"In the analysis of the incidence of influenza-like illness, the data are downloaded from the server every time the analysis code is executed, but there is no guarantee that the download URL remains valid, nor that the returned data will always remain the same. So it is preferable to first make a copy of the data, and then use that copy in the computational document. To ensure traceability of the data, the computational document must nevertheless contain the URL from which the data was obtained."