DigitalNZ metadata dictionary

This metadata dictionary provides detailed information about the most common metadata fields used by the DigitalNZ API to describe items from content partner collections and make them easier to find, share, and use.

What metadata information does DigitalNZ collect?

Metadata is any information that describes digital content. When metadata is added to digital items in a standardised and consistent way (using prescribed metadata field names and schema), the items can be managed and organised so users are better able to discover, share, and use that content.

DigitalNZ’s field names and schema are based on Dublin Core but have been adapted to be more flexible and cater for a range of content and metadata levels.

Settings for metadata fields

Each field has settings that indicate whether the field is required, if it is searchable, and whether it can contain multiple values. These settings are listed for each field type.

Is the field required?

  • Compulsory - all records harvested into DigitalNZ must have these fields

  • Recommended - makes a record considerably more useful and should be provided wherever possible

  • Optional - additional fields available to capture any extra information

Is the field searchable?

  • Searchable - this field can be searched using the DigitalNZ search records API

  • Not searchable - this field cannot be searched

Can the field contain multiple values?

  • Single-valued - this field only has one value, for example, a DigitalNZ item can only have one thumbnail (dnz:thumbnail_url)

  • Multi-valued - this field can have many values, for example, a DigitalNZ item may be created by several different people (dc:creator)

Metadata field types

dnz : category

Compulsory | Searchable | Multi-value

This field is used for filtering, via tabs, on the DigitalNZ.org website. It is also used for the API and possibly other hosted/custom search integrations.

Category Definition Usage notes
Images Image is a primary/major component of resource. Items can also use other categories such as ‘Newspapers’ (eg images from Papers Past), ‘Articles’ (eg Kete), or ‘Books’ (eg for images from books at NZETC).
Videos Video is a primary/major component of resource. Items can also use other categories such as ‘Articles’ (eg a video included in a Te Papa article).
Audio Audio is a primary/major component of resource. Items can also be other categories such as ‘Articles’ (eg Te Ara).
Books Digitised books in various forms, including ephemeral booklets or born digital books. Maybe combined with ‘Images’ if the image exists as significant component of an individual item.
Newspapers Digitised newspapers. Maybe combined with ‘Images’ if the image exists as significant component of an individual item.
Manuscripts Digitised unpublished archival material such as letters, scrapbooks, diaries etc. Usually discrete collections so doesn’t need additional categories. Use dc_subject to add useful descriptors if not already supplied by content partner (eg letters, diaries).
Data Datasets or unprocessed information. Includes raw data, csv files, and downloadable sets of resources (eg map tiles/layers). Due to the diversity of this set there may be times where it is appropriate to combine, eg if maps tiles are available in a usable image format, or if data has been sourced in bulk from a newspaper collection.
Research papers Individual published outputs of research. Usually discrete collections so doesn’t need additional categories. Do check for ‘Videos’.
Articles Single items or groups (eg journals) of news, opinion, fact, or general information. Includes blogs. Not for use for newspaper articles, use ‘Newspaper’ instead. Examples include guides and reports. Can be combined with ‘Audio’, ‘Videos’, or ‘Image’ if there is enough contextual information to warrant it. However if the main focus of the item is the information and the image is only illustrative (ie secondary) do not apply the ‘Image’ category. Do utilise the thumbnail if possible.
Archives Other types of unpublished archival material not covered by ‘Manuscripts’ etc. Examples include Archives NZ NZDF Personnel Files.
Other Items that do not fit into existing categories. Examples include interactive web resources, slideshows, and any other outlying item types.
Stories DigitalNZ Stories created by users. Not strictly an item category, but needed.

dnz : collection

Compulsory | Searchable | Multi-value

The name of the collection(s) and/or website that the item belongs to. This field can also contain sub-collections or groupings within the main collection. This field is searchable but does not actually get displayed directly on DigitalNZ.org. A separate (single value) field named ‘display_collection’ containing the primary top level collection is used for displaying alongside the record and in the collection filter drop downs.

For example:

  • Papers Past

  • Grey River Argus

  • Te Ao Hou

dnz : content_partner

Compulsory | Searchable | Multi-value

The name of the organisation(s), institution(s), or individual(s) making content available through DigitalNZ. Usually the name of the organisation that has agreed to the DigitalNZ Metadata Contribution Terms.

For example:

  • Archives New Zealand Te Rua Mahara o te Kāwanatanga

  • Alexander Turnbull Library

dc : description

Compulsory | Searchable | Multi-value

Descriptive information about the record. Can be a summary, abstract, or any information about the record. In most cases this is limited to 350 characters, except where the provided descriptions are clear abstracts or summaries from trusted sources that are not too much longer than 350 characters.

For example:

  • ‘The drawing is from the book Why Cats Paint: a theory of feline aesthetics. Published by Heather Busch and Burton Silver …’

  • ‘During the late 1980s, Kiwi inventor John Britten developed and built a revolutionary racing motorcycle ...’

dnz : landing_url

Compulsory | Not searchable | Single-value

An HTTP URL resolving to a landing page for the content.

For example:

dnz : thumbnail_url

Compulsory | Searchable | Single-value

A URL resolving to a thumbnail of the content for display in search results list. Preferably around 250 px wide. If the image is larger than 255 px wide or 138 px high it will be resized on the DigitalNZ website.

For example:

dc : title

Compulsory | Searchable | Single-value

A name given to the resource or content object being described. If the titles of a series of resources are all identical then some extra information, like a date or identifier, can be added to the title so that the titles are unique.

For example:

  • US forces at Camp McKay store

  • [Looking east from Ponsonby over Auckland City to harbour,....]

  • The paleoaustral genus Protodolium Wilckens, 1922 (Mollusca: Gastropoda), and a new species from the Late Cretaceous of Chatham Islands, New Zealand

dnz : usage

Compulsory | Searchable | Multi-value

Human-understandable label defined by the DigitalNZ project team, ie ‘CC-BY-NC’ would have ‘Share’ and ‘Modify’; ‘No known copyright’ would have: ‘Share’, Modify’, and ‘Use commercially’. If the status of content is not identified by the content provider the value ‘Unknown’ is used.

For example:

  • A combination of ‘All rights reserved’, ‘Share’, ‘Modify’, ‘Use commercially’, ‘Unknown’

dc : creator

Recommended | Searchable | Multi-value

The name of the person, organisation, institution, or service who created the content, ie the photographer, artist, writer, or author. Often this will be the original copyright holder. Where the content is born digital, this may be the creator of the digital file. Where content has been digitised, it is usually the creator of the analogue source.

For example:

  • John Pine Snadden

dc : date

Recommended | Searchable | Multi-value

Stores date information to be used in the filters/facets. Because it is being used in the filters, this information is transformed during harvesting to a computer readable format. DigitalNZ isn’t able to currently deal with date ranges, so the work around is to take the first date of the date range. Also this date must be expressed down to a single day, so will default to 1st of the 1st if only a year is specified in the original source.

For example:

  • 2011-12-15T00:00:00.000Z

dnz : display_date

Recommended | Not searchable | Single-value

Stores date information in any format. This field does not have to be computer readable like dc:date so can store any text-based date expression.

For example:

  • circa 1960s

  • 1920s to 1930s

  • c1895

  • 15 Jan 1992

dnz : locations

Recommended | Not searchable | Single-value

This field contains geographical location information including latitude/longitude coordinates, text-based location info (ie city/region/street), and details about where the location information comes from (ie ‘Location provided by Museum of New Zealand Te Papa Tongarewa’).

For example:

  • <entry lat="-40.0733" lng="+176.8967" area="North Island, Hawkes Bay" specific_location="750 m South of Tuingara Point, Pourerere, Hawkes Bay" comment="Location provided by Museum of New Zealand Te Papa Tongarewa"/>

dnz : placename

Recommended | Searchable | Multi-value

Human-readable place name. Ideally from a controlled vocabulary such as the Getty Thesaurus of Geographic names or the New Zealand Gazetteer of Official Placenames.

For example:

  • Castlepoint, Wairarapa, New Zealand

dc : rights

Recommended | Searchable | Single-value

Rights information. Can be a rights statement explaining the rights of the record or a link to a webpage with more detailed rights information.

For example:

  • Crown Copyright 2009 administered through the Ministry for Culture and Heritage Te Manatū Taonga.

dnz : rights_url

Recommended | Not searchable | Multi-value

An HTTP URL resolving to a rights or terms of use page statement for the resource. Where records have a rights statement rather than a URL, this is recorded in dc:rights.

For example:

dc : subject

Recommended | Searchable | Multi-value

A repeatable field containing keywords about the record content. The harvester will typically split any comma (or similar character) separated lists into separate elements. A harvester may add useful subjects to records to aid findability and grouping if those terms don’t already exist in the data.

For example:

  • Clothing & dress

  • LB Theory and practice of education

  • Caring for sick

  • World War I

dc : type

Recommended | Not searchable | Multi-value

Terms describing general categories, functions, genres, or aggregation levels for content. This field is one of the main factors used to determine the TYPE search filter (such as memorabilia, magazine, people etc).

For example:

  • MovingImage

  • StillImage

  • Thesis

dc : contributor

Optional | Searchable | Multi-value

If the content comes from an aggregation service, this field can optionally provide more context as to the original organisation that provided the content.

For example:

  • School of Biological Sciences

  • Jordan, Bill

dc : coverage

Optional | Not searchable | Multi-value

The extent or scope of the content of the resource. Coverage can include spatial (geographical) location or temporal period (a period label, date, or date range). This is not the primary field for location or date information within DigitalNZ and is mainly used to match existing metadata field names.

DigitalNZ’s primary spatial/location field is dnz:location (which holds latitude and longitude coordinates if available) and dnz:placename for place names as text.

DigitalNZ’s primary date fields are dc:date (strict formated date) and dnz:display_date (more flexible text based date information).

For example:

  • Auckland Region (N.Z.)

  • 1980

dc : format

Optional | Searchable | Multi-value

Contains information about the physical attributes of the item or record. Often used to match an existing field name from the content source.

For example:

  • Watercolour/pencil

  • 112 x 112 mm

dc : identifier

Optional | Searchable | Multi-value

Keeps information like content provider record IDs, ISBNs, serial numbers etc.

For example:

  • KF5-803

dc : language

Optional | Searchable | Multi-value

Indicates the languages of the resource. Harvested as a language code like ‘en’ .

For example:

  • en

  • bo

dnz : object_url

Optional | Searchable | Single-value

A single HTTP URL resolving to the resource or content object itself, such as a PDF or image file. Where multiple versions exist, use the highest resolution version.

For example:

dc : publisher

Optional | Searchable | Multi-value

The name of the publisher of the digital object, or the original physical object if it has been digitised.

For example:

  • Taylor and Francis

dc : relation

Optional | Searchable | Multi-value

Information about a related resource. It could contain a URL to a related resource, or just be the title of a related collection.

For example:

  • K E Niven and Co :Commercial negatives

dc : source

Optional | Not searchable | Multi-value

Information about where the resource has come from, or a reference to another resource from which the content is derived, such as a book in the case of a digitised book.

For example:

  • Amer. fiction, 1774-1910 (microfilm, 1970-1978 ed.), v. 2 (1851-1875), reel R-19, no. 2144A

  • Archives footage preserved and made available by Archives New Zealand/Te Rua Mahara o te Kāwanatanga