Chi ha a cura la cura dei dati? Lezioni indirette dal Master in Data Management and Curation

Chi ha a cura la cura dei dati? Lezioni indirette dal Master in Data Management and Curation Mariarita de Luca, Federica Bazzocchi, Area Science Park Open Science Cafè, 13 Novembre 2025 10.5281/zenodo.17650417 Master in Data Management and Curation The Master in Data Management and Curation was launched in 2024–2025 as part of a pilot initiative by Area Science Park, CNR-IOM, and SISSA, supported by projects funded under Italy’s National Recovery and Resilience Plan (PNRR). Master in Data Management and Curation Pilot edition 2024-2025 had well defined and clear goals Training new experts in FAIR Data Management devoted to dissemination of FAIR principles, techniques and tools in the experimental laboratories of the PRP and NFFA-DI Research infrastructure Realizing FAIR-by-design pipelines in the laboratories of the involved Research Infrastructures Funded by the National Recovery and Resilience Plan (“PNRR”) within “Missione 4, Istruzione e Ricerca - Componente 2, Dalla ricerca all’impresa - Linea di investimento 3.1, Fondo per la realizzazione di un sistema integrato di infrastrutture di ricerca e innovazione”, with funds from the European Union –NextGenerationEU. Why do we need MDMC? 6 Data are growing faster than ever. The real challenge is not their quantity, but our capacity to manage them responsibly MDMC aims at preparing participants with the theory and practice to succeed in this mission, providing the skills to turn this challenge into opportunity Mission : To train professionals capable of transforming data challenges into value, innovation, and collective knowledge. DIGITAL INFRASTRUCTURES and DATA SPACES will provide the technological and policy framework, people provide the practice. We need a change of perspective from ‘FAIR data’ to ‘FAIR-by-design’ behaviors across labs, repositories, and services. We cannot have a FAIR ecosystem without FAIRskilled people. Education leads to the “human” infrastructure: when training becomes curricular, FAIR becomes culture Training the next generation of FAIR & Data professionals •Open Research Infrastructures •Data Center and HPC ORFEO •Laboratory of Data Engineering – LADE •Technology Park • Theoretical and Scientific Data Science Group •Pioneer in OA (hosted ArXive servers, OA journals) •Other successful Masters programs (MSC, MHPC) Our ecosystem combines infrastructures, research experience, innovation and a strong Open Science culture MASTER IN DATA MANAGEMENT AND CURATION MDMC STUDENTS EDITION 2025-2026 2 calls 37 applications 16 students selected 4 PhD, 8 Master, 2 Bachelor 7 countries Background: Genetics, Electronics, Physics, Data Science, Humanities, Psychology, Economics, Mathematics A black and white logo AI-generated content may be incorrect. Why so much stress on FAIR-by-design? •analyzing the content and structure •defining and choosing a semantic model, •enriching the recoverable metadata • transforming it into an open and linkable format •assigning licenses. FAIRification GO FAIR framework 0 20 40 60 80 100 120 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 time spent Produce the data Images generated by AI LEGO® Data Classification Games for FAIR Data Training https://doi.org/10.5281/zenod o.17183871 •Planning the structure •Planning the ingestion •defining and choosing a semantic model, •Planning how to collect metadata • how to transform data into an open and linkable format •choosing licenses FAIR-by-design Implement the pipeline Images generated by AI 0 20 40 60 80 100 120 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 time spent Great strategy for well managed and curated FAIR data may lead to great achivements!!! BEYOND FUNDING: BUILDING A CULTURE OF CARE Starting from the interest of institutions and research funders in supporting the MDMC edition 2025-2026, and from the growing interest of students in internships within companies, we began to reflect on the real value of data care. From investment to engagement, a wider question arose: what does it really mean to care for data? •For some, it is a strategic or economic investment — for others, a way to preserve unique and irreproducible knowledge, or to enable collaboration in large scientific communities. •Each perspective reveals a different facet of the same idea: data care creates value — but in many different forms. Who and why they care? EUROPEAN COMMISSION EOSC FEDERATION EU NODE COST OF NOT HAVING FAIR DATA European Commission, Directorate-General for Research and Innovation, Cost-benefit analysis for FAIR research data – Cost of not having FAIR research data, Publications Office, 2018, https://data.europa.eu/doi/10.2777/02999 Private Companies •AtroCore –Confronto strumenti MDM ,Osservatori Digital Innovation –Politecnico di Milano •DataCalculus –Raccolta dati ad alto volume Company Type (employee number) Annual Data Volume Annual Cost (K€) Technologies Example Estimated Human Resources needed Estimated Annual Cost/HR (K€) Estimated Total Annual Cost (K€) Italian Revenue Range for Service Comapny( K€) Data Manage ment Impact( %) Micro (<10 ) Low (<100 GB) 3– 8 AtroCore, Pimcore, open - source solutions 1 35 40,5 500 -2000 3,2% Micro (<10 ) Medium (100 GB – 1 TB) 8– 20 Boomi MDM, cloud base, BI tools 1 35 44 500 -2000 3,5% Micro (<10 ) High (>1 TB) 20 -50 Data Lake, Hadoop, Spark 12 35 87,5 500 -2000 7,0% Small (<50) Low (<100 GB) 520 AtroCore , Pimcore, opensource solutions 12 35 65 2000 - 10.000 1,1% Small (<50) Medium (100 GB – 1 TB) 20 -100 Boomi MDM, cloud base, BI tools 12 40 120 2000 - 10.000 2,0% Small (<50) High (>1 TB) 100 -500 Data Lake, Hadoop, Spark 12 50 675 2000 - 10.000 11,3% Medium (<250) Low (<100 GB) 10 -30 Semarchy xDM, Precisely, ibrid solutions 35 40 180 10.000 - 50.000 0,6% Medium (<250) Medium (100 GB – 1 TB) 30 -150 BI avanzata, storage cloud, backup 35 50 290 10.000 - 50.000 1,0% Medium (<250) High (>1 TB) 150 -500 Big Data platforms, compliance GDPR 35 50 525 10.000 - 50.000 1,8% The word is only one …. ENGAGEMENT (and responsibility) Passion drives better science When students learn the principles of Open Science and Data Curation, they don’t just acquire new skills, they embrace a new way of doing science. They feel engaged, motivated, and aware that by taking care of data, they are also taking care of the integrity and future of research. And that passion makes them better scientists. Image generated by AI STRATEGIES LEARNT FROM THE MASTER ➢FAIR-by-design mindset: integrating data care from the very beginning. ➢Interdisciplinary language: learning to communicate across technical, legal, and ethical domains. ➢Hands-on learning: through real-world internships and applied problemsolving. ➢Community approach: learning together, sharing experiences, building trust. ➢Data as a shared value: understanding that caring for data means caring for science itself. STRATEGIES THAT COULD MAKE THE DIFFERENCE ✓Recognise and reward data care: include stewardship efforts in research assessment. ✓Invest in professional roles: support for data stewards, engineers, and curators. ✓Encourage community engagement: foster active participation in open science networks. ✓Highlight success stories and champions: make best practices visible and inspiring. ✓Strengthen the bridge with institutions and companies: align scientific and innovation value. ✓Teach data care early: embed Open Science and FAIR principles in every stage of education. ✓Reward collaboration and openness: celebrate shared processes, not only results. [email protected] [email protected] OUR CONTACTS ✓Website: MDMC - SISSA ✓Available OPEN material: MDMC - Zenodo ✓Git repository: MDMC – GitHub ✓Email : [email protected] MDMC INFORMATION and LINKS