After getting the initial PD, the PD can be re-sampled and calculated, to get error bars around the PD values. This can be continually streamed to the client to improve the error bounds for as long as the connection is open.
GBIF diversity doesn't involve sampling so there are no error bars, but can still make use of streaming by returning points as they are obtained rather than in one go.
After getting the initial PD, the PD can be re-sampled and calculated, to get error bars around the PD values. This can be continually streamed to the client to improve the error bounds for as long as the connection is open.
GBIF diversity doesn't involve sampling so there are no error bars, but can still make use of streaming by returning points as they are obtained rather than in one go.