Workflows
Submission workflow
The workflow for archiving data objects to the CLARIN-DK-UCPH Repository is:
- Depositors submit their data to repository using the defined submission workflow.
- The curator receives a notification via e-mail and performs an assessment of the submission; if necessary, the curator will contact the depositor and request re-submission. The curator works closely with the depositor in the case of insufficient data or metadata quality, and/or licensing issues.
- The submission is approved, published and receives a new persistent identifier (PID).
- The archived submission is retrievable via the repository’s interface and metadata is available for harvesting via the CLARIN Virtual Language Observatory (CLARIN VLO) through OAI-PMH harvesting.
The workflow is valid for all the 4 data types that the repository accepts: corpus, languageDescription, lexicalConceptualResource or toolService.
Curation workflow
There are defined curation rules on how to proceed and deal with deposit submissions:
- The curators first checks if the deposited data seems to fall into the one of the four categories using the information in the metadata. The repository only accepts language-based data falling under the following four data types: corpus, languageDescription, lexicalConceptualResource or toolService.
- The curator then checks the content and validity of the metadata.
- The curator checks the validity of the content data by checking whether the data formats are acceptable, whether the files are properly described and whether there is chosen a proper license.
- The repository curator will review if there are traces of sensitive data by reading the metadata provided with submitted digital objects. In regards to submitted audio and video data, the curator will further check whether consent has been agreed i.e. via consent forms. The repository does not store research data that is classified as sensitive, including personal data, without consent forms.
- The curator then conclude if metadata has to be updated or extended, and if the data are accepted as language-based data falling under the following four data types.
- Concerning the acceptance of very large data collections or data outside the scope of the supported data types, a negotiation will occur between the repository manager and the depositor, see also What submissions do we accept?
Withdrawal request workflow
The workflow for withdrawal from the repository is:
- Users can request data to be withdrawn from the repository by contacting the Help Desk or by sending a mail to info@clarin.dk stating the reason for withdrawal.
- To qualify if there is a reasonable causes for withdrawal, the curator contacts the depositor if possible to clarify the doubts about the validity of the data or attached license information.
- If the depositor cannot or will not communicate with the repository curator, other sources for information to clarify the validity of the withdrawal request are addressed.
- The curator decides together with the repository manager about withdrawal, and logs the decision.
- The curator sends documentation of the decision to the user that requested the withdrawal.
- The national coordinator for CLARIN-DK is informed about the decision.