THE ESFRI ROADMAP AND ITS DEMANDS ON THE E-INFRASTRUCTURE

The ESFRI Roadmap marked a turning point in the evolution of European thinking on research facilities, providing a catalogue of such facilities with their characteristics. In parallel, the ESF (European Science Foundation) completed a questionnaire-based survey of research facilities. Finally, the ERF (European Research Facilities) consortium representing national facilities with international access was formed to parallel EIROForum (the European laboratories funded by international subscriptions). It is becoming increasingly clear that management of these facilities and management of the research process require extensive ICT: for research managers that is provided by CRIS (Current Research Information Systems) and to give researchers additionally access to facilities to control experiments with associated modelling and simulation and access to research datasets and software.


INTRODUCTION
Research facilities have long provided researchers with the ability to 'see further and deeper' and thus push back the frontiers of human knowledge.As the large research facilities become more expensive to construct and operate, there have been two effects: a) national governments have agreed to contribute to one or more facilities not in their own country but available to their researchers; b) national governments have agreed to allow access to their national research facilities by researchers from other countries.The extreme form of this is when a large number of national governments agree (usually via a treaty) to the joint construction and operation of an international facility open to all their researchers.

ESFRI
ESFRI, the European Strategy Forum on Research Infrastructures, is a strategic instrument to develop the scientific integration of Europe and to strengthen its international outreach.The competitive and open access to high quality research infrastructures supports and benchmarks the quality of the activities of European scientists and attracts the best researchers from around the world.
The mission of ESFRI is to support a coherent and strategy-led approach to policy-making on research infrastructures in Europe and to facilitate multilateral initiatives leading to the better use and development of research infrastructures, at the EU and international levels.
ESFRI's delegates are nominated by the research ministers of the member and associate countries and include a representative of the commission, working together to develop a joint vision and a common strategy.This strategy aims at overcoming the limits due to fragmentation of individual policies and provides Europe with the most up-todate research infrastructures, responding to the rapidly evolving science frontiers, advancing also the knowledgebased technologies and their extended use.
ESFRI defines research infrastructures as: "They are facilities, resources or services of a unique nature that have been identified by pan-European research communities to conduct top-level activities in all fields."There are essentially two classes or kinds of research infrastructure facilities considered: EiroForum: European facilities (i.e.funded as Europe) E.g.CERN, ILL, ESRF, EMBL, ESA….and ERF: National facilities with international access / participation e.g.DESY/XFEL, RAL/ISIS… ESFRI does not consider small, individual facilities, e.g., at a university or restricted to national use although any einfrastructure must take account also of these facilities.
A typical research infrastructure facility is a large instrument that provides a method (e.g., through neutrons, electrons, photons) of observing directly natural phenomena (e.g.telescopes) or experimenting with matter to understand their properties (e.g.structure of proteins in cells).
Increasingly, however, led by the thinking behind the UK e-Science initiative (Jeffery, 1999a), research requires: • Observation or experiment with observation; • modelling or simulation; and • comparison of the results.
Moreover, the research commonly is performed not by one individual in one place but by a team of individuals geographically distributed from each other and from the facility (although some representatives of the team may attend the facility to perform the experiments).This leads to the need for ICT support for: • cooperative working / intercommunication between researchers; • interactive distance management of experiments on facilities (this happens most obviously on spacecraft); • data collection, storage and preservation; • data analysis and interpretation; • production of scholarly publications and their distribution; • scholarly discussion; • innovation leading to patents and products; and • synthesis and summarisation leading to inputs to research policy and strategy.
Clearly there is here a research workflow analogous to that in production industry or commerce.An advantage of workflow is that information is gathered incrementally and previously collected information is not re-entered but reused thus gaining both efficiency and accuracy.

ESFRI ROADMAP
ESFRI was initiated following a conference in 2000 and subsequent declarations.It really got under way with the statement from the Competitiveness Council in 2004: "In the context of developing research infrastructures of European interest, the Council of the European Union welcomes the development of a strategic roadmap for Europe in the field of research infrastructures and the role of the European Strategy Forum on Research Infrastructures (ESFRI) in this context.This roadmap should describe the scientific needs for Research Infrastructures for the next 10-20 years, on the basis of a methodology recognised by all stakeholders, and take into account input from relevant inter-governmental research organisations as well as the industrial community.The Council stresses that this roadmap should identify vital new European research infrastructures of different size and scope, including medium-sized infrastructures in the fields of humanities and bio-informatics, such as electronic archiving systems for scientific publications and databases, covering all scientific areas as well as existing ones that need to be upgraded." The 2006 roadmap (ESFRI, 2006) covered existing facilities, policies, size, capabilities and planned facilities, policies, size, and capabilities (including upgrades).It essentially provided a catalogue of facilities existing or planned in near-time.
The 2008 (updated) roadmap (ESFRI, 2008) stressed research infrastructures in new areas such as energy, biological and medical science, and environmental science.However the roadmap also states (p11): "e-Infrastructures are critical to all projects in this roadmap.The e-infrastructure aspect of each project has therefore been assessed in much greater detail than before."

E-INFRASTRUCTURE
The e-Infrastructure can be envisaged at several levels as in the original GRIDs architecture (Jeffery, 1999a) namely: 1. a lower computation and data layer consisting of the network, computational facilities, data storage (and preservation) facilities, and detectors or instruments; 2. a middle information layer where the data is structured as information and there managed, analysed, and displayed; and 3. a higher knowledge layer where -either by humans or by computers using data mining techniques -the information is converted to knowledge (justified belief).
The knowledge from the upper layer can, of course, be used to control the lower layers: for example, due to deduction from information to knowledge indicating that the range of interest of wavelengths of electromagnetic radiation is between x and y.Then the detectors can be set (or tuned) to maximum reception in the x-y range.
The lower layer has similarities to the US-developed GRID concept (Foster & Kesselman, 1998).
Such an architecture meets the requirements of: data collection, data storage and distribution, data preservation (all including metadata), and all with associated capability for processing and all within constraints of trust, security, privacy, and rights of use.Indeed, the metadata (Jeffery, 2000) is the crucial component because it provides a way to inter-relate all the various data sources and ensure their appropriate management using a CRIS, particularly if it uses the EU recommendation to member states CERIF (Common European Research Information Format) (CERIF).
However, the architecture also supports the research infrastructure facilities by providing the capabilities for: experimental control (steering), associated simulation/modelling (understanding), comparison of model with experimental results (gaining insight), and advanced analysis / visualisation tools (aiding interpretation).Alongside the experimental research the e-infrastructure -through digital libraries -also provides access to scholarly publications and thus to the wealth of past and contemporaneous research.This describes the 'researcher's workbench.'However, the researcher also has management or administration tasks and -in the e-infrastructure -these should also be integrated.Typically, these include systems for project management, financial management, research proposals, authoring scholarly publications, etc.Using a CRIS to assist in driving this workbench is a very effective way of reducing the workload of the researcher because the CRIS contains much information that is requested time and time again from the other systems, so the CRIS can be used to produce this information in place of repeated input by the researcher.
In a complementary way, research managers -both in research institutions and in funding organisations -will wish to analyse research information in order to support strategic decision-making and management of research.This is usually based on output metrics from the research such as publication citations, license fees from patents, spin-out company turnover, invited presentations, work for the community (e.g., reviewing publications or research proposals), etc.A CERIF-CRIS contains exactly such information in a structured way so that valid statistical analyses (with appropriate graphic outputs as required) can be performed.