Final session #APE2021 4th Paradigm: Data-Intensive Scientific Discovery: >10 yrs later. Irina Sens @TIBHannover Chairs: Claudia Draxl @HumboldtUni; James Milne @ACSPublications & Chairman of @STM_Publishing Board; @KarelLuyben Chair of #EuropeanOpenScienceCloud Board

First, Claudia Draxl @HumboldtUni who has been active in applying FAIR data principles to materials research, especially big data driven materials research. She argues that negative results in one area might still inform research in another area, so should be published #APE2021

Claudia notes it's impossible to get everything from a project in a paper, there's more to be shared, e.g. unsuccessful synthesis recipes; candidate materials that were not selected for further work; also results that are considered intellectual property Claudia Draxl #APE2021

Particularly in high throughput experiments, there can be 1000s of materials tested, but only e.g. 3 are used for further research and published. This huge waste drove the forming of NOMAD: https://nomad-lab.eu/  which houses 100s of 1000s of calculations Claudia Draxl #APE2021

Now there is NFDI, the [German] National Research Data Infrastructure 'to systematically index, edit, interconnect and make available the valuable stock of data from science and research'. https://www.nfdi.de/en-gb  Claudia Draxl #APE2021

Next, James Milne, Chairman of @STM_Publishing Board on 'STM Research Data Year 2020 - A Review'. The key role of publishers in encouraging data sharing led STM to dedicate 2020 as the Research Data Year. James reviews how publishers have supported FAIR data. #APE2021

What we see though is that while there is good growth in numbers of journals with data policies from 29% to 52%, and papers with Data Availability Statements (DAS) from 7% to 15%, it's clear there is still much to do. https://www.stm-researchdata.org/  #APE2021 @STM_Publishing

Finally we hear from @KarelLuyben, President of #EuropeanOpenScienceCloud Association. 'To move how we do science forward, we need people with data skills, we need to amend how we reward researchers, and we need to amend the metrics by which we assess research.' #APE2021

#EOSC "will be a federation of existing services, not a database or repository" and needs to be inclusive for all stakeholders, and recognise that countries have different infrastructures. @KarelLuyben #APE2021

Question: what success would you report in 2031? @KarelLuyben 50% of research data is FAIR and researchers know what it is, might take another 10 years to reach 100%. James Milne - I would hope researchers engage in data sharing naturally, via a simple process that is not a chore

James Milne continues, "we have to keep it simple, from a rough calculation, if a data sharing step added 1 minute of work to our @ACSPublications submissions process, it would use up 2 years worth of research time across all of our submissions" #APE2021