Modeling suggestions

[dgarijo]

(This is aimed at the next iteration of the specification, but I am adding some initial requirements here).
Credit: Arnaud Duyme. As DMP users at we thought about test examples/ use cases described below.

Source(PANOSC): https://openaireeu.sharepoint.com/:w:/r/sites/OSTrails/_layouts/15/Doc.aspx?sourcedoc=%7BCA692F41-B379-4C31-B404-6D60C9D29159%7D&file=PaNOSC%20-%20Architecture%20Feedback%20DMP-SKG-FAIR%20Tool.docx&action=default&mobileredirect=true

[MaudMed]

Here are some additional thoughts (Inria, French national pilot)

• test on software archiving: is the software archived in a trusted repository / an archive? (user has to provide the git URL where the code is located) "The source code is located on Github / Gitlab /etc. but is not archived on Software Heritage. => then instructions on how to archive software"
• test on software PID (very much related to the test above, I am not sure if we need two tests or just the one): does the software possess a PID? (user has to provide the git URL where the code is located) "The source code is located on Github / Gitlab /etc. but is not archived on Software Heritage and therefore possesses no persistent ID. => then instructions on how to archive software"
• test on software credits (very similar to the test on licence Renaud describes above): are software contributors credited? "This software exists, but no information about authors is available. Please add an AUTHORS.md (possibly accompanied by a CONTRIBUTORS.md) file at the root of your git repo listing all people involved in the creation of the source code."

We could also test the presence of a code meta file in the git repo, containing certain metadata (which ones do we want to test, the question is open).