Data DOI Minting
Curtin University Library can mint digital object identifiers (DOIs) for data using the Australian National Data Service’s Cite My Data service.
Definition of DOI
A digital object identifier DOI is a character string or ‘digital identifier’ used to uniquely identify an object such as an electronic document. Metadata about the object is stored with the DOI name and may include a location or URL where the object can be found. The DOI for a document remains fixed over the lifetime of the document, whereas its location and other metadata may change. Referring to an online document by its DOI provides more stable linking than a URL, because if the document’s URL changes, the publisher need only update the DOI metadata to link to the new URL. A DOI takes the form of an alphanumeric string starting with ’10.’, such as 10.4225/06/553721D2E190D. You can resolve a DOI to the object it identifies by adding it to http://doi.org/, such as http://doi.org/10.4225/06/553721D2E190D. For more information about DOIs see the Wikipedia page.
Criteria for datasets
In order for Curtin University Library to mint a DOI, the dataset must fulfil these essential criteria:
- Affiliated with Curtin University
At least one of the dataset’s aggregators must be a staff member or student of Curtin University.
The Library will create a public metadata record in the process of minting a DOI. This metadata record will be published to the Research Data Australia (RDA) discovery portal. Every dataset aggregator including those not affiliated with Curtin, that requires credit in the metadata record must have an Open Researcher and Contributor ID (ORCID) freely available from ORCID’s website. An example of an RDA record for a Curtin dataset can be found here.
The dataset must be openly accessible on the web with an appropriate licence, such as Creative Commons. Alternatively, access to the dataset may be mediated by Curtin University or its representatives.
A DOI is forever, so the dataset must be made available without any time limitations. An archival copy of the dataset must be stored on centralised Curtin storage under the joint custodianship of Curtin Information Technology Services (CITS) and the Library. If a dataset is withdrawn, the DOI will be redirected to a placeholder landing page explaining that the dataset has been withdrawn.
The dataset must remain unchanged from when the DOI is minted. Small, administrative changes may be made such as a change in name or addition/removal of authors, but nothing that may invalidate the results of a publication citing the dataset. A new version of a dataset including the addition of more data, will require a new DOI.
In some instances, these criteria may not be appropriate. For example, extremely large datasets may need specialised storage. Any exceptions to the above criteria must be negotiated with the Library and approved by the Associate Director, Research and Collections Services.
Getting a DOI
To publish your data and have a DOI allocated to your dataset, please complete the form located here. If you need further information on the DOI minting process, please email firstname.lastname@example.org.