BibliographicReference). When this is used, the node represents a reference and not the thing being referenced; the @id contains a relative URI, representing this reference

the name of the resource being referenced

• This value is intended for display.
• This can be, but is not required to be, the same title given in the metadata record describing the resource

a local identifier representing this resource

• This identifier is expected to point to an up-to-date description of the resource as known to the local system. The properties associated with that identifier may be different those given in the current record.

the URL for accessing the resource

a recommended label or title to display as the text for a link to the document

• This is intended to be briefer than a citation. It should be short enough to use as tag to an entry in reference list as inserted into a document's text

The date the referenced resource was issued

• This is intended for determining relative newness--i.e. being newer or older than the referring resource.

a brief, human-readable description of what this reference refers to and/or why it is being referenced.

a string indicating a version of the release of this resource

• label could contain the same value

a full formated citation string for the reference, appropriate for inclusion in a bibliography

the type of relationship that this document has with the resource

• This is equivalent to the Datacite relationType in that the term is a predicate that connects the resource as the subject to the referenced document as the object (e.g. resource IsDocumentedBy referenced-doc)
• The DCiteReference type sets DataCite terms as controlled vocabulary

a term indicating the nature of the relationship between this resource and the one being referenced

• Note that with this term, the subject of relationship is the resource described by this NERDm record and the object is the referenced resource given by the @id property in this node. Although this violates the JSON-LD semantics that properties in this node should describe what's given with the local @id--the referenced resource, in this case--it is more consistant with their use in the DataCite schema.

IsDocumentedBy

The referenced document provides documentation of this resource.

IsSupplementedTo

The referenced document is a supplement to this resource.

IsCitedBy

The referenced document cites the resource in some way.

Cites

This resource cites the referenced document.

IsReviewedBy

The referenced document reviews this resource.

IsReferencedBy

The resource is used as a source of information by the referenced document.

References

The referenced document is used as a source of information by the resource.

IsSourceOf

The resource is the source of upon which the referenced resource is based.

IsDerivedFrom

The referenced document is the source upon which the resource is based.

IsNewVersionOf

The referenced resource is a previous version of this resource.

IsPreviousVersionOf

The referenced resource is a newer version of this resource.

IsVariantOf

The referenced resource contains the content of this resource in a different form.

a label indicating whether the value refers to a concept or a thing

Concept

label indicating that the value refers to a concept (as in owl:Concept)

Thing

label indicating that the value refers to a named person, place, or thing (as in skos:Thing)

a URI that identifies the controlled vocabulary, registry, or identifier system that the value is defined in.

the unique identifier identifying the concept or thing

a short, display-able token that locally represents the concept or thing

• As a token, it is intended that applications can search for this value and find all resources that are talking about the same thing. Thus, regardless of whether the @id field is provided, all references to the same concept or thing should use the same tag value.

IRI for the JSON-LD data type. This should be org:Organization for each publisher

The plain-text name of the organization

A parent organizational entity

full name of the contact person, role, or organization

The email address of the resource contact

the contact mailing address

the contact telephone number

• Complete international dialing codes should be given, e.g. '+1-410-338-1234'

the time zone where the contact typically operates

a local identifier representing this person

• This identifier is expected to point to an up-to-date description of the person as known to the local system. The properties associated with that identifier may be different those given in the current record.

a label indicating the system that the identifier adheres to

• this label may imply a particular resolver to use

the value of the identifier

• if no scheme is provided, a URI form of the identifier should be given

the algorithm used to produce the checksum hash

the checksum value

a human-readable description of the format

a URI that identifies the format type registry or identification system that the value is defined in.

the unique identifier for the format

a short, display-able token or abbreviation for the format

• As a token, it is intended that applications can search for this value and find all files having the same format. Thus, regardless of whether the @id field is provided, all references to the same format should use the same tag value.

the class indicating that this is a Person

the author's full name in the preferred format

the author's given name

• Often referred to in English-speaking conventions as the first name

the author's family name

• Often referred to in English-speaking conventions as the last name

the author's middle names or initials

• Often referred to in English-speaking conventions as the first name

the author's ORCID

The institution the person was affiliated with at the time of publication

a local identifier representing this person

• This identifier is expected to point to an up-to-date description of the person as known to the local system. The properties associated with that identifier may be different those given in the current record.

sub-units of the main organization the that the person is a member of

• The order of the array elements should be treated as significant. Typically (though not required), each element will reflect a more specific unit contained in unit nameed in the previous element.