Have a way to add metadata that we generate to better describe tools and datasets

[cmadjar]

Purpose

Proposed implementation on the tool side of things:

• create a separate JSON file that links the zenodo ID of the tool to metadata generated by us (name of those fields to be determined)
• store that JSON mapping file in a separated GitHub repo that could be read by the CONP portal and by CBRAIN so that the extra metadata can be pulled on both platforms

[cmadjar]

@natacha-beck @glatard @tkkuehn @emmetaobrien Does that summarize the solution we talked about today?

[tkkuehn]

I think that covers the proposed implementation.

For the record, my understanding of the issue is:

• Some pipelines have descriptions that are shorter or less descriptive than we'd like, but we don't control those pipelines at the source (Zenodo), so we can't update them accordingly.
• We'd like to have a way to expand on these and other metadata fields to make them more usable without needing to change the source datasets.
• We'd like to propagate those annotations to CBRAIN so users have a uniform experience across both platforms.

(Some of that might not be right but I thought it was worth writing down)