Member Nodes

I have contacted the node operator Adam with an example 10.6067:XCV8GH9MV1_meta$v=1533165864386 in the record found at https://search.dataone.org/view/doi:10.6067:XCV8GH9MV1_meta$v=153316586438 to ask if there is a delay in activating new DOIs. Adam indicates that the example "10.6067:XCV8GH9MV1_meta$v=1533165864386 “ isn’t a valid DOI. It’s a DataONE identifier. the valid DOI for that record is: 10.6067/XCV8GH9MV1. For DataONE we changed the / to a : and added the suffix with the type (“metadata”) and the creation date (the $v=…)."

This suggests a flaw in the assumption that the appearance of a DOI in the PID will indicate that the PID is a DOI, and the assumption that the value of the PID will be equal to the value of the DOI as provided in the metadata record.

A second point: I did happen to also run a diff report finding 222 pids on the MN not on the CN. Due to that tDAR is a custom build node which does not retain immutable copies of metadata records, once a record is "versioned", all old versions are lost. There are sometimes hiccups in how their system translates in this way to DataONE reality. Have notified Adam of the attached pids and he'll take a look when he can.