Storing and processing data has never offered more choice - formats, databases, cloud providers, APIs. The harder question sits one level up: once teams and partners multiply, how do you find, describe, and safely exchange data across them - not just inside your own systems, but across organizations that use different tools, follow different conventions, and answer to different legal and confidentiality constraints?
I call this the inter-organizational data sharing and exchange problem. My PhD (University of Bamberg, 2023) is an attempt to work it out from first principles - grounded in six years inside a multi-partner EU research consortium where it was a daily, unsolved reality, not a theoretical exercise.
Most approaches sit at one of two extremes. Platform-first solutions (such as the International Data Spaces project) bundle the answer into one technology stack - which tends toward vendor lock-in and makes interoperability between vendors harder, not easier. Standards-first approaches - the FAIR principles, W3C’s Data on the Web Best Practices, or architectural ideas like the data lake, data mesh, and data catalogue - point in the right direction but stay high-level and more prescriptive than constructive: strong on what, thin on how.
The Basin Network aims for the middle: a constructive architectural pattern with enough structure to build against, that also satisfies the major standards-first recommendations rather than competing with them.
The pattern treats sharing as a matter of describing and exchanging data, not moving or merging systems - you publish a surrogate that stands in for a data asset, and the asset itself stays where it is. It rests on two pieces:
The same machinery works inside one team, across teams, and across companies - the only thing that changes is who can see what.
I tested the pattern against three real domains: collaborative aerospace/automotive manufacturing, IoT in smart agriculture, and crowd-sourced mobility data.
Since this work was completed, the industry has converged on much of the same ground from a different direction - the Open Data Contract and Data Product Standards (ODCS/ODPS), DCAT, and the dataspace initiatives now driven by Catena-X and the EU Data Act. I find that convergence more validating than the original novelty: the field arrived, a few years later, at a similar shape. The enduring contribution of the thesis is less any single abstraction than the framing: describe and govern data across boundaries without integrating the systems beneath it.
The full thesis is published here.