03 Apr 2019

PaNOSC: Sharing data to speed up science


Florian Gliksohn,
ELI-DC Associate Director

On January 15th 2019, the Photon and Neutron Open Science Cloud (PaNOSC) project officially kicked off. ELI is one of the six European research infrastructures that cooperate in this project, which is aimed at allowing universal and cross-disciplinary open access to data. PaNOSC coordinator Andy Götz from the European Synchrotron Radiation Facility (ESRF), and ELI-DC’s Associate Director Florian Gliksohn tell ELInes about the project, and the role ELI will play in it.


PaNOSC is a European project financed by the European Union’s Horizon2020 research and innovation program. It brings together six strategic ESFRI research infrastructures (RIs) – besides ESRF and ELI, the other beneficiaries are CERIC-ERIC, the European Spallation Source, European XFEL and Institut Laue-Langevin – and the e-infrastructures EGI and GÉANT. The ultimate aim of the project is to align the efforts of the existing and new photon and neutron sources to make FAIR (Findable, Accessible, Interoperable and Reusable) data a reality and to get ready for and contribute to the European Open Science Cloud, which is currently under definition.

Treat data better
Andy Götz: ‘There is a general movement in science to treat data better. Until about five years ago, users came to a facility, conducted their experiments, and left with their data, which were subsequently deleted at the research infrastructure itself. The user was responsible for the data afterwards. But individual research groups don’t have the resources to make data available in a useable way. Being one of the more mature European research infrastructures, at ESRF we have experimented with different methods for data management over the past years. This EU call was a golden opportunity to connect our efforts with those of other ESFRI RIs.’

Cultural change
PaNOSC first and foremost aims to achieve a cultural change, state both Götz and Florian Gliksohn. ‘In the old days, some scientists perceived data as their private property. But by making experimental data easy to find, evaluate and reuse, we can speed up science significantly. Moreover, when data obtained with expensive facilities like those of the ESFRI RIs is used more widely, that enhances the return on these massive investments.’

Challenges
The project will address different challenges, Götz explains. ‘We will have to find ways and means to archive hundreds of petabytes of data for several decades. This imposes both financial challenges in terms of cost and sustainability, and technical challenges in terms of interoperability and metadata. How can we annotate data in such a way that they can be used by researchers who were not involved in the initial experiments, or who work in a different field?’ But let’s first walk before we start running, he states. ‘The PaNOSC project runs until 2022. In this period, we want to define a new framework for FAIR data policies, and explore which services our photon and neutron sources need to provide to enable scientists to reuse the data in a meaningful way.’

Perfect timing
The timing of this project is spot-on for ELI, Gliksohn says. ‘The project’s lifespan coincides with the initial operation years of ELI. As a young facility, we can benefit from the experience previously gained by some of the more established partners in this project, like ESRF. ELI’s data policy, which will define the core principles of our data management practices and strategy, will be adopted shortly after ELI ERIC is established. PaNOSC provides much needed expertise for ELI in this context. By the end of the project, we aim at having the fundamentals of our data management standards and data services fully in place at ELI.’

Think out of the box
PaNOSC will trigger ELI to interact with its users to determine priorities in the field of data management and get an idea of what data-related services they need. ‘But we want to go beyond that,’ Gliksohn says. ‘It is like with so many innovations: it’s hard to realize what you need before it has been invented. Therefore, we are looking for people who can think out of the box, and are able to imagine possible use of data only few have thought of before. For the laser community, open data is a fairly new practice, since most laser labs in Europe are not big enough to bear the costs of Open Data. ELI has the critical mass to establish a more ambitious data management policy, and to develop an offer of advanced data services (remote data access, data analysis, simulation and computation, visualization). Our main goal with this project is to get the laser community to embrace the idea of open data, and to explore new opportunities and practices in this field.’ As a first step in the direction of a unified ELI data vision, ELI will organize a workshop later this spring addressing this topic.

Take the wheel
‘PaNOSC is a golden opportunity to help shape the European Open Science Cloud,’ Götz concludes. ‘We represent a large community: over 30,000 scientists visit our facilities and use our instruments to produce scientific data. We are at a tipping point in the history of science, in which both publications and data are changing to do justice to the open nature of science. This is a very interesting time to take the wheel and help shape scientific practice for decades to come.’


Group photo taken during the Kick-off meeting at Grenoble, France