You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
At the CW18 workshop hackday we've created a prototype web app to scrape GitHub and GitLab repos for metadata, edit the metadata and produce CITATION.cff files from them. These can be saved in the git root. Also, a useful extension for that app would be to push directly back to Git**b.
Downstream, CFF files should always be converted to codemeta.json in an automated fashion to be able to leverage their exchange format strengths (will write up considerations and ideas early next week after hols).
As for GitHub willing to implement automatisms I think we'd have to identify people to talk to as there seems to be no obvious replacement for Arfon at GH. I have no direct contacts at GH myself.
In order for versions of software that have not been published by its authors to citable with full metadata, that metadata needs to be created and stored by the software authors. (See https://danielskatzblog.wordpress.com/2017/09/25/software-heritage-and-repository-metadata-a-software-citation-solution/)
The metadata could be stored in a codemeta.json file, or via the citation file format work.
In any case, the subject of this issue is how do we get authors to start doing this?
One method would be to create such a file automatically when a new repo is created. Would GitHub be willing to do this?
Or could this be combined with the standard README?
The text was updated successfully, but these errors were encountered: