To provide dataset PIs we should provide periodic reports (like "tax documents" by banks) but timing would vary across dandisets. So I feel we should add explicit metadata fields so dandiset would gain ability to instruct archive to produce such reports at some given interval.
Aspects/Concerns:
- alternative would be to implement that at the level of archive DB and not bother with metadata record
- although feels like "not for public eye data" I think there is nothing "private" about it, so IMHO ok to have in generic dandiset metadata record.
- what if multiple awards support specific dandiset. Which reporting period should be used? how "contributions" are split between awards?
WDYT @satra -- in dandischema or directly in the archive DB?
To provide dataset PIs we should provide periodic reports (like "tax documents" by banks) but timing would vary across dandisets. So I feel we should add explicit metadata fields so dandiset would gain ability to instruct archive to produce such reports at some given interval.
Aspects/Concerns:
WDYT @satra -- in dandischema or directly in the archive DB?