Skip to Content
ublogo print

University at Buffalo Libraries

Digital Collections

Metadata in Omeka

Omeka uses Dublin Core metadata. While this is also true of CONTENTdm, the CMS we’ve been using for the last ten years, we’ve had more flexibility in Cdm because it allows one to customize field names and then map them to a Dublin Core field. Since we no longer have this ability, it’s important to consider where your metadata best fits in Dublin Core.


In addition to the standard Dublin Core Element Set, we are using the Dublin Core Extended plugin in Omeka, which allows for more detailed and comprehensive metadata.

The Digital Collections Team have a spreadsheet template that lists the fields as presented in Omeka. Contact Stacy Person (ssperson@buffalo.edu, 645-0774) for a copy of the spreadsheet. You can remove from the template any fields you will not use.

Guidelines for use of Basic Dublin Core for our digital collections can be found below.

Note that the appropriate authorities and controlled vocabularies should be used whenever possible.

A Note on Good Metadata

What is the "Dumb-Down" Principle? (from http://dublincore.org/resources/faq/)

The so called "Dumb-Down Principle" simply means that in any use of a qualified DC element, the qualifier may be dropped and the remaining value of the element should still be a term that is useful for discovery.

For example, there are several date qualifiers that might be used to enhance the precision of various dates associated with a resource. Dropping the date qualifier (say, for example, Date-Created) will still leave a useful date for discovery, though perhaps not quite as useful as if the qualifier were included.

Similarly, the specification of a subject term from LCSH, for example, is still useful even if one does not know it was selected from a controlled vocabulary.

The basic idea is that qualifiers should improve the precision of a piece of metadata, but the metadata should still be useful even without that extra precision (that is, dropping the qualifier has 'dumbed-down' the metadata).


Basic Dublin Core Element Notes
Title Title of the item. (Required. If you don't know the title, use [Untitled] or [Unknown], whichever is more appropriate.
Subject It is not necessary to specify from which authority the subject is derived. Proper names should use the name given in the LCNAF or, if unavailable, last name first.
Description
CreatorAn entity primarily responsible for making the [original] resource..
Source
Publisher In addition to the publisher of the original, this field should also contain the publisher of the digital asset (e.g., University at Buffalo. University Archives.)
Date

use the ISO-8601 format, YYYY-MM-DD:

  • YYYY is the year [all the digits, i.e. 2012]
  • MM is the month [01 (January) to 12 (December)]
  • DD is the day [01 to 31]

For approximate dates use "circa" (e.g., circa 1954); a range of dates can also be used (e.g., 1954-1962)

Contributor name of the person or department who supplied and is responsible for the digital collection
Rights Conditions for use either stated within the metadata, or a link to the appropriate webpage that explains conditions for use.
Relation
Format
Language
Type
Identifier Unique identifier for the item. This should be consistent throughout the collection. If a native identifier doesn't exist, the identifier of Collection ID plus incremental number should be used (e.g. LIB001_001, LIB001_002, etc.)
Coverage

The following fields are not part of the basic Dublin Core, but should be included in your metadata spreadsheet.

File Name [Identifier] Whenever possible, the file name should match the Identifier, with the appropriate file extension appended. THIS IS A REQUIRED FIELD.
IsPartOf The Collection Name and Collection ID should go into the IsPartOf field. The Collection ID is assigned by a member of the Digital Collections Team, who maintain a master list of the IDs in use in our digital collections.
Scripto StatusScripto is a plugin that allows for crowd-sourcing of transcriptions. The default setting for Scripto is "on," meaning that, unless specified otherwise, every file loaded into Omeka will have a link to "Transcribe this item." If you're uploading images that, by nature, don't require a transcription, please include a Scripto Status field in your metadata, filling it with either To transcribe or Not to transcribe.

If your metadata requires HTML (for such things as making a URL a hyperlink, formatting text as bold or italic, or making an email link clickable), the HTML markup must appear in the metadata.