Dataspaces aim at enabling inter-organizational data exchange, emphasizing interoperability and data sovereignty of data assets. While current implementations focus on providing a foundational framework to enable secure, standards-based data sharing and sovereignty, they lack the robust query processing features needed to address emerging demands in distributed and federated data ecosystems. We present a vision for advancing dataspace technology by incorporating sophisticated query processing mechanisms and integrating features that ensure data sovereignty within traditional data management platforms such as data lakes.
The original idea of dataspaces as envisioned by Franklin
et al. [
In 2015, Fraunhofer in Germany started the Industrial
Dataspace (IDS) initiative [
Gaia-X has evolved from the IDS concept by extending
its focus on data sharing and sovereignty into a broader
framework that integrates cloud services, edge computing,
and data ecosystems through standardized frameworks and
governance mechanisms [
The priority for trust and data sovereignty is a significant
strength, it also imposes limitations on the ability to
support data processing across a distributed data ecosystem [
A lack of these features constrains the practical utility of dataspaces in scenarios where data-driven decision-making depends on seamless and secure integration of data in a distributed data ecosystem.
The evolution from data lakes to data meshes and data
fabrics reflects a significant transformation in how
organizations approach data management to address issues of
scalability, governance, and accessibility [
Data meshes emerged to solve these issues by decentralizing data governance. This paradigm treats data as a product, where responsibility for the quality, usability, and governance of data lies with domain-specific teams. This domain-driven ownership model ensures scalability while addressing the shortcomings of centralized approaches, such as those found in traditional data lakes.
In parallel, data fabrics focus on creating an interconnected layer that integrates metadata across disparate systems. By employing technologies such as AI, automation, and knowledge graphs, data fabrics enable seamless data discovery, improved lineage tracking, and enhanced integration across an organization’s diverse data landscape. This approach prioritizes metadata-driven governance and context-aware connectivity, providing a more dynamic and agile data ecosystem that supports advanced analytics and decision-making processes.
The concept of modern dataspaces aligns with these paradigms. Similar to the ‘data as a product’ philosophy in data meshes, dataspaces emphasize treating data assets as shared, governed resources designed for collaboration and interoperability. In both cases, the focus is on ensuring the quality, contextual relevance, and accessibility of data for specific stakeholders or applications. This shared emphasis underlines the mutual goal of enabling robust, domainaware data collaboration across organizational boundaries.
However, sharing data products in a dataspace does not
imply centralized governance or uniform data quality
control. Similar to data meshes, data governance should be
organized in a decentralized manner. Therefore, each participant
should manage their own policies through self-governing
data products. This approach aligns with the dataspace
architecture, where data owners define and enforce usage
policies for their data assets [
Additionally, dataspaces as well as data fabrics rely on
semantic models to support semantic interoperability. In data
fabrics, knowledge graphs serve as a foundational tool for
modeling relationships and enriching metadata, allowing
for enhanced data discovery, integration, and query
capabilities. Similarly, dataspaces employ semantic models to
achieve interoperability among heterogeneous data sources
and domains (e.g., the IDS information model [
The interplay between these paradigms suggests a path toward convergence, where dataspaces could incorporate the principles of both data meshes and data fabrics. By blending the domain-centric ownership and product-oriented data management of data meshes with the semantic and automation-driven integration of data fabrics, dataspaces could emerge as a comprehensive framework for addressing modern data challenges. This evolution reflects a growing recognition of the need for distributed, interoperable, and semantically enriched data ecosystems capable of supporting diverse organizational and cross-domain needs.
The challenge lies in finding the optimal balance between
unified semantic models for describing data assets and
decentralized governance. In many dataspace projects, we
have observed that a centralized approach to defining the
core information model significantly slows down the
bootstrapping process. A decentralized approach, as envisioned
in data meshes, could accelerate this process but comes with
the risk of diverging semantics. To mitigate these issues,
collaborative ontology engineering methodologies need to
be applied [
Enhancing data lakes with dataspace-inspired features
can bridge the gap between centralized data repositories
and the decentralized nature of dataspaces. Specifically,
integrating features for data sovereignty and advanced query
processing can yield transformative capabilities. By
incorporating mechanisms like usage policies [
On the other hand, enhancing dataspace frameworks,
such as the Eclipse Dataspace Components3, with more
sophisticated federated query processing for heterogeneous
datasets could ofer a better usability of dataspaces. Data
scientists require easy solutions for creating a Pandas data
frame over heterogeneous data: an API as provided in
Apache Spark, combined with Delta Sharing, and enriched
with usage policies, could facilitate a true sovereign data
science framework that integrates heterogeneous data
access, data integration, and machine learning. Although it
might be still challenging to merge all these features into
one platform, we can leverage large-language models to
support users in executing their tasks in such a platform
[
We advocate for a paradigm shift in dataspace technology by prioritizing advanced query processing and seamless integration with traditional data management platforms. Leveraging existing innovations, such as the Delta Sharing Protocol, and extending platforms like SEDAR, can help realize the vision of a unified, sovereignty-aware data management ecosystem. However, several research challenges must still be addressed to fully enable this vision.
Policy-aware query execution requires embedding data sovereignty rules directly into query execution plans. Queries should be executed in compliance with access restrictions, data-sharing agreements, and computational constraints defined by data owners. User-centric interfaces should allow non-expert users to interact efectively with the dataspace. Many existing dataspaces sufer from poor usability, limiting their adoption and accessibility. Furthermore, usage policies must be extended beyond basic access control to include restrictions on query processing itself. Finally, data quality management remains a significant challenge in many dataspaces. A decentralized data quality framework, incorporating objective and standardized quality metrics, could help assess and improve data reliability while allowing participants to retain autonomy over their data assets.
This work has been sponsored by the German Federal Ministry of Education and Research in the funding program “Forschung an Fachhochschulen”, project I2DACH (grant no. 13FH557KX0) and in the funding program “KIAnwendungshub Kunststofverpackungen – nachhaltige Kreislaufwirtschaft durch Künstliche Intelligenz”, project KIOptiPack (grant no. 033KI111).
AI Disclosure Statement During the preparation of this work, the author used ChatGPT 4o in order to improve writing style, check grammar, and spelling. After using this tool, the author reviewed and edited the content as needed and takes full responsibility for the content of the publication.