The UK’s E-Science Programme has been hailed a success
by leading researchers at a BCS Thought Leadership Debate. But they
warned that momentum must be maintained to reap the long-term
benefits.
E-science refers to the research that is made possible when the
resources held on computers at dispersed locations are pooled via
high-speed networks. Launched five years ago, the E-Science
Programme has funded more than 100 projects to develop e-science
techniques for a diverse range of applications.
The programme’s work using grid computing has enabled scientists
to do their jobs more quickly and share more data, the debate
heard. For example, a computer can now perform a piece of work in
four hours that would have formed the subject of a PhD 10 years
ago.
E-science has also enabled the combination of technologies with
models. In the environmental science area, for example, e-science
was used in conjunction with a Met Office model for a global
warming project. Starting from different initial conditions,
simulations were run several thousand times. The results showed
that the range of temperature increases by the end of the century
could be significantly greater than earlier predicted.
E-science has also facilitated the validation of scientists’
methods and data. In one project, grid computing enabled a whole
community to interrogate oceanography observations. Thanks to so
many people looking at the data, the degree of confidence in it was
greatly improved.
The BCS debate heard that the programme had also brought
together more traditional scientists with computer scientists,
providing them with useful networking opportunities.
However, delegates generally felt the programme had not been
running for sufficient time. One delegate said the point of the
E-Science Programme was to change the way people think by using
large computers but he doubted whether the programme had been going
long enough to provoke a change in thinking.
Another participant said the programme could take 10 years to
feed through into mainstream science and 30 years to reach
commerce.
Going forward, the key objective is to sustain the momentum
created by the e-science programme, delegates agreed.
Some research councils have already embedded e-science in their
methods, while others are only starting to use it. They all need to
reach the same stage and continue promoting e-science, the debate
heard.
Research councils could also specify that data and code from
projects have to be made available when papers are published.
The BCS could also get involved to encourage research councils
to offer cross-discipline funding as many projects are crossing
boundaries. The society could also help by closing the boundaries
between the scientists and computer scientists, for example, by
setting up links with other scientific societies, such as chemists,
meteorologists or civil engineers. Communication between the groups
will be key to further progress, the debate concluded.
What is the E-Science Programme?
E-science refers to the large-scale scientific research that is
increasingly being carried out through distributed global
collaborations enabled by the internet. These collaborative
scientific enterprises typically require access to very large data
collections, major computing resources and high-performance
visualisation technology.
In November 2000, the government announced funding for the
E-Science Programme, with allocations to programmes within each of
the UK’s Research Councils. In addition, a core cross-council
programme was set up to develop generic IT systems and middleware
to enable e-science and form the basis for commercial e-business
software.