There are cases where the data set may be a database. Do we want live databases to be part of what is kept in the manifest? I think we do; even as an optional parameter. This will support the full lifecycle of people working with live datasets through the emerging JupyterHub/Quarry/Wikibase architecture.
There are cases where the data set may be a database. Do we want live databases to be part of what is kept in the manifest? I think we do; even as an optional parameter. This will support the full lifecycle of people working with live datasets through the emerging JupyterHub/Quarry/Wikibase architecture.