I imagine a nice synergy between git-annex / DataLad and Onedata serving as a special remote (third-party data infrastructure) for them. I'm thinking especially about scenarios of collaborative data sharing across organizations. Imagine a couple of federated organizations with their heterogeneous storage systems, hosting datasets for research purposes. Collaborating users may use DataLad in this setup, but every user has to configure access to all different storage systems holding the data (picture 1). Onedata can virtualize access to all the sites in a single namespace and serve as a special remote, this way everyone would configure just one set of credentials and be able to share the data freely across the federation (picture 2).
I imagine a nice synergy between git-annex / DataLad and Onedata serving as a special remote (third-party data infrastructure) for them. I'm thinking especially about scenarios of collaborative data sharing across organizations. Imagine a couple of federated organizations with their heterogeneous storage systems, hosting datasets for research purposes. Collaborating users may use DataLad in this setup, but every user has to configure access to all different storage systems holding the data (picture 1). Onedata can virtualize access to all the sites in a single namespace and serve as a special remote, this way everyone would configure just one set of credentials and be able to share the data freely across the federation (picture 2).
I'll be happy to hear your thoughts and feedback.