Minutes of the 15th TAG

The e-Science Core Programme

The e-Science Technical Advisory Group

Minutes of the 17^th meeting

20 January 2005

DTI, 151 Buckingham Palace Road, London

(Agenda)

Present:

Prof. Tony Hey (Director, Core Programme, DTI/EPSRC) Chairman
Dr Jamil Appa (BAE Systems Advanced Technology Centre)
Prof. Malcolm Atkinson (NeSC)
Prof. Peter Dew (Leeds University
Dr Alistair Dunlop (OMII)
Prof. Geoffrey Fox (University of Indiana)
Dr Paul Jeffreys (Oxford University Computing Services)
Prof. Ron Perrott (Queens University Belfast)
Prof. Nigel Shadbolt (Southampton University)
Prof. Ian Sommerville (Lancaster University)
Prof. Paul Watson (Newcastle University)
Prof. David Williams (CERN)

In Attendance:

Deborah Miller (PPARC) Secretary
Audrey Canning (DTI)
Dr Jim Fleming (EPSRC)
Dr Neil Geddes (CLRC)
Dr David Salmon (UKERNA)
Dr Anne Trefethen (EPSRC)

Apologies:

Dr Graham Cameron (EBI)
Prof. Jon Crowcroft (Cambridge University)
Dr Glenn Gapper (BAE Systems Advanced Technology Centre)
Dr Mike Kirton (QinetiQ)
Linda McCormick (Glasgow University)
Dr Angela Sasse (UCL)
Steve Tuecke (Argonne National Laboratory)

Minutes for the 16^th meeting

1

The minutes were agreed as an accurate record of the meeting subject to the following amendment.

List of Members Present: “…Prof. Paul Jeffreys.”

Actions and matters arising from the minutes

2

The following matters arising from the minutes were discussed:

a

Minute 2e: Malcolm Atkinson to invite John Taylor to join the DCC Management Board. Done

b

Minute 2f: Presentation to TAG information on smaller activities. Jim Fleming said that there would be a presentation on the Centre Industrial Projects at this meeting and that an Open Call and IRC Projects meeting was planned for later in 2005. Other events/presentations would be arranged if members requested them.

c

Minute 2j: Circulate details of the UCISA/JISC campus security activity website. Outstanding. Action Linda McCormick

d

Minute 3c: Tony Hey to circulate the Report on the SAKAI programme. Done.

e

Minute 3g: Graham Cameron to send details of the EU Network of Excellence, EMBRACE to Tony Hey.   Done.

f

Minute 4a: Neil Geddes said that he had not yet been invited to a meeting of the CCP Steering Committee. He also said one CCP had informed him that it would not be using the NGS and a new CCP had been proposed in bioinformatics.

g

Minute 5b: Neil Geddes and Alistair Dunlop to discuss the role of the ETF in evaluating alternative systems to OMII and their plans for development. Done.

h

Minute 7e: Jim Fleming to forward details of EPSRC projects on fundamental computer science to Debbie Crawford. Done.

i

Minute 8a: Jim Fleming to forward details of the two new EPSRC pilot projects to Tom Rodden. Done.

j

Minute 9d: Jim Fleming to look into the requirement for EPSRC e-Science research grant holders to sign the e-Science security policy document. Jim Fleming reported that the new e-Science projects had been announced before the security policy document had been produced Also the projects’ security requirements were, in fact, more rigorous than the e-Science security policy.

k

Minute 10: Tony Hey said that the allocation of funds from SR2004 would be discussed at the next TAG meeting.

l

Minute 10c: Tony Hey to discuss with Lesley Thompson EPSRC’s role in supporting R and D in the underpinning technologies. Done.

m

Minute 10c: Neil Geddes to suggest suitable amendments to the Multidisciplinary Collaborative Science and Innovation paper. Done. The paper had been revised incorporating TAG and eSSC comments and sent to Sir Keith O’Nions as a White Paper.

n

Minute 11a: Jim Fleming to look into the costs of the UK e-Science stand at SC2004. Done. NeSC/RAL had been asked to produce a final report.

o

Minute 11b: Anne Trefethen to look into how SC2004 could be used to promote the UK e-Science Programme. Done.

Chairman's report

3

Tony Hey presented his report (.ppt ) The following issues arose:

a

The question of how generic activities within the individual DTI Technology Fund projects would be coordinated.

b

In the “FEC” model for research grants funding, the research Councils would be providing 80% of the indirect costs.

c

There were significant challenges in providing the operating infrastructure for the “plug and play” vision. Having every user making choices on components would be unsustainable. “Configure and compute services” might be a better description of the vison. Members were asked to suggest names to brand the e-Science grid vision. Action: All

d

Further information was requested on the Sisters Project. Action: Jim Fleming {Secretary’s note – Further details are annexed to these minutes) Tony Hey said that, if there were sufficient funds in the next phase of the e-Science Programme, he intended to widen international collaboration. A report on the Sisters Projects would be presented to the July TAG meeting. Action: Tony Hey

e

Malcolm Atkinson suggested that a meeting of all the Sisters Projects might be beneficial, perhaps via a BoF session at the All Hands meeting. Action: Malcolm Atkinson

f

Peter Henderson would be looking after links with the GridCoord project. Malcolm Atkinson was asked to provide him with the appropriate briefing for this role. Action: Malcolm Atkinson

g

Ron Perrott asked why the proposed MoU with China did not include HPC activities. Tony Hey said the MoU between EPSRC and MOST covered HPC.

h

Tony Hey said that China did not realise just how strong it was in e-Science and the UK could assist in helping China exploit this expertise, for example China was not strongly involved in the GGF. Geoffrey Fox suggested that setting up some joint architectural groups might also be valuable. On security research, TAG also agreed that it would be useful to engage with the Crown Project, funded by NSF China. Nigel Shadbolt said that China was also strong in the semantic web and it was planning to set up a large conference on that topic to be held first in China and then in the UK. Tony Hey agreed to develop a strategy for engagement with China. Action: Tony Hey

I

A call was out for Data Intensive demonstrator projects at SC2005.

J

A further meeting was planned with the Health Protection Agency to follow up on initial ideas for joint activities.

Grid Operations and Support Centre and the National Grid Service (GOSC&NGS)

4

Neil Geddes presented his report (.ppt) on the Grid Operations and Support Centre (GOSC) and the National Grid Service (NGS). The following issues arose:

a

Average usage across the nodes was around 25%. Nearly all the users were EPSRC researchers, but NERC usage was increasing. No plans were currently in place for marketing the service. Peter Dew said that this needed to be addressed and a training programme rolled out.

b

The NGS affiliation documentation should help local grids configure themselves to fit the NGS.

c

A Campus Grid Workshop would be held at NeSC on 16^th/17^th June.

d

Geoffrey Fox said that the NGS looked a more elegant system than TeraGrid.

e

The UK ETF was integrated with the GOSC, providing 0.5 FTE at each regional centre to evaluate software distributed by OMII, EGEE etc and to feedback to the ETF. The evaluation documents would be made public at some point for educational purposes. The aim of this activity was to maintain compatibility in the light of convergent forces e.g. OMII, gLite, GT4.

f

Industrial companies could get access to the NGS via their academic project partners. Also UKERNA had some flexibility regarding industrial access to JANET. Potential industrial users should discuss their requirements directly with UKERNA.

g

Ian Sommerville said that the current funding model for the GOSC/NGS was not viable in the long term and possibly the only way to make it sustainable would be for users to pay for access.

h

Alistair Dunlop said that e-Science projects were encouraged to adopt WS technology, but the NGS was running GT2 infrastructure. A rapid evolution to WS was required. Neil Geddes said this would be decided at the next GOSC Board meeting in March.

Open Middleware Infrastructure Institute (OMII)

5

Alistair Dunlop presented his report (.pdf .ppt) on the OMII. In discussion the following issues arose:

a

The provision of suitable training, and the training of trainers, were vital to bring down barriers for new users of the OMII. The proposed involvement of NeSC personnel in this would be helpful. Malcolm Atkinson said he was preparing an EU proposal on e-Science education and was looking for potential collaborators.

b

Paul Watson said that the NGS should support at least the OMII scheduling interface and more work was needed on this.

c

David Williams said that the OMII had had a very encouraging first year, but the challenge for TAG and the eSSC was developing the relationship between OMII/NGS and the e-Science projects.

d

Nigel Shadbolt said that the OMII was looking at the knowledge life-cycle for e-scientists and this topic was just as important as the development of middleware.

e

There was a danger of the project being pulled in a particular direction by users’ requirements resulting in key developments being overlooked e.g. resource brokering was important but had not been requested by users.

f

The report had not discussed how the OMII would get industry on board for long-term sustainability. Alistair Dunlop said he had given this issue some thought, but it was proving difficult to address. For the next year OMII would be concentrating on its integration with key e-Science projects.

Future Evolution of OMII

6

Tony Hey presented his report. .ppt In discussion the following matter arose:

a

It might be difficult for the OMII to generate revenue by using middleware in commercial products if the software had been developed in an HEI. This could be resolved by dual licenses or by charging for services rather than software.

b

The management of the “hub and spoke” model and funding of the “spokes” needed to be addressed. Tony Hey agreed to put together a more detailed plan for the next TAG meeting. Action: Tony Hey

c

There was a need to open negotiations with potential partners in OMII-International. The UK was not in a position to build all the required middleware, so a common international infrastructure was required. However, much of the work could be done in the UK, especially in collaboration with China. It would also be important to maintain links with the US.

d

The closing date for the EU call for e-Infrastructure Grids was 7^th September 2005, so plans needed to be in place quickly if this deadline was to be met.

Data Curation Centre Update

7

Malcolm Atkinson gave an oral report. The following items were reported:

a

The DCC launch was held on 2^nd November and was very successful. The day included the formal launch followed by a series of discussions about scientific digital curation research and strategy.

b

The Director of the DCC, Chris Rusbridge, had been appointed and would take up his post on 18^th February. He was a former employee of JISC.

c

5 of 6 of the research posts had been filled. The post at Glasgow was still unfilled.

d

Rennie Miller of IBM Research visited the DCC in January. The Duke of Edinburgh would be visiting on 17^th February and Dr David Rosenthal of LOCKSS would be visiting on 20-27^th February.

e

The DCC was contributing to the Digital Curation Consortium Away Day on 9^th February.

f

There would be a DCC conference in Bath during September (possibly overlapping with AHM).

g

The Digital Curation manual was under construction and would be published on the website in a few weeks.

h

A meeting with JISC funded repositories would be held on 7^th March.

i

The DCC was leading three of the tracks at IASIST, the Social Sciences IT conference, in May.

j

The work of the EU Information Structures Group had been passed to the DCC, which would take it forward as the first component of the registry of standards.

k

The DCC was attempting to set up an International Journal of Digital Curation

l

The DCC would provide a full report on the first 12 months by the end of February. The Office was asked to clarify the DCC reporting lines with the e-Science Core Programme. Action: Deborah Miller The DCC Advisory Board had not yet met.

m

It was suggested that Chris Rusbridge be invited to a future meeting (July 2005) of TAG to present a report. Action: Deborah Miller

Security Issues

8

Anne Trefethen gave an oral report. The following items were reported:

a

The next meeting of the Security Task Force would be discussing a paper, developed by Peter Henderson and the OMII, which reviewed security technologies. This was required input for the White Paper on Security.

b

UKERNA was putting together a survey paper on the issues the UKERNA community needed to consider regarding grids, e.g. the advantages and disadvantages of various technologies.

c

Angela Sasse was working on the usability aspects of security.    The results of this work would be presented to the next TAG meeting. Action: Angela Sasse

d

Malcolm Atkinson asked when drafts of these papers would be available as the OGSA-DAI Technical Review Board would like to see copies at its meeting on 18^th February. Anne said that the OMII review of security technologies and the UKERNA paper would be available in 2-3 weeks but the usability paper would not be available until mid-March at the earliest.

e

EPSRC wanted to put funds into security research and had convened a sub-group to flesh out what might be included in such a research programme.

f

Tony Hey said he had been discussing with Alan Robiette the possibility of presenting the results of JISC funded security evaluations to the community. Malcolm Atkinson was asked to flag this up as possible e-Science event at NeSC. Action: Malcolm Atkinson

DTI Technology Programme Update

9

Audrey Canning gave a brief update of the programme. The following issues arose:

a

The e-Science community had done well in the April 2004 competition but several very good projects did not get through as the call was very competitive. In the November 2004 competition two themes would be of interest to the e-Science Community; Design, Simulation and Modelling and Pervasive Computing including Networks and Sensors.

b

Ian Sommerville said that a company he had persuaded to apply to the April competition had been disappointed by the lack of feedback from the DTI. Audrey said improved feedback processes were being developed. It was also suggested that referee comments be fed back directly to applicants, as was standard practice within EPSRC. Audrey said she would take this suggestion back to the DTI.

E-Science Regional Centres

10

Anne Trefethen presented a report .ppt .pdf on recent visits to the e-Science Regional Centres undertaken by herself and Audrey Canning. During discussion the following points were made:

a

The DTI funding of these centres had been more than matched by additional industrial funding and many centres had attracted significant EU funding.

b

Audrey Canning was currently developing criteria for evaluating the final reports of these projects.

AOB

12

The next meeting would be held on 27 April 2005.

Action List:

From the meeting held on 18^th October 2004

2c:

Linda McCormick to circulate details of the UCISA/JISC campus security activity website.

From this meeting:

3c:

All Members to suggest names to brand the e-Science grid vision.

3d:

Jim Fleming to provide further information on the Sisters Project.

3d:

Tony Hey to present a report on the Sisters Projects to the July 2005 TAG meeting.

3e:

Malcolm Atkinson to consider holding a BoF sessions for the Sisters Projects at the All Hands meeting.

3f:

Malcolm Atkinson to brief Peter Henderson on his role of liaising with the GridCoord project.

3h:

Tony Hey to develop a strategy for engagement with China.

6b:

Tony Hey to present a plan for the management and funding of OMII Phase 2 at the April 2005 TAG meeting.

7l:

Deborah Miller to clarify DCC reporting lines with the e-Science Core Programme.

7m:

Deborah Miller to invite Chris Rusbridge to the July 2005 TAG meeting.

8c:

Angela Sasse to present a paper on the usability aspects of security to the April 2005 TAG meeting.

8f:

Malcolm Atkinson to flag up the presentation of the results of the JISC security evaluations as a future e-Science event at NeSC.

Sisters Projects Annex to TAG minutes 20^th January 2005

PI

Organisation

e-Science Project

Funders

Title

Sister Project

Abstract

Moreau, Professor L

University of Southampton

CONOISE-G/PASOA/VO

EPSRC

SOCA Sister Project: SOuthampton Chicago Activity

Argonne

The Grid is a large scale computer system that is capable of coordinating resources that are not subject to centralised control, whilst using standard, open, general-purpose protocols and interfaces, and delivering non-trivial qualities of service. While a number of projects have been reporting success stories, in which "Grid computing" was applied to highly visible scientific applications, a tremendous number of challenging issues still remain to be investigated in order to address this vision fully.
Southampton and Chicago, which both have teams actively involved in multiple Grid and e-Science collaborations, share a remarkable number of common interests in the area of resource discovery, provenance, workflow languages, negotiation and virtual organisations. These different topics are being investigated in various projects in Southampton (myGrid, PASOA, EU Provenance, Conoise-G, Grimoires, Virtual Organisations for e-Science) and Chicago (GriPhyN, iVDGL, Globus). Additionally, the partners bring to the table complementary expertise from different communities: Grid and High Performance Computing, Agent-based computing and Semantic Web and Grid technologies.
SOCA, the SOuthampton Chicago Activity, is a sister project that sets up a formal collaboration between Southampton and Chicago in order to investigate some fundamental issues related to Grids. The SOCA project will establish bridges between American and UK communities, but also between Grid and Agent communities, and between Grid and Semantic Grid communities.

Frey, Dr JG

University of Southampton

CombeChem/e-Bank

EPSRC /JISC

The Crystal Grid Network

Reciprocal Net (US)

Grid based interactions with central crystallography facilities have been developed separately in the UK (CombeChem/e-Bank) and the USA (Reciprocal Net, Collaborative Multi-Scale Chemistry Sciences - CMCS). Initial meetings have demonstrated that these projects have many complimentary features in design and architecture. We propose to collaborate on the formation of an international standards and implementation for grid based data collection, monitor & control of laboratory environment and data services for crystallography. We will achieve these aims efficiently by sharing and integrating the complimentary components of the grid services developed for the EPSRC National Crystallography Service (NCS, UK) and the Smart Labs systems with those of Reciprocal Net and CMCS (USA). We aim to similarly integrate the disseminations systems of e-Bank and Reciprocal Net and extend these to the new e-science community in Australia.

Davies, Dr JW

University of Oxford

OeSC, CancerGrid

CP/MRC

International Collaboration in Data Standards and Software Engineering for Cancer Bioinformatics

caBIG

1. Foster and cement links between CancerGrid and the NCICB caBIG project.

2. Extend the Cancergrid-caBIG collaboration to other relevant user-groups within the project and the community it serves, including development staff from the project, trial coordinators, statisticians and software engineers from clinical trials units served by the project,

and bioinformaticians working on the project's principle demonstrators.

3. Promote the caBIG project to the cancer and wider bio-medical research communities in the UK, enhancing opportunities for further collaboration, integration and reuse of its approach and software products.

Gavaghan, Dr DJ

University of Oxford

IB

EPSRC

a Truly Global Virtual Research Environment for the Modelling and Simulation of the Heart

Tulane University (US), UCLA (US), UCSD (US)

The primary aim of the Integrative Biology project is the development of the Computational Infrastructure using state of the art technologies (termed "Grid Middleware") required to support post-genomic research in biology and medicine. Integrative or systems approaches to research in biology are evolving very rapidly, driven by a pressing need to understand how the components that make up a biological system interact to allow biological function, such as the beating of the human heart, to emerge, and to express that understanding in quantitative (i.e. mathematical) terms. It is now widely accepted that this can only be achieved by close collaboration between experimentalists, mathematicians, and computer scientists.
The IB consortium initially represented a collaboration between 6 leading UK Universities (Oxford, Nottingham, Leeds, UCL, Birmingham and Sheffield), the Bioengineering Institute at the University of Auckland, the Central Laboratories of the Research Councils and the IBM corporation. To test fully the computational or Grid infrastructure that we intend to develop, we are now requesting funding through this proposal to extend the consortium to include three leading heart-modelling groups in the USA led by Professors Natalia Trayanova at Tulane, Andrew McCulloch at UCSD, and Alan Garfinkel at UCLA. This will be achieved through a series of annual visits by UK researchers to the US to install middleware and to train users, and an annual workshop in the UK at which researchers from the US will report on their experiences of using the system and suggest improvements. The final deliverable of the project will be a truly global computational grid infrastructure supporting the research of a leading international research community.

Atkinson, Professor M

University of Edinburgh

OGSA-DAI

CP

DIALOGUE - Data Integration Applications: Linking Organisations to Gain Understanding and Experience

DataCutter (Ohio), dQUOB (Indiana), GridMiner (Vienna), SDSC

Thanks to major investment by the UK e-Science Core Programme, the UK is at the forefront of work that provides convenient access to databases and other data resources via the grid and web services. This has the goal of supporting complex data integration from heterogeneous, independently administered, distributed data resources. Better support for such integration is becoming essential as the scale, diversity and quantity of data resources is growing very rapidly. This rich international environment of data has huge potential for discovery by combining independently collected data. Before this potential knowledge can be harvested it is necessary to develop data access and integration technology that can cope with this complex and ever changing environment of data resources.
The ability to combine information conveniently, economically and reliably is vital for science, business, e-Health, e-Government, engineering and many other socially and economically significant activities. It will be an essential component of the UK's and Europe's e-Infrastructure.
Building this e-Infrastructure is a major challenge. Each group collects its data in its own way. So data is stored in very different forms. Each group changes its arrangements for storing data, changing the structure, the way in which the things it observes are represented, and its policies for allowing data to be used. Imagine the different ways in which geologists have named rocks, the number of ways they collect samples, the different coordinate systems they use, and even the different ways they talk about time. Then imagine all the modern ways of collecting more data: fixed and mobile arrays of seismophones, satellite observations that use lidar, radar and GPS to monitor earth surface movements, vertically and horizontally. Add in the different ways in which paleobiologists describe species, chemists describe rock, meteorologists describe weather, and so on. Now in an area of great interest, data from all such sources may be collected and combined, as they are in the GEON project in California for the Yellowstone region.
New "production engineering" is needed to combine data for a scientist who has an idea and wishes to test that idea by comparing data from many sources. New "flexible production engineering" is needed to continue to supply the scientist with her data as the data sources change.
No one group posses the knowledge of all the kinds of data in use, of all the ways in which people want to combine data, and of all the methods we might use in the process. DIALOGUE will gather that expertise, as each partner has specialist knowledge covering a range of data challenges and has already developed methods they can apply. Using the framework of our existing technologies we will plan how we could build the complete "flexible production engineering" system (or at least a major part of it).

Naylor, Professor T

University of Exeter

e-Star

PPARC/CP

Grid Interoperability for Heterogeneous Telescopes Networks

NASA, RAPTOR, MONET (US/SA)

Modern astronomical telescopes are increasingly automated, and communicate via the internet. There are now a small number of networks of such telescopes, and our objective is to be able to unify them into a single, global network that can be accessed either directly by astronomers, or by software "agents" working for them. To achieve this we need to agree the "languages" these telescopes speak and the protocols they adhere to. We are therefore proposing an international workshop and follow-up bilateral visits to create such a network.

The e-Science Core Programme The e-Science Technical Advisory Group Minutes of the 17^th meeting 20 January 2005 DTI, 151 Buckingham Palace Road, London (Agenda)
Present:	Prof. Tony Hey (Director, Core Programme, DTI/EPSRC) Chairman Dr Jamil Appa (BAE Systems Advanced Technology Centre) Prof. Malcolm Atkinson (NeSC) Prof. Peter Dew (Leeds University Dr Alistair Dunlop (OMII) Prof. Geoffrey Fox (University of Indiana) Dr Paul Jeffreys (Oxford University Computing Services) Prof. Ron Perrott (Queens University Belfast) Prof. Nigel Shadbolt (Southampton University) Prof. Ian Sommerville (Lancaster University) Prof. Paul Watson (Newcastle University) Prof. David Williams (CERN)
In Attendance:	Deborah Miller (PPARC) Secretary Audrey Canning (DTI) Dr Jim Fleming (EPSRC) Dr Neil Geddes (CLRC) Dr David Salmon (UKERNA) Dr Anne Trefethen (EPSRC)
Apologies:	Dr Graham Cameron (EBI) Prof. Jon Crowcroft (Cambridge University) Dr Glenn Gapper (BAE Systems Advanced Technology Centre) Dr Mike Kirton (QinetiQ) Linda McCormick (Glasgow University) Dr Angela Sasse (UCL) Steve Tuecke (Argonne National Laboratory)

Minutes for the 16^th meeting
1	The minutes were agreed as an accurate record of the meeting subject to the following amendment. List of Members Present: “…Prof. Paul Jeffreys.”
Actions and matters arising from the minutes
2	The following matters arising from the minutes were discussed:
	a	Minute 2e: Malcolm Atkinson to invite John Taylor to join the DCC Management Board. Done
	b	Minute 2f: Presentation to TAG information on smaller activities. Jim Fleming said that there would be a presentation on the Centre Industrial Projects at this meeting and that an Open Call and IRC Projects meeting was planned for later in 2005. Other events/presentations would be arranged if members requested them.
	c	Minute 2j: Circulate details of the UCISA/JISC campus security activity website. Outstanding. Action Linda McCormick
	d	Minute 3c: Tony Hey to circulate the Report on the SAKAI programme. Done.
	e	Minute 3g: Graham Cameron to send details of the EU Network of Excellence, EMBRACE to Tony Hey. Done.
	f	Minute 4a: Neil Geddes said that he had not yet been invited to a meeting of the CCP Steering Committee. He also said one CCP had informed him that it would not be using the NGS and a new CCP had been proposed in bioinformatics.
	g	Minute 5b: Neil Geddes and Alistair Dunlop to discuss the role of the ETF in evaluating alternative systems to OMII and their plans for development. Done.
	h	Minute 7e: Jim Fleming to forward details of EPSRC projects on fundamental computer science to Debbie Crawford. Done.
	i	Minute 8a: Jim Fleming to forward details of the two new EPSRC pilot projects to Tom Rodden. Done.
	j	Minute 9d: Jim Fleming to look into the requirement for EPSRC e-Science research grant holders to sign the e-Science security policy document. Jim Fleming reported that the new e-Science projects had been announced before the security policy document had been produced Also the projects’ security requirements were, in fact, more rigorous than the e-Science security policy.
	k	Minute 10: Tony Hey said that the allocation of funds from SR2004 would be discussed at the next TAG meeting.
	l	Minute 10c: Tony Hey to discuss with Lesley Thompson EPSRC’s role in supporting R and D in the underpinning technologies. Done.
	m	Minute 10c: Neil Geddes to suggest suitable amendments to the Multidisciplinary Collaborative Science and Innovation paper. Done. The paper had been revised incorporating TAG and eSSC comments and sent to Sir Keith O’Nions as a White Paper.
	n	Minute 11a: Jim Fleming to look into the costs of the UK e-Science stand at SC2004. Done. NeSC/RAL had been asked to produce a final report.
	o	Minute 11b: Anne Trefethen to look into how SC2004 could be used to promote the UK e-Science Programme. Done.
Chairman's report
3	Tony Hey presented his report (.ppt ) The following issues arose:
	a	The question of how generic activities within the individual DTI Technology Fund projects would be coordinated.
	b	In the “FEC” model for research grants funding, the research Councils would be providing 80% of the indirect costs.
	c	There were significant challenges in providing the operating infrastructure for the “plug and play” vision. Having every user making choices on components would be unsustainable. “Configure and compute services” might be a better description of the vison. Members were asked to suggest names to brand the e-Science grid vision. Action: All
	d	Further information was requested on the Sisters Project. Action: Jim Fleming {Secretary’s note – Further details are annexed to these minutes) Tony Hey said that, if there were sufficient funds in the next phase of the e-Science Programme, he intended to widen international collaboration. A report on the Sisters Projects would be presented to the July TAG meeting. Action: Tony Hey
	e	Malcolm Atkinson suggested that a meeting of all the Sisters Projects might be beneficial, perhaps via a BoF session at the All Hands meeting. Action: Malcolm Atkinson
	f	Peter Henderson would be looking after links with the GridCoord project. Malcolm Atkinson was asked to provide him with the appropriate briefing for this role. Action: Malcolm Atkinson
	g	Ron Perrott asked why the proposed MoU with China did not include HPC activities. Tony Hey said the MoU between EPSRC and MOST covered HPC.
	h	Tony Hey said that China did not realise just how strong it was in e-Science and the UK could assist in helping China exploit this expertise, for example China was not strongly involved in the GGF. Geoffrey Fox suggested that setting up some joint architectural groups might also be valuable. On security research, TAG also agreed that it would be useful to engage with the Crown Project, funded by NSF China. Nigel Shadbolt said that China was also strong in the semantic web and it was planning to set up a large conference on that topic to be held first in China and then in the UK. Tony Hey agreed to develop a strategy for engagement with China. Action: Tony Hey
	I	A call was out for Data Intensive demonstrator projects at SC2005.
	J	A further meeting was planned with the Health Protection Agency to follow up on initial ideas for joint activities.
Grid Operations and Support Centre and the National Grid Service (GOSC&NGS)
4	Neil Geddes presented his report (.ppt) on the Grid Operations and Support Centre (GOSC) and the National Grid Service (NGS). The following issues arose:
	a	Average usage across the nodes was around 25%. Nearly all the users were EPSRC researchers, but NERC usage was increasing. No plans were currently in place for marketing the service. Peter Dew said that this needed to be addressed and a training programme rolled out.
	b	The NGS affiliation documentation should help local grids configure themselves to fit the NGS.
	c	A Campus Grid Workshop would be held at NeSC on 16^th/17^th June.
	d	Geoffrey Fox said that the NGS looked a more elegant system than TeraGrid.
	e	The UK ETF was integrated with the GOSC, providing 0.5 FTE at each regional centre to evaluate software distributed by OMII, EGEE etc and to feedback to the ETF. The evaluation documents would be made public at some point for educational purposes. The aim of this activity was to maintain compatibility in the light of convergent forces e.g. OMII, gLite, GT4.
	f	Industrial companies could get access to the NGS via their academic project partners. Also UKERNA had some flexibility regarding industrial access to JANET. Potential industrial users should discuss their requirements directly with UKERNA.
	g	Ian Sommerville said that the current funding model for the GOSC/NGS was not viable in the long term and possibly the only way to make it sustainable would be for users to pay for access.
	h	Alistair Dunlop said that e-Science projects were encouraged to adopt WS technology, but the NGS was running GT2 infrastructure. A rapid evolution to WS was required. Neil Geddes said this would be decided at the next GOSC Board meeting in March.
Open Middleware Infrastructure Institute (OMII)
5	Alistair Dunlop presented his report (.pdf .ppt) on the OMII. In discussion the following issues arose:
	a	The provision of suitable training, and the training of trainers, were vital to bring down barriers for new users of the OMII. The proposed involvement of NeSC personnel in this would be helpful. Malcolm Atkinson said he was preparing an EU proposal on e-Science education and was looking for potential collaborators.
	b	Paul Watson said that the NGS should support at least the OMII scheduling interface and more work was needed on this.
	c	David Williams said that the OMII had had a very encouraging first year, but the challenge for TAG and the eSSC was developing the relationship between OMII/NGS and the e-Science projects.
	d	Nigel Shadbolt said that the OMII was looking at the knowledge life-cycle for e-scientists and this topic was just as important as the development of middleware.
	e	There was a danger of the project being pulled in a particular direction by users’ requirements resulting in key developments being overlooked e.g. resource brokering was important but had not been requested by users.
	f	The report had not discussed how the OMII would get industry on board for long-term sustainability. Alistair Dunlop said he had given this issue some thought, but it was proving difficult to address. For the next year OMII would be concentrating on its integration with key e-Science projects.
Future Evolution of OMII
6	Tony Hey presented his report. .ppt In discussion the following matter arose:
	a	It might be difficult for the OMII to generate revenue by using middleware in commercial products if the software had been developed in an HEI. This could be resolved by dual licenses or by charging for services rather than software.
	b	The management of the “hub and spoke” model and funding of the “spokes” needed to be addressed. Tony Hey agreed to put together a more detailed plan for the next TAG meeting. Action: Tony Hey
	c	There was a need to open negotiations with potential partners in OMII-International. The UK was not in a position to build all the required middleware, so a common international infrastructure was required. However, much of the work could be done in the UK, especially in collaboration with China. It would also be important to maintain links with the US.
	d	The closing date for the EU call for e-Infrastructure Grids was 7^th September 2005, so plans needed to be in place quickly if this deadline was to be met.
Data Curation Centre Update
7	Malcolm Atkinson gave an oral report. The following items were reported:
	a	The DCC launch was held on 2^nd November and was very successful. The day included the formal launch followed by a series of discussions about scientific digital curation research and strategy.
	b	The Director of the DCC, Chris Rusbridge, had been appointed and would take up his post on 18^th February. He was a former employee of JISC.
	c	5 of 6 of the research posts had been filled. The post at Glasgow was still unfilled.
	d	Rennie Miller of IBM Research visited the DCC in January. The Duke of Edinburgh would be visiting on 17^th February and Dr David Rosenthal of LOCKSS would be visiting on 20-27^th February.
	e	The DCC was contributing to the Digital Curation Consortium Away Day on 9^th February.
	f	There would be a DCC conference in Bath during September (possibly overlapping with AHM).
	g	The Digital Curation manual was under construction and would be published on the website in a few weeks.
	h	A meeting with JISC funded repositories would be held on 7^th March.
	i	The DCC was leading three of the tracks at IASIST, the Social Sciences IT conference, in May.
	j	The work of the EU Information Structures Group had been passed to the DCC, which would take it forward as the first component of the registry of standards.
	k	The DCC was attempting to set up an International Journal of Digital Curation
	l	The DCC would provide a full report on the first 12 months by the end of February. The Office was asked to clarify the DCC reporting lines with the e-Science Core Programme. Action: Deborah Miller The DCC Advisory Board had not yet met.
	m	It was suggested that Chris Rusbridge be invited to a future meeting (July 2005) of TAG to present a report. Action: Deborah Miller
Security Issues
8	Anne Trefethen gave an oral report. The following items were reported:
	a	The next meeting of the Security Task Force would be discussing a paper, developed by Peter Henderson and the OMII, which reviewed security technologies. This was required input for the White Paper on Security.
	b	UKERNA was putting together a survey paper on the issues the UKERNA community needed to consider regarding grids, e.g. the advantages and disadvantages of various technologies.
	c	Angela Sasse was working on the usability aspects of security. The results of this work would be presented to the next TAG meeting. Action: Angela Sasse
	d	Malcolm Atkinson asked when drafts of these papers would be available as the OGSA-DAI Technical Review Board would like to see copies at its meeting on 18^th February. Anne said that the OMII review of security technologies and the UKERNA paper would be available in 2-3 weeks but the usability paper would not be available until mid-March at the earliest.
	e	EPSRC wanted to put funds into security research and had convened a sub-group to flesh out what might be included in such a research programme.
	f	Tony Hey said he had been discussing with Alan Robiette the possibility of presenting the results of JISC funded security evaluations to the community. Malcolm Atkinson was asked to flag this up as possible e-Science event at NeSC. Action: Malcolm Atkinson
DTI Technology Programme Update
9	Audrey Canning gave a brief update of the programme. The following issues arose:
	a	The e-Science community had done well in the April 2004 competition but several very good projects did not get through as the call was very competitive. In the November 2004 competition two themes would be of interest to the e-Science Community; Design, Simulation and Modelling and Pervasive Computing including Networks and Sensors.
	b	Ian Sommerville said that a company he had persuaded to apply to the April competition had been disappointed by the lack of feedback from the DTI. Audrey said improved feedback processes were being developed. It was also suggested that referee comments be fed back directly to applicants, as was standard practice within EPSRC. Audrey said she would take this suggestion back to the DTI.
E-Science Regional Centres
10	Anne Trefethen presented a report .ppt .pdf on recent visits to the e-Science Regional Centres undertaken by herself and Audrey Canning. During discussion the following points were made:
	a	The DTI funding of these centres had been more than matched by additional industrial funding and many centres had attracted significant EU funding.
	b	Audrey Canning was currently developing criteria for evaluating the final reports of these projects.
AOB
12	The next meeting would be held on 27 April 2005.
Action List:
	From the meeting held on 18^th October 2004
2c:	Linda McCormick to circulate details of the UCISA/JISC campus security activity website.
	From this meeting:
3c:	All Members to suggest names to brand the e-Science grid vision.
3d:	Jim Fleming to provide further information on the Sisters Project.
3d:	Tony Hey to present a report on the Sisters Projects to the July 2005 TAG meeting.
3e:	Malcolm Atkinson to consider holding a BoF sessions for the Sisters Projects at the All Hands meeting.
3f:	Malcolm Atkinson to brief Peter Henderson on his role of liaising with the GridCoord project.
3h:	Tony Hey to develop a strategy for engagement with China.
6b:	Tony Hey to present a plan for the management and funding of OMII Phase 2 at the April 2005 TAG meeting.
7l:	Deborah Miller to clarify DCC reporting lines with the e-Science Core Programme.
7m:	Deborah Miller to invite Chris Rusbridge to the July 2005 TAG meeting.
8c:	Angela Sasse to present a paper on the usability aspects of security to the April 2005 TAG meeting.
8f:	Malcolm Atkinson to flag up the presentation of the results of the JISC security evaluations as a future e-Science event at NeSC.

PI	Organisation	e-Science Project	Funders	Title	Sister Project	Abstract
Moreau, Professor L	University of Southampton	CONOISE-G/PASOA/VO	EPSRC	SOCA Sister Project: SOuthampton Chicago Activity	Argonne	The Grid is a large scale computer system that is capable of coordinating resources that are not subject to centralised control, whilst using standard, open, general-purpose protocols and interfaces, and delivering non-trivial qualities of service. While a number of projects have been reporting success stories, in which "Grid computing" was applied to highly visible scientific applications, a tremendous number of challenging issues still remain to be investigated in order to address this vision fully. Southampton and Chicago, which both have teams actively involved in multiple Grid and e-Science collaborations, share a remarkable number of common interests in the area of resource discovery, provenance, workflow languages, negotiation and virtual organisations. These different topics are being investigated in various projects in Southampton (myGrid, PASOA, EU Provenance, Conoise-G, Grimoires, Virtual Organisations for e-Science) and Chicago (GriPhyN, iVDGL, Globus). Additionally, the partners bring to the table complementary expertise from different communities: Grid and High Performance Computing, Agent-based computing and Semantic Web and Grid technologies. SOCA, the SOuthampton Chicago Activity, is a sister project that sets up a formal collaboration between Southampton and Chicago in order to investigate some fundamental issues related to Grids. The SOCA project will establish bridges between American and UK communities, but also between Grid and Agent communities, and between Grid and Semantic Grid communities.
Frey, Dr JG	University of Southampton	CombeChem/e-Bank	EPSRC /JISC	The Crystal Grid Network	Reciprocal Net (US)	Grid based interactions with central crystallography facilities have been developed separately in the UK (CombeChem/e-Bank) and the USA (Reciprocal Net, Collaborative Multi-Scale Chemistry Sciences - CMCS). Initial meetings have demonstrated that these projects have many complimentary features in design and architecture. We propose to collaborate on the formation of an international standards and implementation for grid based data collection, monitor & control of laboratory environment and data services for crystallography. We will achieve these aims efficiently by sharing and integrating the complimentary components of the grid services developed for the EPSRC National Crystallography Service (NCS, UK) and the Smart Labs systems with those of Reciprocal Net and CMCS (USA). We aim to similarly integrate the disseminations systems of e-Bank and Reciprocal Net and extend these to the new e-science community in Australia.
Davies, Dr JW	University of Oxford	OeSC, CancerGrid	CP/MRC	International Collaboration in Data Standards and Software Engineering for Cancer Bioinformatics	caBIG	1. Foster and cement links between CancerGrid and the NCICB caBIG project. 2. Extend the Cancergrid-caBIG collaboration to other relevant user-groups within the project and the community it serves, including development staff from the project, trial coordinators, statisticians and software engineers from clinical trials units served by the project, and bioinformaticians working on the project's principle demonstrators. 3. Promote the caBIG project to the cancer and wider bio-medical research communities in the UK, enhancing opportunities for further collaboration, integration and reuse of its approach and software products.
Gavaghan, Dr DJ	University of Oxford	IB	EPSRC	a Truly Global Virtual Research Environment for the Modelling and Simulation of the Heart	Tulane University (US), UCLA (US), UCSD (US)	The primary aim of the Integrative Biology project is the development of the Computational Infrastructure using state of the art technologies (termed "Grid Middleware") required to support post-genomic research in biology and medicine. Integrative or systems approaches to research in biology are evolving very rapidly, driven by a pressing need to understand how the components that make up a biological system interact to allow biological function, such as the beating of the human heart, to emerge, and to express that understanding in quantitative (i.e. mathematical) terms. It is now widely accepted that this can only be achieved by close collaboration between experimentalists, mathematicians, and computer scientists. The IB consortium initially represented a collaboration between 6 leading UK Universities (Oxford, Nottingham, Leeds, UCL, Birmingham and Sheffield), the Bioengineering Institute at the University of Auckland, the Central Laboratories of the Research Councils and the IBM corporation. To test fully the computational or Grid infrastructure that we intend to develop, we are now requesting funding through this proposal to extend the consortium to include three leading heart-modelling groups in the USA led by Professors Natalia Trayanova at Tulane, Andrew McCulloch at UCSD, and Alan Garfinkel at UCLA. This will be achieved through a series of annual visits by UK researchers to the US to install middleware and to train users, and an annual workshop in the UK at which researchers from the US will report on their experiences of using the system and suggest improvements. The final deliverable of the project will be a truly global computational grid infrastructure supporting the research of a leading international research community.
Atkinson, Professor M	University of Edinburgh	OGSA-DAI	CP	DIALOGUE - Data Integration Applications: Linking Organisations to Gain Understanding and Experience	DataCutter (Ohio), dQUOB (Indiana), GridMiner (Vienna), SDSC	Thanks to major investment by the UK e-Science Core Programme, the UK is at the forefront of work that provides convenient access to databases and other data resources via the grid and web services. This has the goal of supporting complex data integration from heterogeneous, independently administered, distributed data resources. Better support for such integration is becoming essential as the scale, diversity and quantity of data resources is growing very rapidly. This rich international environment of data has huge potential for discovery by combining independently collected data. Before this potential knowledge can be harvested it is necessary to develop data access and integration technology that can cope with this complex and ever changing environment of data resources. The ability to combine information conveniently, economically and reliably is vital for science, business, e-Health, e-Government, engineering and many other socially and economically significant activities. It will be an essential component of the UK's and Europe's e-Infrastructure. Building this e-Infrastructure is a major challenge. Each group collects its data in its own way. So data is stored in very different forms. Each group changes its arrangements for storing data, changing the structure, the way in which the things it observes are represented, and its policies for allowing data to be used. Imagine the different ways in which geologists have named rocks, the number of ways they collect samples, the different coordinate systems they use, and even the different ways they talk about time. Then imagine all the modern ways of collecting more data: fixed and mobile arrays of seismophones, satellite observations that use lidar, radar and GPS to monitor earth surface movements, vertically and horizontally. Add in the different ways in which paleobiologists describe species, chemists describe rock, meteorologists describe weather, and so on. Now in an area of great interest, data from all such sources may be collected and combined, as they are in the GEON project in California for the Yellowstone region. New "production engineering" is needed to combine data for a scientist who has an idea and wishes to test that idea by comparing data from many sources. New "flexible production engineering" is needed to continue to supply the scientist with her data as the data sources change. No one group posses the knowledge of all the kinds of data in use, of all the ways in which people want to combine data, and of all the methods we might use in the process. DIALOGUE will gather that expertise, as each partner has specialist knowledge covering a range of data challenges and has already developed methods they can apply. Using the framework of our existing technologies we will plan how we could build the complete "flexible production engineering" system (or at least a major part of it).
Naylor, Professor T	University of Exeter	e-Star	PPARC/CP	Grid Interoperability for Heterogeneous Telescopes Networks	NASA, RAPTOR, MONET (US/SA)	Modern astronomical telescopes are increasingly automated, and communicate via the internet. There are now a small number of networks of such telescopes, and our objective is to be able to unify them into a single, global network that can be accessed either directly by astronomers, or by software "agents" working for them. To achieve this we need to agree the "languages" these telescopes speak and the protocols they adhere to. We are therefore proposing an international workshop and follow-up bilateral visits to create such a network.