describe/annotate our example INM-7 dataset with metadata
provide a simple-as-possible way for catalog contributors to submit metadata for their datasets
The goal is to use the existing tools that we have available, while leaving as much flexibility as we can for future changes to the tech stack at any level.
We have the following available methods for providing/extracting metadata:
tabby (dataset- and file-level, supports semantic fields, works with datalad-tabby and scripts in tabby-utils)
metalad_core (dataset- and file-level, works with datalad-metalad and scripts in sfb1451-projects-dataset)
studyminimeta (dataset-level, works with datalad-metalad and scripts in sfb1451-projects-dataset)
cff (datase-levelt, works with datalad-metalad and scripts in sfb1451-projects-dataset)
bids_dataset (datase-levelt, works with datalad-metalad+neuroimaging and scripts in sfb1451-projects-dataset)
datacite_gin (datase-levelt, works with datalad-metalad and scripts in sfb1451-projects-dataset)
Out of these options, tabby is the only one with a pipeline that can operate outside of a datalad dataset.
If we go with tabby as the starting format, we will likely need to use a specific convention, which could of course be as generic as we want it to be.
We want to be able to:
The goal is to use the existing tools that we have available, while leaving as much flexibility as we can for future changes to the tech stack at any level.
We have the following available methods for providing/extracting metadata:
datalad-tabby
and scripts intabby-utils
)datalad-metalad
and scripts insfb1451-projects-dataset
)datalad-metalad
and scripts insfb1451-projects-dataset
)datalad-metalad
and scripts insfb1451-projects-dataset
)datalad-metalad+neuroimaging
and scripts insfb1451-projects-dataset
)datalad-metalad
and scripts insfb1451-projects-dataset
)Out of these options, tabby is the only one with a pipeline that can operate outside of a datalad dataset.
If we go with tabby as the starting format, we will likely need to use a specific convention, which could of course be as generic as we want it to be.