Building a UK Foundation for the Transformative

Enhancement of Research and Innovation

Panel MembersDaniel Atkins, Chair

Nathan BindoffChristine Borgman

Mark EllismanStuart Feldman

Ian FosterAlbert Heck

Dieter HeermannJulia Lane

Luciano MilanesiJayanth Paraki

Wolfgang von RüdenAlexander Szalay

Paul TackleyHan Wensink

Anders Ynnerman

Report of the International Panel for the 2009 Review of the UK Research Councils

e-Science ProgrammeD. E. Atkins, Chairatkins@umich.edu

Town Hall Presentation, February 9, 2009, London

International Panel for 2009 Review of the RCUK e-Science Program

(left to right)Anders Ynnerman

Linköping University, Sweden

Paul TackleyETH Zürich, Switzerland

Albert HeckUtrecht University, Netherlands

Dieter HeermannU. of Heidelberg, Germany

Ian FosterANL & U. of Chicago, USA

Mark EllismanU. of California, San Diego, USA

Wolfgang von RüdenCERN, Switzerland

Christine BorgmanU. of California, Los Angeles, USA

Daniel AtkinsUniversity of Michigan, USA

Alexander SzalayJohns Hopkins University, USA

Julia LaneUS National Science Foundation

Nathan BindoffUniversity of Tasmania, Australia

Stuart FeldmanGoogle, USA

Han WensinkARGOSS, The Netherlands

Jayanth ParakiOmega Associates, India

Luciano MilanesiNational Research Council, Italy

Text

Please Note:

Send any comments or corrections you would like considered for the final version of the report to

atkins@umich.eduNOT LATER THAN February 13, 2010

The slides for this presentation will be made available in pdf on the EPSRC website. If you want Mac Keynote of MS Powerpoint versions contact me via email.

Background and Context

What we are talking about today is very

important.

Original UK Definition of e-science bySir John Taylor

Definition should likely now be made more agnostic wrt

technology

Sir John TaylorProf. Dan Atkins

D. E. Atkins • University of Michigan • atkins@umich.edu

Cyberinfrastructure Genealogy & MovementCollaboratories

E-science

CyberscienceIT & Future of Higher Education

GRIDS

2nd Editionwww.mkp.com/grid2

Digital Libraries

PACI

ITR

KDI

HPCC

http://www.nsf.gov/oci

NSF Blue Ribbon Advisory Panel on CyberinfrastructureDaniel E. Atkins, Chair

University of Michigan Kelvin K. Droegemeier

University of Oklahoma Stuart I. Feldman

IBM Hector Garcia-Molina

Stanford University Michael L. Klein

University of Pennsylvania David G. Messerschmitt

University of California at Berkeley

Paul MessinaCalifornia Institute of Technology

Jeremiah P. OstrikerPrinceton University

Margaret H. WrightNew York University

“a new age has dawned in scientific and engineering research, pushed by continuing progress in computing, information, and communication technology, and pulled by the expanding complexity, scope, and scale of today’s challenges. The capacity of this technology has crossed thresholds that now make possible a comprehensive “cyberinfrastructure” on which to build new types of scientific and engineering knowledge environments and organizations and to pursue research in new ways and with increased efficacy.”

e-science (research enabled bye-infrastructure/ICT) is

increasingly essential for meeting 21st century challenges in

scientific discovery and learning

Time Scale (seconds)

10-15

10-9 10-6 10-3100 103

10-12 109

Num

ber

Sca

le (

over

siz

e sc

ale

from

Angst

rom

s to

Km

)A

tom

s B

iop

oly

mer

s

Geologic &EvolutionaryTimescales

106

Org

anis

ms

Ab initioQuantum Chemistry

First PrinciplesMolecular Dynamics

Empirical force fieldMolecular Dynamics

EnzymeMechanisms

ProteinFolding

Homology-basedProtein modeling

EvolutionaryProcessesEcosystems

andEpidemiology

Cell signallingCel

ls

100

103

106

100

103

106

100

103

106

100

103

106

Organ function

DNAreplication

Finite elementmodels

Electrostaticcontinuum models

Discrete Automatamodels

The Complexity of Biosystems

The inherent complexity,multi-scale, and multi-science

nature of today’s frontier science challenges.

The accompanying requirement for multi-disciplinary, multi-investigator,

multi-institutional approach(often international in scope).

And the need to engage more students in high quality, authentic,

passion-building science and engineering education.

The increased scale and value of data and demand for semantic federation, active curation and

long-term preservation of access.

The high data intensity and heterogeneity from simulations, digital instruments, sensor nets,

and observatories.

!!"#$%%&'($)'*+

!,-#'+.*#/*01(,/

2*330+')4#(1,$),5#

1'(6#3,5'$#+$11$),5#)*01/

7$18,#)'&,5#'3$8,#5$)$/,)/

!"##$%&'()%

*%$%+),'%

-.(/01(234(5

*%$%+),'%

&'678%19/:1(1%0&'()%

*%$%+),'%

&$,(/

;6<67($

&=5

&"1>%5

;6<67($?@($,A(1B$$?&=5&"1>%5C

D617"($?E(17.

&"1>%5+

FB&B@$(/%7(159A(<%?

&"1>%5+C

G6=6'%06(

@"#$6+.%0?'('%1+

B+71,'.5+6)+

;(7(#(+%?

!"#$%&

'%()*+,+-.

'&/)/0/(1(

H7.%1+?C

@1,0")%0?#5?(+71,/,A%1+I?(+71,'.5+6)6+7+?J?%0")(7,1+?

:1,AK?!(1>(10I?&'()%?*%$%+),'%I?-.(/01(I?&'678%1I?B0$%1I?

FHBHI?!(50%/I?L,./+?!,'=6/+I?-($?*%).?%7?($M?

!,-#'3$8,#/*01(,/

N"$76'$%?0(7(?+,"1)%+O5?,#P%)7I?Q(>%$%/<7.I??:,1A(7I

%7)M/%%0%0

Learning & Work Force Needs & Opportunities

Virtual Environments

Data & Visualization/

Interaction

High Performance Computing

e-science

CI-enabled science

Note: Many other reports on discipline-specific visions of and drivers for e-Science are available at www.nsf.gov/oci

•••

a new vision for Science

Global challenges with high societal impact

Data deluge… wet-labs versus virtual-labs

Improved scientific process

Cross-disciplinarity

Virtual Research Communities

networkinggridsinstrumentationcomputingdata curation…

Te

chn

olo

gy

pu

sh

value added of distributed

collaborative research (virtual

communities)

Ap

plica

tion

pu

ll

revolution

in science &

engineering,

research &

education

•••

building Science through collaboration

Research Communities

common goals, complementary and shared information, tools and knowledge, awareness of research protocols, effective means of collaboration, interest in being part of the community

Virtual research

from empirical, experimental, theoretical and computational science… to intensive use of data… abstraction… models… simulation… e-Science

Virtual communities

no geographical, time or institutional boundaries

Globalisation

Global challenges, global dimension, win-win situation

•••

ICT for Science: e-Infrastructures

Linking at the speed of the light

Sharing computers, instruments and applications

Sharing and federating scientific data

. . . . . . .

Astrophysics

community

Connecting the finest mindsSharing and federating the best scientific resources

Building global virtual communities

Biomedics

community

WeatherForecast

community

The Fourth Paradigm:Data-Intensive Scientific Discovery

http://research.microsoft.com/en-us/collaboration/fourthparadigm/

1 JANUARY 2010 VOL 327 SCIENCE www.sciencemag.org 36

POLICYFORUM

The mission of the National Institutes

of Health (NIH) is science in pursuit

of fundamental knowledge about the

nature and behavior of living systems and

the application of that knowledge to extend

healthy life and to reduce the burdens of ill-

ness and disability. The power of the molec-

ular approach to health and disease has

steadily gained momentum over the past

several decades and is now poised to cata-

lyze a revolution in medicine. The founda-

tion of success in biomedical research has

always been, and no doubt will continue to

be, the creative insights of individual inves-

tigators. But increasingly those investiga-

tors are working in teams, accelerated by

interdisciplinary approaches and empow-

ered by open access to tools, databases, and

technologies, so a careful balance is needed

between investigator-initiated projects and

large-scale community resource programs.

For both individual and large-scale efforts,

it is appropriate to identify areas of particu-

lar promise. Here are fi ve such areas that are

ripe for major advances that could reap sub-

stantial downstream benefi ts.

High-Throughput Technologies

In the past, most biomedical basic science

projects required investigators to limit their

scope to a single aspect of cell biology or

physiology. The revolution now sweep-

ing the fi eld is the ability to be comprehen-

sive—for example, to defi ne all of the genes

of the human or a model organism, all of the

human proteins and their structures, all of the

common variations in the genome, all of the

major pathways for signal transduction in the

cell, all of the patterns of gene expression in

the brain, all of the steps involved in early

development, or all of the components of

the immune system. Further development of

technologies in areas such as DNA sequenc-

ing, imaging, nanotechnology, proteomics,

metabolomics, small-molecule screening,

and RNA interference are ripe for aggressive

investment. Furthermore, these technologies

will spur the production of massive and com-

plex data sets and will require major invest-

ments in computational biology.

As one example, the Cancer Genome

Atlas ( 1) is now poised to derive comprehen-

sive information about the genetic underpin-

nings of 20 major tumor types. This infor-

mation will likely force a complete revi-

sion of diagnostic categories in cancer and

will usher in an era where abnormal path-

ways in specific tumors will be matched

with the known targets of existing therapeu-

tics. Another example is the opportunity to

understand how interactions between our-

selves and the microbes that live on us and in

us (the “microbiome”) can infl uence health

and disease ( 2).

Translational Medicine

Critics have complained in the past that NIH

is too slow to translate basic discoveries into

new diagnostic and treatment advances in the

clinic. Some of that criticism may have been

deserved, but often the pathway from molec-

ular insight to therapeutic benefi t was just not

discernible. For many disorders, that is now

changing. Three major factors have contrib-

uted to this: (i) the discovery of the fundamen-

tal basis of hundreds of diseases has advanced

dramatically; (ii) with support from the NIH

Roadmap, academic investigators supported

by NIH now have access to resources to

enable them to convert fundamental observa-

tions into assays that can be used to screen

hundreds of thousands of candidates for drug

development; (iii) public-private partnerships

are being more widely embraced in the drug-

development pipeline to enable biotech and

pharmaceutical companies to pick up prom-

ising compounds that have been effectively

“de-risked” by academic investigators and to

bring them to clinical trials and U.S. Food and

Drug Administration (FDA) approval.

As one example, the NIH Therapeutics for

Rare and Neglected Diseases (TRND) ( 3) pro-

gram will allow certain promising compounds

to be taken through the preclinical phase by

NIH, in an open environment where the world’s

experts on the disease can be involved. Fur-

thermore, as information about common dis-

eases increases, many are being resolved into

distinct molecular subsets, and so the TRND

model will be even more widely applicable.

The fi rst human protocol (for spinal cord

injury) involving human embryonic stem

cells (hESCs) was approved by the FDA in

2009, and the opening up of federal sup-

port for hESC research will bring many

investigators into this field. The capabil-

ity of transforming human skin fi broblasts

and other cells into induced pluripotent stem

cells (iPSCs) opens up a powerful strategy

for thera peutic replacement of damaged or

abnormal tissues without the risk of trans-

plant rejection ( 4– 6). Although much work

remains to be done to investigate possible

risks, the iPSC approach stands as one of the

most breathtaking advances of the last sev-

eral years, and every effort should be made

to pursue the basic and therapeutic implica-

tions with maximum speed.

Benefi ting Health Care Reform

U.S. expenditures on health care now rep-

resent 17% of our Gross Domestic Product,

are continuing to grow, and are excessive as

a percentage of per capita gross income com-

Opportunities for Research and NIHRESEARCH AGENDA

Francis S. Collins

The promise of fundamental advances in

diagnosis, prevention, and treatment of

disease has never been greater.

CR

ED

ITS: (L

EF

T T

O R

IGH

T) O

DO

M A

ND

O’H

ALLO

RA

N/N

OR

TH

WE

ST

ER

N U

NIV

ER

SIT

Y; LIN

DA

BA

RT

LE

TT

/NC

I, N

IH; B

ILL B

RA

NSO

N/N

CI,N

IH; N

ASA

; B

ILL B

RA

NSO

N/N

IDC

R, N

IH

National Institutes of Health, Bethesda, MD 20892, USA.

E-mail: collinsf@mail.nih.gov

Published by AAAS

on

Janu

ary

7, 2

010

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fro

m

Opportunities for Research and NIHby Francis S. Collins, DirectorThe power of the molecular approach to health and disease has steadily gained momentum over the past several decades and is now poised to catalyze a revolution in medicine. The foundation of success in biomedical research has always been, and no doubt will continue to be, the creative insights of individual

investigators. But increasingly those investigators are working in teams, accelerated by interdisciplinary approaches and empowered by open access to tools, databases, and technologies, so a careful balance is needed between investigator-initiated projects and large-scale community resource programs. For both individual and large-scale efforts, it is appropriate to identify areas of particular promise. Here are five such areas that are ripe for major advances that could reap substantial downstream benefits.

1 Jan 2010, vol. 326 SCIENCE www.sciencemag.org

Interdisciplinary, team-based e-Science

Five Special Opportunities for Biomedical Research by Francis S. Collins, Director US NIH

High-throughput Technologies for comprehensive studies

The revolution now sweeping the field is the ability to be comprehensive—for example, to define all of the genes of the human or a model organism, all of the human proteins and their structures, all of the common variations in the genome, all of the major pathways for signal transduction in the cell, all of the patterns of gene expression in the brain, all of the steps involved in early development, or all of the components of the immune system.

Translational MedicineMaking the pathway from molecular insight to therapeutic benefit more discernible

Focusing More on Global HealthIt is also critical to go beyond the focus on the “big three” diseases to neglected tropical diseases of low-income countries that contribute to staggering levels of morbidity and mortality

Benefiting Health Care Reform

Comparative effectiveness researchPrevention and personalized medicineHealth disparities researchPharmacogenomicsHealth research economics

Reinvigorating and Empowering the Biomedical Research Community to be More Innovative

Emphasis on innovationRobust training programs for next generation of basic and clinical scientists

1 Jan 2010, vol. 326 SCIENCE www.sciencemag.org

e-Infrastructure and e-Science as a Platform for Meeting Grand Challenges

e-science

Dig

ital e

cono

my

e-research e-arts & humanities

e-infrastructure (shared & specific)

e-urgent response e-learning

Ener

gy

Livi

ng w

ith

envi

ronm

enta

l ch

ange

Glo

bal S

ecur

ity

Agi

ng r

esea

rch:

lif

elon

g he

alth

&

wel

lbei

ng

Nan

osca

le:

engi

neer

ing

to a

ppFurther Opportunities??

Grand Challenge Research Initiatives

e-Science

e-Humanities & Arts

e-Learning

TheResearchUniversity

in the Digital Age

e-Development

Transforming American Education: Learning Powered by TechnologySoon to be released Report to the President and Congress from the U.S. Department of Education

This report was in part motivated by and highly leverages e-science and e-learning

research

e-infrastructure has history and theory

GLOBAL

TECHNICAL

LOCAL

GLOBAL

SOCIAL

Transparency

Embodiment of standards

Reach & scope

Links with conventions of

practice

Learned as part of

membership

EmbeddednessBecomes

visible upon breakdown

Built on an installed base

http://deepblue.lib.umich.edu/handle/2027.42/49353

UK e-infrastructure for Science and Innovation

Executive Summary

The growth of the UK’s knowledge-based economy depends significantly upon the continued support of the research community and in particular its activities to engage with industry and to apply its world-leading innovations to commercial use. A national e-infrastructure for research provides a vital foundation for the UK’s science base, supporting not only rapidly advancing technological developments, but also the increasing possibilities for knowledge transfer and the creation of wealth.

http://www.nesc.ac.uk/documents/OSI/index.html

sets out the requirements for a national e-infrastructure to help ensure the UK maintains and indeed enhances its global standing in science and

innovation in an increasingly competitive world.

Six Working Groups: Data & information creation • Preservation &

curation • Search & navigation • Virtual research communities • Networks, compute

and data storage • Authentication, authorisation, accounting, middleware, and

digital rights management

Background on Report and Process

of Its Creation

Terms of Reference:Role of Review Panel

Assess the impact of the programme on research areas nationally and internationally; on broader wealth creation and quality of life; and on e-Science itself;

Assess and compare the quality of the UK research base in e-Science with the rest of the world via triangulation of data, panel and community perception;

Comment on the added value of this programme;

Present findings and recommendations about the strength, weakness and opportunities for the future to the research community and Councils.

Panel’s Concept of the Process and Expectations

Review Week in Oxford

Evidence Fram

ework

200 Page Evidence Documentation

Provided by RCUK

Presentations, Conversations, Observation

Panel’s collective expert judgement

Findings

Context, Vision, Opportunity Space

Recommendations

Draft Panel Public Report

+Town Hall Meeting

Documentation Requested by Panel Feedback

+Distributed group

writing anddiscussions

Final Report

e-InfrastructureReport

Next Steps

e-Science Evidence Framework(FOGIES)

F - What are the future opportunities for UK e-Science?

O - Other observations not included in the above.

G - How does UK e-Science activity compare globally?

I - What has been the impact (accomplished and potential) of the UK e- Science Programme?

E - Did the UK e-Science Programme build a Platform which enables e- Science tools, infrastructure and practises to become incorporated into mainstream research in the UK?

S - How did the Programme Strategy (having a Core and individual Research Council Programmes, developing tools and applications in parallel) affect the outputs from UK e-Science?

EFeldmanFoster

Wensink

GMilanesi EllismanBindoff

IBorgmanYnnermanHeermann

FHeck Szalay

Tackley

SLane

Parakivon Rüden

OAtkins

Project Presentations

Field Trips

Evidence Documents

Conversations

Panel Knowledge

Other Sources

Panel Explore and Integrate Matrix

At every opportunity we have in our group private sessions (including after some meals), we will report-out

findings of the day and the designate teams will mine these reports for findings and recommendations relevant

to their topic in the (E-GIFSO OR FOGIES) evidence framework. Please be proactive in getting your input to

the appropriate teams.

Findings and Recommendations

The UK has created a “jewel” -- a pioneering, vital activity of enormous strategic importance to the pursuit of scientific knowledge and the support of allied learning.

The UK e-Science Programme is in a world-leading position.

Investments are already empowering significant contributions in the UK and beyond.

The UK must decide whether to create the necessary combination of financial, organisational, and policy commitments to capitalise on their prior investments,.

e-Science is an organic, emergent process requiring ongoing, coordinated investment from multiple funders and coordinated action by multiple research and infrastructure communities. It is both an enabler of research and an object of research,

The UK should continue to nurture a robust infrastructure that couples and balance research, application development, and training processes.

None of this is easy, but done well could achieve enormous reward.

Golden Jubilee Diamond

The UK has created a “jewel” -- a pioneering, vital activity of enormous strategic importance to the pursuit of scientific knowledge and the support of allied learning.

Twinkling in the sky is a diamond star of 10 billion trillion trillion carats, astronomers have

discovered.The diamond star completely outclasses the

largest diamond on Earth, the 546-carat Golden Jubilee...

http://news.bbc.co.uk/2/hi/sci/tech/3492919.stm

Produced competent people for both academia and industry.Created and enriched interdisciplinary collaborations (faculty & students) that are increasingly important for progress at the frontiers of scientific discovery.Created interdisciplinary e-science doctoral training programs.Stimulated industry up-take of e-science.Situated the UK reputation in e-science among the top nations. Accelerated the penetration of e-science capability in UK and Europe.Contributed significantly to establishing standards and tools for national and european consortial networked science projects.

SIGNIFICANT FINDINGS: STRENGTHS (1)

Stimulated university leaders to establish e-science buildings and organizations on campus.Promoted recognition of digital data as an asset and the need for greater attention to its stewardship.

But still lacking adequate repository capacity and the trained human capacity that is needed.

Supported directly and indirectly scientific international collaborations.But very little (any?) participation by developing countries.

Enabled science not otherwise possible.Initiated a national framework for sharing large-scale resources that will be increasingly important to both large and small scale research and education.

SIGNIFICANT FINDINGS: STRENGTHS (2)

Supported important work in the humanities and social sciences.Promoted increasing standardization for data interoperability and aggregation.Promoted interdisciplinary work and provided mechanisms to link projects to projects on an on going way. (All-hands Meetings; community building activities)Leveraged investments from other sources (regional, industrial, international)Stimulated local/regional economies.Accelerated the productivity of researchers and science processes.Accomplished some knowledge/technology transfer but there could be more.

SIGNIFICANT FINDINGS: STRENGTHS (3)

Too early and rapid reduction of core funding and high level leadership, resulting in

Sense of abandonment by some researchers and career shift away from e-scienceFailure to reap the benefits of prior investments.

The time constants for real transformative impact and significant competitive advantage is decades.

Reduced momentum of an important movement in which UK established a global leadership rolePerception that RCs don't appreciate the strategic importance of e-science and the special handling that it still needs

SIGNIFICANT FINDINGS: WEAKNESS (1)

Lack of models for sustaining/maintaining software, platform/infrastructure operations, and data resources

e-infrastructure providers are being forced to compete for research funding (rather than targeted infrastructure/facilities funding) every two years

Missed opportunities to transfer successful software tools/systems and best practices to other fields. Lack of regularized processes and organizations to do so.

SIGNIFICANT FINDINGS: WEAKNESS (2)

Lack of adequate structure, leadership, and resources to promote constructive interplay and achieving balance (1) between shared/generic and specific e-infrastructure and (2) between e-science core and disciplinary research.

This situation raises barriers to interdisciplinary work; to developing and exploiting common infrastructure, middleware, tools, systems; and to interoperability between disciplines, projects, platforms and shared use of resource

Overly focused on computational "grid" platforms Good fit to physics but not necessarily all other domains. But there are variations in what "grid" means. (computational, data, all types of resources)

Knowledge transfer (KT) potential higher than was actually achieved. Need more systematic and staffed KT processes.

SIGNIFICANT FINDINGS: WEAKNESS (3)

Shortage of women participants (as judged by data in evidence document and gender balance during review)Gaps in the e-science program conceptualization.

sensor networksusability, human-computer-interactionunderstanding of social/behaviors barriers to technology mediate science collaboration and ways to design systems and processes to reduce barriers

SIGNIFICANT FINDINGS: WEAKNESS (4)

Crossing the Chasm – Modified

e-Science Prog.

Recommendations for the future within a 5 year

window, informed by a

10 year vision

Next 5 YearsFrom Research to Sustainability

Maintain critical-size centers already established Partnership with academic institutions already committed to e-Science

Dramatically expand (x10 or more) involvementBoth researchers and users (industry, society)

Provide mid-term career paths for current e-science personnelUse existing mechanisms like 5-year UKRC fellowships

Build stronger ties/bridges to the HPC community (e-Science should include simulation, modelling, prediction, and data mining activities.)

Durham cosmology simulations, Oxford effortDevelop increasingly shared infrastructure that is reliable, mature and sustainable

Software is a sustainable facility, and the capital investmentHardware is becoming the consumable

Next 5 YearsSustain and Grow the Emerging Community

Strong national leadership, stable cross-council coordination with real authority.

Community building and training

National and regional e-Science Centers, All Hands Meetings

Summer schools, Doctoral Training Centers

Systematic dissemination of best practice

Knowledge transfer organization

Explicitly reward “re-use” of existing components

Emphasise packaged and hosted service-based e-Science

Keep up with emerging/evolving new technologies

Shared support center, explore clouds, GPUs, etc

Next 10 YearsEvolutionary Challenges

Transformations in different disciplines will not happen all at the same time

HEP … astro … genomics … environment … health … (most disciplines)Continuous evolution in underlying technologies

Grid … SOA.. cloud … ??? Innovative “sensors” tied to computational tools(sequencing, gene chips, remote sensing, medical imaging, sensing data for social sciences)

Stable hierarchical distribution of resources, with large national or international scale facilities at the top

Group … University … Regional … National … InternationalSanger, EBI, IVOA, LHC Tiers (within and across disciplines)

Next 10 YearsData Challenges

Data sets: “new kinds of instruments”LHC, Sanger, Oceanography, AstroGrid, Medical records, 4th paradigm

Self-amplification of dataThe easier to collect and analyze large data, the more people collect…Data doubling times go from 1 year to 3-4 months

Simulations larger and in more disciplinesHeart, brain, cell, astro, LHC, socialMore pressure on the infrastructure

Data growth beyond any current imaginationMany 100s of PB in 10 yearsCapacity issues hard even on national scale, impossible locallyNot only technical but a socio-political issue

Next 10 YearsSociological Challenges

Cultural changes cause e-Science to become “socially” accepted in more and more disciplines (to stay competitive)

Many authors, data sharing, novel kinds of contributions e-Science buildings and departments are first signs of wider acceptance, but still small fraction of papers impacted

e-Science (“computational”) thinking becomes an integral part of the everyday curriculum across all disciplines

7 Doctoral Training Centers alreadyE-Science will become mainstream when integrated into the educationframework as a whole

Substantial fraction of the research community exposed to e-ScienceBiggest changes will happen through demographics (“progress one grave at a time”)

A Dozen Recommendations

1. Structure and Leadership

Continue to treat e-Science as a designated strategic initiative spanning all Research Councils and having ongoing designated funding.

Provide high-quality dedicated leadership for a strategic e-Science Programme, and provide the leader with adequate authority at a high enough level and with resources to catalyse real synergy and co-investment.

Establish more systematic, ongoing, coordinated investments in the infrastructure, development, and adoption of e-Science at even a higher level than across Research Councils. This higher level coordination could include several components of BIS: the Research Councils, JISC, the Technology Strategy Board, and the funders of higher education and facilities. e-Science Programme leadership should also seek coordination with e-Science-related funding from outside of government including the Wellcome Trust, the EU, the US NSF, and others.

Balance of Funding RegionsSpecific

Shared/Generic

Research Infrastructure

1. Domain Specifc

Research Programs

2. Domain Specific

e-Infrastructure Programs

4. Shared (multi-domain) e-

Infrastructure Programs

3. Collaborative, (interdisciplinary)

Research Programs

Intellectual Assets e-infrastructure Assets

Based on work by Suzi Iacono & Diana Rhoten at NSF

e-science management

structure

2. Industry-academic collaborations

Establish more systematic and better supported mechanisms, including targeted funding, to nurture collaboration and bi-directional knowledge transfer between academia and industry in the creation, provisioning, and application of e-Science.There are multiple goals here:

(1) accelerate transfer of research outputs to beneficial innovation;

(2) enlist industry in the creation and provisioning of the e-infrastructure platform for e-Science; and(3) to help industry adopt and tailor best practices and services from e-Science to enhance their own productivity.

http://www.networkworld.com/community/node/27219

http://www.nsf.gov/news/news_summ.jsp?cntn_id=116336&org=NSF&from=news

http://www.nytimes.com/2010/02/05/science/05cloud.html

Recent novel examples of government, industry, academia partnerships on e-Science research

3. RCUK e-Science Centre Network

Sustain and strengthen the RCUK network of e-Science Centres.Challenge and support these Centres to both serve regional constituencies and be members of a network for the common good.Establish the remit of various centres in a way that stresses complementary expertise and sharing. (give and take)Emphasise and assess expectations that the Centres be proactive in engaging with each other, with UK industry, and with other countries. Establish polices for sustaining this network but in a way that periodically (e.g. every 5 years) requires re-competition to enable new entries into the field and to retire less effective activities.

4. Sustaining advanced e-infrastructure

Sustain the operational e-infrastructure for e-Science created to the present. Informed by an ongoing review of future needs, evolve it to: (1) higher capacity; (2) more complete function; and (3) leading-edge, distributed system architecture.

In particular the UK needs to invest in e-infrastructure in ways that (1) anticipates the continuing exponential growth in scientific data and the increasing ability to extract knowledge from it; (2) provides the UK research community access to petascale-level computing and anticipates the future needs for access to exascale; (3) continues to build on and strengthen the grid model of distributed computing, but also explores the adoption of emerging models of cloud computing for research.

5. Supporting complementary personnel roles

Recognise in programme calls and funding policies that there are people in several complementary roles that need to be funded and rewarded appropriately. There are people

(1) seeking to innovate in the application of e-Science method to their field;

(2) seeking to innovate better e-Science methods, practises, and services;

(3) who design and build reliable production-grade systems; and

(4) who administer and operate the supporting e-infrastructure.

Need to better define and reward new professional identities and job types within academia that span role 2 and 3 above.

Transformative Application - to enhance research, learning,

and societal engagement.

R&D to provide technical and social enhancements in future e-science environments

Provisioning - to create, deploy and operate

advanced e-infrastructure

How do we enhance a culture of mutually-beneficial balanced participation and synergy between e-infrastructure provisioning, research application, and e-science/e-infrastructure innovation ?

Borromean Ring: The three rings taken together are inseparable, but remove any one ring and the other two fall apart. See www.liv.ac.uk/~spmr02/rings/

6. Sharing for cost- and science- effectiveness

Continue funding policies that strongly encourage or require the creation and adoption of shared e-infrastructure.

Important not only from a cost-effective, efficient-energy use, and environmental-impact perspective, but also for facilitating intellectual interoperability between disciplines, institutions, facilities, and data resources essential for grand-challenge research.

Since scientific research is intrinsically global, place great emphasis on creating UK e-infrastructure that harmonises with the e-infrastructure in other countries.

7. Role for arts and humanities

Encourage and support even more participation of the arts and humanities research communities in the e-Science program. (We saw some excellent beginnings in our review.) Arts and humanities are poised to achieve large benefit from e-science methods and infrastructure as the human record becomes increasingly digitised and multimedia.

8. Role for social sciences

Encourage and support even greater leadership by the social science research community in the adoption of e-Science methods particularly, for example, given capabilities to explore enormous data sets, analyse social networks, and explore very complex systems through simulation and modelling.

The social science community should also be encouraged to contribute more to deeper understanding of more principled ways to design effective virtual research environments, collaboratories, (four-quadrant organisations).

9. Crossing the chasm; refreshing innovation

Develop a dual strategy that both (1) accelerates the adoption of e-Science methods in the “mainstream market” of researchers and (2) refreshes the investments in the “early market” to produce the next wave of innovation in e-Science services and application.

Goal (1) involves training and tailoring current services to more specific needs of disciplinary and interdisciplinary communities. This process needs an integrated formative assessment activity that includes continuous monitoring and that helps ground and inform the activities of “early market” communities and supports participatory design methods.

10. Data stewardship at enormous scale

Continue the strong focus on creating practices and services for appraisal, curation, federation, and long-term access to scientific data.

Complement or broaden the activities of the Digital Curation Centre with the creation of coordinated and sustained production services for curation and stewardship of scientific data.

Consider a highly centralised, large data centre model for storing the information and preserving the bits, together with a distributed model for curation by disciplinary specialists. Seek and promote international cooperation.

11. Openness as a general policy

Establish policies for openness: open-source code, open data and meta-data, and open courseware. Encourage reuse.

Dimensions

"Open" Resources for Research and

Learning

Intellectual access

Technical access

Monetary payment or

notAnonymous or

not

IP terms and

conditions for use

Default copyright

Creative Commons Licenses

Attribution Commercial or not

Derivative works or not Share Alike See http://

creativecommons.org/

Extent and method of

credentialling/warranting

Organizational form for

collaborative creation & use/

remix

Type of objects/

collection

Software (open source) Courseware Online

laboratoriesDigital capture

of lectures

Venues/tools for

participatory learning

Journal Publications Data

Raw data Meta data Analysis tools

12. Towards functionally complete, four-quadrant, research environments

Pursue capacity for collaborative, international, interdisciplinary team science to occur routinely in “functionally complete, four-quadrant environments” built upon e-infrastructure.

“A four-quadrant environment” refers to a blended virtual-physical environment in which the activities of a group can flow easily between all four quadrants in a 2-by-2 matrix with same versus different for the two dimensions of both time and place. It subsumes the concept of virtual research environment.

“Functionally complete” means that the environment supports access to all the people, the information and data services, the observatories and facilities, the computational services, and the collaboration and communication services necessary for a scientific team (or more generally, a community of practice) to carry out its work.

Daniel E. Atkinsatkins@umich.eduUniversity of Michigan P: people, I: information, F: facilities, instruments

P: Physical mtgsI: Print-on-paper books, journalsF: Wet labs, studios, shops

P: Shared physical artifactI: Library reserveF: Time-shared physical labs, ...

P: AV conferenceI: Web searchF: Online instruments

P: EmailI: KnowbotsF: Autonomous observatories

TimeSame

(synchronous)Different

(asynchronous)

Geo

gra

phi

c P

lace

Sam

eD

iffer

ent

Physical + Virtual,Not Physical vs.

Virtual

Four Quadrant Organizations (virtual organizations) offer additional modes of interaction between People, Information, and Facilities

Daniel E. Atkinsatkins@umich.eduUniversity of Michigan P: people, I: information, F: facilities, instruments

P: Physical mtgsI: Print-on-paper books, journalsF: Physical labs, studios, shops

P: Shared physical artifactI: Digital librariesF: Time-shared physical labs, ...

P: AV conferenceI: Web searchF: Online instruments

P: EmailI: KnowbotsF: Autonomous observatories

TimeSame(synchronous)

Different(asynchronous)

Geo

gra

phi

c P

lace

Sam

eD

iffer

ent

Need to discover how to best map collaborative workflows onto various paths through these collaboration states.

Beyond Being There: A Blueprint for Advancing the Design, Development, and Evaluation of Virtual Organizations

Available at: www.ci.uchicago.edu/events/VirtOrg2008/

Use e-Science as a Pathway to New Forms of High Performance Collaboration for Discovery and Learning

Creating a Transformative e-Science Programme is a Complex Balancing Act

also

Creating a Transformative e-Science Programme is a Complex Balancing Act

Discussion & Questions

e-Science