If you are involved in science, IT, European politics or just have your ear to the ground, you may have heard about it by now: The European Open Science Cloud.
As a long term investment, doing science is worse than buying a car. Exact numbers vary, but it is possible that as much as 50% of published science can not be replicated. Most source data cannot be found after a couple of years.
To solve this problem, large funding schemes such as Horizon 2020 will soon demand a Data Management Plan as part of the research project as well as open access to and sharing of the data used. But without the necessary means to share data in a sensible, organised and properly structured fashion, such a requirement will only ad to scientist workload, without much added benefit to the community. The vision of a EOSC is a vision to provide such a structure.
Previously we have seen other infrastructure projects where the European Union has strived to build connections across borders where no single country could have done it themselves. I have previously written about one such project, of which I am not entirely un-skeptical. But whatever you think of these bridging efforts, they constitute the core of what a union should do: Build bridges over gaps rivers to wide for any single nation to cross on its own.
So what is the European Open Science Cloud? As the chair of the High Level Expert Group (HLEG) recently pointed out during the Combined CHARME – EMBnet and NETTAB 2016 Workshop in Rome in October:
“it is not European, its not open, it’s not science, and it’s definitely not a cloud”
It is not European
The aim of the EOSC is not to create an island of open data sharing only within Europe or only for European scientists. Science is global and so must the data be.
It is not Open
… as in free for all with no obligations, restrictions or limitations. Included in the project is the formulation of Rules of Engagement for access to the EOSC, provision of services and considerations of data security and privacy. The HLEG estimate that approximately 5% of a total research budget should be spent on managing and maintaining data and research infrastructure, making it possible for scientists to focus on the actual science.
It is not Science
The EOSC has nothing to do with how science is performed, and should have no say in who executes the science, as long as it is done in a responsible, ethical and financially sound fashion. It aims to construct an infrastructure for science, to make it possible for scientists to focus on their respective fields of expertise, instead of managing data and systems.
It is not a Cloud
Many of us use cloud solutions today as more or less structured dumping grounds for anything we are too lazy to back up on our own hard drive anymore. In terms of sharing, this can potentially make the situation resemble an internet without links, where everything exists but nothing can be found. Many times, due to restrictions in national data privacy laws, the system must also be capable of sufficient security. The EOSC strives to provide an infrastructure for data sharing and reuse, with both archiving capabilities and a functioning indexing system by integrating leading existing resources across Europe.
The usefulness of science is ultimately defined by its purpose, not its means and infrastructure is nothing but a means to an end. A new bridge can be walked across, but you can also continue to swim. Or build a boat of your own. But if you DO walk across, the purpose of that bridge will be determined by what you walk towards. That is up to you, not the bridge.
Thanks to Prof. Barend Mons for some of the background material.