Open jon-bell opened 5 years ago
To what extent should reviewers try to extend an artifact (build on a tool, analyze a dataset in a new way, or provide new inputs/configurations to that tool or analysis), in comparison to simply running it?
I would consider running/verifying the artifact as the baseline and everything else nice to have, but not required.
To what extent should reviewers try to extend an artifact (build on a tool, analyze a dataset in a new way, or provide new inputs/configurations to that tool or analysis), in comparison to simply running it?