Technology that supports collaborative science must be sound, secure, simple and compatible globally before academics will adopt it widely, says Judy Redfearn
E-Science requires an e-infrastructure - the collective term for networks, grids, data centres, collaborative environments and the software and organisations that tie them together neatly. An e-infrastructure can be used for activities such as e-learning, e-government, e-health and e-business as well as e-science and e-research. One of its greatest advantages is that it allows people to pool resources and work collaboratively, often on big projects across great distances.
The UK Government recognised its importance in the Investment Framework for Science and Innovation 2004-14, and the Office of Science and Innovation has taken the lead in drawing up a road map to chart its development. Many of the elements are in place, if still in fledgling form. They were brought to life over the past five years thanks to the UK e-Science Programme.
Now the Joint Information Systems Committee, guided by the OSI road map, is driving further development. The aim is to make the e-infrastructure so easy to use that researchers in all disciplines and others in academia and industry, even those with little computer experience, will use it to good end. Matthew Dovey, Jisc's director of research, says: "The UK e-Science Programme produced many proofs of concept for e-infrastructure and demonstrated its potential. The challenge now is to realise that promise and bring together the components in a way that is usable by a wide community."
The main elements include:
* SuperJANET, the high-bandwidth academic network that will soon be upgraded to SuperJANET5 and will incorporate UKLight, the switched optical network for research
* The National Grid Service
* The Open Middleware Infrastructure Institute UK, which re-engineers the proof-of-concept software that knits together an e-infrastructure and makes it easier to use
* The Digital Curation Centre, which is pioneering ways of preserving the vast wealth of data produced by science for the future
* The National Centre for Text Mining, which is providing tools and services to help researchers discover new facts and relationships within the published research literature
* Tools for collaboration.
In addition, individual large e-Science projects have established their own e-infrastructures, the largest being GridPP, the particle physics grid for analysing the vast amount of data expected from the Large Hadron Collider particle accelerator when it comes on line at Cern in Geneva early next year. By using middleware built on common standards, these projects will be compatible with the national e-infrastructure, which will in turn be compatible with international e-infrastructures such the Europe's Enabling Grids for E-Science initiative and America's TeraGrid.
An e-infrastructure will foster quicker or better research and new collaborations only if the different elements can be organised and interconnected. For example, appropriate tools and software can allow researchers to work in virtual research environments or facilitate the archiving, retrieval and processing of information in digital repositories.
Crucial to broadening uptake are security and ease of access and use. Users want simple sign-on procedures but also want to know that security is tight when they share their resources. Tools for collaborative working and better security for shared resources are two important areas for development under Jisc's three-year e-infrastructure programme. Other key areas involve increasing the uptake and ease of use of the NGS, the UK's core computing grid, and organising knowledge so that resources available through the e-infrastructure are easier to discover.
"There's a misconception that e-Science is just about large number-crunching. But we are seeing more work involving semantics and knowledge management," Dovey says. The real challenge will be to make the technology as invisible as possible to attract widespread use. "It will need to be as ubiquitous and accepted as e-mail is now before there is broad adoption within many communities. For that to happen, there will need to be social as well as technological change. People will need be able to rely on it before they come to accept it," Dovey says.
Back to ICT index page