I'm a UX researcher for Dataverse and occasionally help manage Harvard Dataverse's OAI-PMH harvesting.
Yes, we can use Dataverse's OAI-PMH harvesting features to achieve the kind of bidirectional mirroring you described.
My colleagues and I created all of the repository's OAI-PMH harvesting sets you found after folks requested us to. We probably haven't created a set that contains information about the datasets that your user published in Harvard Dataverse. We can do that a number of ways, like telling Dataverse to create a set with information about all datasets in a particular Dataverse collection, if your user's datasets are and will be published in a collection, or telling Dataverse to create a set with information about datasets that have particular characteristics, like being authored by your user.
Or, instead of coordinating with my colleagues and I, you can tell your Dataverse repository to query DataCite to get information about datasets your user has published in Harvard Dataverse, if your Dataverse repository is running one of the more recent releases of the Dataverse software that includes this feature. That feature's described in our guides at
https://guides.dataverse.org/en/6.10.1/admin/harvestclients.html#harvesting-from-datacite.
If you'd like us to create a set for your user, I can email you directly to figure out how to do this. Or if you'd like help telling your Dataverse repository how to query DataCite to harvest dataset info from Harvard Dataverse, we can email directly about that, too.
Lastly, has your user also published datasets in your repository, or will they? And if so, should information about those datasets also show up in Harvard Dataverse?
I hope this helps!
Julian
Julian Gautier (he/him)
Product Research Specialist,
IQSS
