To maximise efficiency, utilisation and uptake of European e-infrastructures, new models of interaction with a diverse ecosystem of data and platforms must be established. The introduction or generation of huge and complex datasets makes the tasks of acquisition, transformation and analysis counterproductively hard. Instead of focusing on delivering end-user-facing solutions top-down, DARE puts teams of research developers and scientists, creating advanced data-powered methods, at the forefront of scientific progress and innovation and enables them to craft solutions required by their communities.
Empowering developers to make the most of what European e-infrastructures and recent technological advances offer, requires programmable methods and tools. Such methods need to encapsulate high-level data and computation concepts, to be reusable and verifiable. They need to respect the time, effort and methodologies of research developers and scientists – they need to be user-friendly and effective. At the forefront of innovation, they will use the most advanced technologies and software stacks. As these inevitably change rapidly, the encoding of advanced methods must be abstracted and automatically mapped to the most appropriate targets. They need to allow for the composition of arbitrary applications and solutions, making use of data and infrastructures as transparently as possible, i.e. without explicit naming and instance details. They need to be easily extensible, to adapt to the changing landscape provided by e-infrastructures and platforms and to the variety of scientific methodologies. They need to be discoverable, accessible and communicable, and encourage agile development and continuous integration. Last, they must be offered in computer languages developers have built expertise in.
However important the above technical requirements may be, they cannot be addressed in isolation from the requirements of the scientific and innovation communities. DARE will facilitate agile and continuous communication between types of user. It will interact with other entities, such as e-infrastructures and resource providers, to continuously improve cross-discipline understanding and, as a result, the platform, its methods, tools and ultimately the applications and services developed. The DARE project will itself be the testbed of this agile co-design, initially working with the EPOS and IS-ENES RIs.
The DARE concept can be summarised as follows: to effectively tap the huge data resources available via technological products and advances – user-facing applications, services or large experiment campaigns – one must empower those who develop methods and their encapsulations so that they can help their user communities. DARE delivers a unified context for their work in order to catalyse innovative data-driven science, making transparent use of e-infrastructures and other resources.