Most of the work in making sure all nodes are being harvested correctly should fall out of the schema.org work, but it might be useful to characterize the repos in the list above into three categories:
Defunct: repositories that never fully came online or never posted content (eg Cornell eBird)
Need fixing: repositories that still actively post content but it is not being harvested for some reason (eg Dryad)
Up to date: repositories whose content is up to date with the CN - even if they have not posted any new data in a long time (eg GOA)
Repositories that post content in our supported dialects (EML, ISO, DataCite, schema.org) should be priorities to be sure that harvests are up to date.
MN status is here: https://monitor.dataone.org/_production_history/
Most of the work in making sure all nodes are being harvested correctly should fall out of the schema.org work, but it might be useful to characterize the repos in the list above into three categories: