Demo showing what RELIANCE project has achieved on Open Science, FAIR and EOSC

This demo shows what we have achieved on Open Science and FAIR. 

- Starting from OpenAIRE EXPLORE, we search for "Copernicus air quality" and find lots of resources, mostly publications and only 2 software. The reason is that to be "classified" as "Software", we have to add specific metadata when publishing.

-  The "Software" we found is a "EOSC Jupyter notebook" created by Simone Mantovani with a DOI and additional metadata so that OpenAIRE explore can "associate" it to a specific EOSC service, namely EGI Notebook. 

- When we click on "EOSC Service: EGI Notebook", we are re-directed directly to the service that has been used to generate the original scientific results we found in OpenAIRE explore.

- Any EOSC service needs to be requested and you have to plave an "order" to get access to it, where you may have to explain why you would like to access this EOSC service. To authenticate to any EOSC service, you can use for instance your ORCID identifier. if you do not have one, we suggest to register: this is very handy for EOSC services and you keep your ORCID identifier when you move from one institution to another (in addition, your institutional login may not work).

- You will get notified by email (check your SPAM folder!) when you got access to an EOSC service.

- We login to EGI notebook using ORCID identifier and upload (manually) the jupyter notebook we found in OpenAIRE (following the link e.g. from zenodo (https://doi.org/10.5281/zenodo.5554786)

- The Jupyter notebook uses CAMS European air quality analysis from Copernicus Atmosphere Monitoring Service. The input data is accessible through an external service called the ADAM platform (Advanced geospatial Data Management platform). It hosts datacubes (easy and fast access to large amount of data).

- We can re-execute the Jupyter notebook but more importatnly we can create derivative work. However, make sure you check the license of the original result you find in OpenAIRE explore: it needs to have a license that allows you to create derivative work. Also make sure the Jupyter notebook is well documented.

- We duplicate the Jupyter notebook and customize it. To bring the Open Science aspect from the beginning and not only when publishing the Jupyter Notebook, we need to use storage that can be shared. We use another service called "EGI datahub".

- As when collaboratively writing scientific papers, we agree on how to organize the data: we create an "input folder" (containing all the input datasets used in the Jupyter notebook), an "output" folder with all the outputs we generate  and a tool folder with the Jupyter notebook.

- The new analysis is very similar to the previous one but over a different geographical area (France). 

- Finally, we create a Research Object that aggrgate all the resources. We use another external service called RoHub  (Research Object Hub) and create and "executable Research Object" which we hope will be found, accessed and reused!