Project:MaRDITaskAreas

From MaRDI portal
Project:MaRDITaskAreas


MaRDI's
Task Areas
The structure behind the MaRDI project.

MaRDI's work is organized into seven task areas spanning mathematical research, infrastructure, outreach, and governance. Each area addresses a distinct set of community needs while contributing to MaRDI's shared mission of making mathematical research data FAIR — Findable, Accessible, Interoperable, and Reusable.

TA1: Computer Algebra

TA1 focuses on making mathematical data generated by computer algebra systems reusable across different software environments and over time. A central challenge is that data produced by one system today often cannot be read by another system in the future. TA1 works towards solving this through the mrdi File Format, a JSON-based, open file format designed to store and exchange data from computer algebra systems like Oscar, Macaulay2, Magma, and Sage. The long-term goal is workflows where computational results — Gröbner bases, polynomial ideals, group structures — can be cited and reused just as a theorem from the literature would be.

TA2: Scientific Computing

TA2 addresses interoperability in numerical and scientific computing, where workflows increasingly combine components from different programming languages and solver frameworks. Switching from one numerical solver to another — or coupling two independently developed codes — often requires significant redevelopment effort. TA2 develops tools like MaRDI Open Interfaces, which provides a common communication layer for scientific computing codes across C, Julia, and Python, and contributes to MaRDIFlow, a framework for designing and documenting FAIR computational workflows. The task area also works on the Portal/T1/MaRDI Packaging System (MaPS), which allows software environments to be packaged and deployed reproducibly with a single command.

TA3: Statistics and Machine Learning

TA3 works on FAIR research data practices in statistics and machine learning, where large collections of benchmark experiments have grown rapidly but often lack the metadata needed to compare results reliably. The task area maintains and develops OpenML, a community platform for sharing datasets, tasks, models, and experiment results, and contributes the mlr3 framework for unified, interoperable machine learning in R. TA3 also curates the Graphical Modelling and Causal Inference Zenodo community and its associated library of statistical notebooks.

TA4: Cooperation with Other Disciplines

TA4 bridges mathematics with other scientific disciplines, focusing on use cases where mathematical models are applied to domain-specific research problems — from computational biology to physics and engineering. The task area develops MathModDB, a knowledge graph of mathematical models that connects models to the research problems they address, the quantities they involve, and the computational tasks they support. TA4 also develops and maintains MaRDMO, MaRDI's data management planning tool, which guides researchers through structured documentation of mathematical workflows, models, and algorithms and links outputs directly to the MaRDI Knowledge Graph.

TA5: MaRDI Portal & MaRDI Knowledge Graph

TA5 is responsible for MaRDI's core digital infrastructure: the MaRDI Portal and the MaRDI Knowledge Graph. The Knowledge Graph integrates over 7 million items and 800 million relationships from sources including zbMATH Open, arXiv, swMATH, DLMF, CRAN, PolyDB, and OpenML, creating a semantically connected resource for mathematical research. TA5 also develops the MaRDI Knowledge Graph API, the MaRDI Knowledge Graph Query Service (SPARQL), the MaRDI Portal Storage (IPFS-based), the MediaWiki Math Search Extension, and the MediaWiki Math Rendering Extension — the latter now deployed across Wikipedia.

TA6: Data Culture and Community Integration

TA6 focuses on building a culture of good research data practice within the mathematical community and beyond. This includes the MaRDI Help Desk, MaRDI's first point of contact for support, consulting, and training; the Best Practices programme, which helps researchers and institutes make their mathematical work FAIR; and the MaRDI Station, a gamified, portable educational tool for research data management. TA6 also handles MaRDI's broader outreach, community events, and integration with library communities.

TA7: Governance and Consortium Management

TA7 provides the administrative and strategic backbone of the MaRDI project. Hosted at WIAS, the governance office supports the MaRDI Board, MaRDI Council, and Advisory Board, coordinates cross-task-area activities, manages the consortium's relationship with the broader NFDI landscape, and ensures that project milestones and deliverables are met. TA7 also coordinates MaRDI's participation in NFDI-wide structures and represents MaRDI in international collaborations with organizations such as SIAM, RDA, and the Australian Research Data Commons.