Publishing code
Publishing code is done in the same way as publishing other research datasets:
- First, the researcher creates metadata about the code in the university's Converis data section (see the "How to describe research data in Converis" guide at the bottom of the page).
- A suitable licence is selected for the code (see "How do I select a suitable licence?" at the bottom of the page).
- The code to be published is packaged by the researcher in a .zip file and submitted for publication either via the Nextcloud sharing link to be filled in the Converis data or via email.
When publishing the code, two different options should be considered, or both if necessary:
- Publishing the generic descriptive metadata of the code repository (e.g. GitLab), or,
- publishing a snapshot of the code at a given time of release.
Publishing a snapshot of code in JYX
A DOI for the current version of the code can be obtained when the code is published in JYX by downloading a copy (snapshot) of the current version of the code, which is published in JYX along with the code description information. Once you have uploaded a copy of the current status of the project from GitLab, follow the instructions in the "Publishing research data in JYX" section linked at the bottom of this page.
If you anticipate the need for multiple versions of the code, it is possible to publish the metadata for the entire repository as a "parent repository" and publish individual snapshots of the code as sub-repositories of the repository.
If there is an expected need for multiple versions of the code, it is possible to publish the metadata of the whole repository as 'parent' data and to publish individual snapshots of the code as sub-dataset.
Publishing code in Zenodo
Code on GitHub can also be published in Zenodo via GitHub's automatic Zenodo connection. In Zenodo, a snapshot of the code is assigned a permanent identifier (DOI) according to FAIR principles. .