Libraries supporting e Science : combining cultures

(1)

Libraries supporting e-Science

---…

combining cultures …

Pauline Simpson

National Oceanography Centre University of Southampton, UK

Digital Libraries à la Carte 2007 Tilburg University

(2)

Overview



_Not

_{about Libraries supporting research}

per se



_e-Science



_{Open Data}



_{Digital Repositories and Open Access}



_{Vision of ‘joined up research’}



_{Issues for e-Science}



_{Combining cultures, connecting people}

–

_{New roles for libraries}

(3)

Whoo are you?

 Librarian

 Documentalist

 Information Specialist  Information Scientist  Information Manager • Information Advisor

 Data Librarian

 Computer Specialist

 Data Manager

 Data Technician

 Data Processor

Anoraks

(4)

Scientific evolution

Digital Libraries à la Carte 2007



Thousand years ago: Thousand years ago:

Experimental Science

- - description of natural phenomenadescription of natural phenomena



Last few hundred years: Last few hundred years:

Theoretical ScienceTheoretical Science

- Newton’s Laws, Maxwell’s Equations …Newton’s Laws, Maxwell’s Equations …



Last few decadesLast few decades::

Computational ScienceComputational Science

- - simulation of complex phenomenasimulation of complex phenomena



Today:Today:

e-Science or Data-centric Science

- unify theory, experiment, and simulation - unify theory, experiment, and simulation

- requires data exploration and data mining- requires data exploration and data mining

(With thanks to Jim Gray)



‘

e-Science’ is a shorthand for a set of technologies to support collaborative networked science

 HPC and Information Management are key technologies to support this e-Science revolution

(5)

(6)

e-Science – not only



e-Science - data driven

(Natural and Physical Sciences)



e-Research

(includes e-Science and Arts & Humanities now joining in – ACLS – report 2004 – great opportunity to bring new analytic and interpretive power to

humanities and social science)



Cyberinfrastructure

(

NSF : Revolutionizing science and engineering through CyberInfrastructure, 2003 (Atkins Report)

 _{describes the new research environments in which advanced computational,}

collaborative, data acquisition and management services are available to researchers through high- performance networks … more than just

hardware

and software, more than bigger computer boxes and wider network wires.

 It is also a set of supporting services made available to researchers by their home

institutions as well as through federations of institutions and national and international disciplinary programs.

(7)

Key elements of e-Infrastructure

• _{Research Network}

• _{International}

authentication and

authorisation

• _{OS Middleware}

Engineering and Software

Repository

• _{Access to international}

Data Sets and Publications

• _{Portals and Discovery}

Services

• _{Digital Curation and}

Preservation

• _{Remote Access to}

Large Scale facilities

• _{Interoperable}

Institutional and

Thematic Repositories

• _{Support for}

International

Standards

• _{Tools and Services to}

support collaboration

• _{International Grid}

(8)

Early Vision of the Grid

J.C.R Licklider -

The Computer as a Communication Device

(1968)

predicted the use of computer networks to support

communities of common interest and collaboration without regard to location. Foretold of graphical computing, point-and-click interfaces, digital libraries, e-commerce, online banking, and software that would exist on a network and migrate

wherever it was needed.

“Lick had this concept of the intergalactic network which he

believed was everybody could use computers anywhere and get at data anywhere in the world. …, but he had the same concept – all of the stuff linked together throughout the world, that you can use a remote computer, get data from a remote computer, or use lots of computers in your job. The vision was really Lick’s

originally.”

Larry Roberts – Principal Architect of the ARPANET

(9)

1990’s The Web

• Tim Berners-Lee developed the Web at CERN as a tool for

exchanging information between the partners in physics

collaborations

• It was the international particle physics community who

first embraced the Web

• The first Web Site in the USA was a link to the SLAC

Library Catalogue (Stanford Linear Accelerator Center)

(10)

Web+

 Scientists developing collaboration technologies that go far beyond the capabilities of the Web

– To use remote computing resources

– To integrate, federate and analyse information from many disparate, distributed, data resources

– To access and control remote experimental equipment

 Capability to access, move, manipulate and mine data is the central requirement of these new collaborative science applications

– Data held in file or database repositories

– Data generated by accelerator or telescopes

– Data gathered from mobile sensor networks

 Grid = set of services for sharing computing power and data storage. Use middleware to handle the complex authentication and scheduling, linking together applications, devices and computing resources as seamlessly as possible

(11)

Web 2.0



Web as a platform

– Regardless of operating system – open network enables collaboration and communication – serves + involves

– Social software all relevant to scientific research

• File sharing - much used documents, videos, slides

• Tagging (for later retrieval)

• Folksonomies (informal ontologies developed by community)

• Virtual Worlds eg Second Life - education, conferences

• Wikis (OpenWetWare), Blogs (Useful Chemistry)etc

– Library 2.0 (Web 2.0 + Library) – reinvention of the Library?

– Scientific Web?

• Web invented for sharing scientific communication – relatively few scientists have embraced the potential

• Barriers: social, psychological, technical

(12)

Some e-Science projects



Particle Physics

–

_{global sharing of data and computation}



Astronomy

–

_{‘Virtual Observatory’ for multi-wavelength}

astrophysics



Chemistry

–

_{remote control of equipment and electronic logbooks}



_{Bioinformatics}

–

_{data integration, knowledge discovery and workflow}



_Healthcare

–

_{sharing normalized mammograms}



_Environment

–

_{climate modelling}

–

_{Undersea sensors}

(13)

climateprediction.net

Since September 2003:

61,000 registered participants in 130 countries have…

(14)

(15)

(16)

Data Workbench

(17)

(18)

The ‘Data Deluge’ – science is turning to

e-Science

 _{In next 5 years e-Science projects will produce more}

scientific data than has been collected in the whole of human history

 _{Some normalizations}

:



The Bible = 5 Megabytes



Annual refereed papers = 1 Terabyte



Library of Congress = 20 Terabytes



Internet Archive (1996 – 2002) = 100

Terabytes

 _{New high throughput devices, sensors and surveys}

 In terms of bytes - moving beyond giga 109 through tera

1012 onto peta 1015 and onto exabytes 1018

(petabyte (PB) 1015 = 10005 = quadrillion bytes)

(19)

Data-centric 2020 vision resulting from

Microsoft ‘Towards 2020 Science’

(2006) ………..

Nature 440, (23 March 2006) |

Data gold-mine

‘Multidisciplinary databases also provide a rich environment for

performing science; that is, a scientist may collect new data,

combine them with data from other archives, and ultimately

deposit the summary data back into a common archive. Many

scientists no longer 'do' experiments the old-fashioned way.

Instead they 'mine' available databases, looking for new patterns

and discoveries, without ever picking up a pipette.’

(20)

Data Loss



_{e-Science is about improving the use and reuse of}

research data



_{Huge amounts of research data cease to exist each}

year.

– Hardware, software obsolescence

– Represents the loss of expensive intellectual resources; a huge opportunity cost for comparative and longitudinal research



_{Unintentional data loss in the sciences due to:}

– lack of incentives to maintain them, or due to neglect (benign and otherwise – forget where or what it is!

– personal computers, Web sites, blogs, wikis, e-mails, digital photo and film etc.

(21)

Digital Preservation



_“

_{Digital information lasts forever, or five}

years, whichever comes first

”

Jeff Rothenberg,

RAND 2001

 Medium Practical Physical Lifetime Av. Time to obsolete

Optical (CD 5-59 years 5 years

Digital tape 2-30 years 5 years Magnetic disk 5-10 years 5 years

(22)

Preservation - Trusted Repositories



Preservation

– UK Digital Curation Centre: advice, tools & services

• RepInfo Registry – representation information adding meaning to data for preservation

– EU CASPAR Integrated Project http://www.casparpreserves.info/pages/1/index.htm

– EU Task Force on the Permanent Access to the Records of

Science http://tfpa.kb.nl/

– EU projects DPE and PLANETS

(leverage library and archive experience)



_{Long-term access: trust, responsibility, policy}

– Trusted DR Audit Checklist for Certification Draft -

Research Libraries Group-NARA Taskforce 2005

• Defined criteria under 4 categories

– Organisation

– Functions, processes & procedures

– Designated community & usability

– Technologies & technical infrastructure

• Can these concepts be extended to data repositories?

(23)

Key drivers for e-Science



_{Access to Large Scale Facilities}

and Data Repositories

– eg CERN, EBI , etc



_{Need for production quality, open}

source versions of open standard

GRID middleware

–

_eg

. OMI, NMI, C-Omega



_{Imminent ‘date deluge’ :}

Particle physics, astronomy, bioinformatics



_{Data Loss/Preservation}



_{Open Access movement}

(24)

Open Data

(25)

(26)

Open Access to Data – European Research

Council

 It is the firm intention of the ERC Scientific Council to issue specific guidelines for the mandatory deposit in open access repositories of research results -- that is, publications, data

and primary materials -- obtained thanks to ERC grants, as

soon as pertinent repositories become operational.

 The ERC Scientific Council moreover hopes that research

funders across Europe will join forces in establishing common open-access rules and in building European open access

repositories that will help make these rules operational.

 To facilitate this process for EU funded research, it

recommends that the European Commission sets up a task force including representatives from the various FP7

programmes … to develop an operational FP7 policy on open access by the end of 2007 …

http://ec.europa.eu/erc

(27)

Organization for Economic Co-operation and

Development (OECD)

 _{Jan 2004 : Promoting Access to Public Research Data for} Scientific, Economic, and Social

Development

 Open access to, and unrestricted use of, data promotes scientific progress and facilitates the training of researchers

 Open access will maximise the value derived from public investments in data collection efforts

 The risk that undue restrictions on access to and use of research data from public funding could diminish the quality and efficiency of scientific

research and innovation

 _{Dec 2006 : Recommendation of the Council concerning} Access to Research Data from Public

Funding

– “each Member country, to develop policies and good practices related to the accessibility, use and management of research data”

(28)

Open Access to Data following OA for

Publications



_{1990’s – Subject Repositories (high energy physics,}

economics, mathematics etc).



_{HEP (ArXiv) v.successful}



_{Economics (RePEc) - successful}



_{Limited success otherwise}



_{1994 ‘Subversive proposal’ (Harnad)}



_{2000’s - Institutional Repositories}



_{Powered by Project funding, driven by the}

Information Community (Libraries)

 Libraries already supporting e-Science by

development of OA digital repositories of research publications – providing global and immediate

discovery and access to new research

(29)

Repository Growth

112 Repositories in 2002

2007

Over 900/1400 repositories

Over 20 Different Software Systems

ROAR

-Registry of Open Access

Repositories

Directory of Open Access Repositories(DOAR)

(30)

(31)

Repository Landscape

• Subject - arXiv, Cogprints, RePEc,

• Institutional – Universities, Research Institutes -Southampton, Glasgow, Nottingham (SHERPA), Max Planck

• National - DARE (all universities in the Netherlands), Scotland IRIS),

• National / Subject - OceanDocsAfrica

• International - Internet Archive ‘Universal’, OAIster (Harvester)

• Regional - White Rose UK

• Consortia - SHERPA-LEAP (London E-prints Access Project)

• Funding Agency – NIH (PubMed), Wellcome Trust (UK PubMed), NERC (NORA)

• Project - Public Knowledge Project EPrint Archive

• Conference - 11th Joint Symposium on Neural Computation, May 2004

• Personal – Peer to peer

• Media Type - VCILT Learning Objects Repository, NTDL (Theses), Museum Objects, Exhibitions

• Publisher – Journal archives

• Data Repositories - UK Data Archive; World Data Centre System; National

(32)

Support - Declarations on Open Access

 _{Berlin Declaration in Support of Open Access 2003}

(

₅₀ + signatories)

Germany: Fraunhofer Society, Wissenschaftsrat, HRK, Max Planck Society Leibniz Association, Helmholtz Association, German

Research

Foundation, Deutscher Bibliotheksverband France : CNRS, INSERM

Austria : FWF Der Wissenschaftsfonds

Belgium : Fonds voor Wetenschappelijk Onderzoek – Vlaanderen) Greece : National Hellenic Research Foundation

 The IFLA Statement on Open Access to Scholarly Literature

and Research Documentation 2004

http://www.ifla.org/V/cdoc/open-access04.html

 India, Australia, China, Africa, USA …

 Scotland (2005) 16 Universities and Research Orgs  Russell Group (UK Universities) 2005

 Buenos Aires, British Columbia, Bethesda Statement (2003)

 Budapest Open Access Initiative Feb 2002 (Soros Open Society Institute)

(33)

Support - Mandates for Open Access



_{Real mandates:}

– _{Wellcome Trust}

– _{RCUK (Research Councils UK)}

– _{Universities – Southampton UK, Minho Portugal, NIT India, CERN …}



_{Proposed mandates: public funders}

– DFG, Germany; FWF, Austria; DARE network; Finland; USA; Sweden (?)

– Canada, Australia, India, S.Africa, Ukraine etc

 NIH: Strengthening now very likely, Require not request

 _{CURES Act}_:_{6-month delay to OA permitted but must deposit at acceptance}  _FRPAA_:_{Mandatory deposit: all research funded by the}

largest agencies Federal Research Public Access Act  EU : Petition for guaranteed public access to publicly-funded

research results, Feb 2007

(34)

Open Access Research Repositories -

developing

• There will be many types of repository software and more powerful interoperability

protocols such as OAI-ORE

(need more than OAI-PMH to enable sharing and reuse)

• Thematic and Institutional Repositories contain not only full text versions of research papers but also ‘grey’

literature such as technical reports and theses

• In addition, repositories in the future will also contain data, images and software

(35)

UK JISC Projects –linking text and data

Source-to-Output Repositories

CITATION, LOCATION, And DEPOSITION IN DISCIPLINE &

INSTITUTIONAL

REPOSITORIES

New JISC Call for data

related projects 2007

(36)



_{Publications, working papers, primary data,}

audiovisual, images



Hardware in research labs will automatically

deposit experimental data



Desktop tools will deposit content



_{Rich data flow between networks of repositories}



Rich data flows between repositories and other

components in information landscape

(37)

Federation of Digital Repositories

• Global

• Inter-disciplinary

• Cross-sectoral

• Multiple format types

• Data, publications, images… • e-Research Framework

• Defining common services + domain-specific services + repository services

From Andy Powell: http://www.ukoln.ac.uk/distributed-systems/jisc-ie/arch/presentations/ jiie-jcs-2005

fusion layer ‘repository federator’

repository repository repository repository repository

portal portal portal portal portal

heterogeneous - metadata formats, content formats, identifiers, packaging standards

homogeneous - metadata formats, content formats, identifiers, packaging standards

(38)

(39)

(40)

Issues for e-Science



Macro and micro issues are

similar for

both text and data

repositories:



_{IPR and Licenses **}



Distributed over many

researchers



_{Over National boundaries}



Lack of awareness amongst

researchers



Cultural roots and resistance to

change



_Funding



Policies



_Standards



Interoperability



_Vocabularies

**

Necessary to understand science

practices:

technical social and communicative

structure in order to adapt licensing

solutions to the practice of

e-Science

.

Arzberger, P. et al Science 2004 303. 1777 – 1778. DOI:

10.1126/science.1095958 Research Issues: information retrieval, information modelling, ontologies, authentication, systems interoperability, and policy issues

(41)

Combining Cultures



_{NSF Report “Long lived digital}

data collections”. 2005

– “Data scientist” - hybrid skills



_{Facilitate collaboration}

– “Multidisciplinary teams: computer scientists, domain scientists, digital library experts, statisticians/modellers

– Lessons learnt: e-Science Human

Factors Audit Report (to be published

Many of the same research issues

that the international digital

library

(42)

Combining Cultures

NSF Report :

 “It is timely to seriously consider the role that digital libraries can and should play in this emerging e-Science computational

infrastructure”.

 _{“Bringing the digital library and the emerging scientific}

infrastructure worlds together can lay the foundation for providing truly integrated support for the entire process of science, from

formulation of research questions to the publication of the outcomes”.

 “Specifically, the e-Science and digital libraries research communities need to work together to identify the

potential contributions of each of these communities for supporting the conduct of science and to articulate a

shared research agenda”

(43)

Calls for Combining Cultures for e-Science

needs

– EU Framework 7 – e-Research Infrastructure development

– UK – JISC – Report on future requirements for curation and

preservation

– Australia DEST – e-Infrastructures Reflection Group

includes CAUL (Council of Australian University Librarians) member - Interim Report

http://www.dest.gov.au/sectors/research_sector/policies_issues_reviews/ key_issues/e_research_consult/interim_report.htm

– UK –CURL/SCONUL Joint e-Research Task Force (2006)

– USA – ARL Libraries and Changing Research Practices

(44)

CURL/SCONUL Joint e-Research Task Force

Nov 2006

1. To raise awareness and understanding of the issues associated

with support of e-research in CURL and SCONUL member libraries and to stimulate discussion about them at institutional level.

2. To position CURL and SCONUL member libraries’ staffs to engage

with their local e-research stakeholders and to encourage them to make appropriate inputs at the research proposal stage.

3. To identify skills gaps in relation to support of e-research and to assist member libraries in addressing them.

4. To work with other e-research stakeholders, including the DCC, RLN and BL, to ensure that information management to support e-research is a high priority for future investment by funders.

5. To advise the CURL Board and the SCONUL Executive Board on

matters relating to the support of e-research.

6. To monitor, and report on, the Group’s progress against an action plan agreed annually by the CURL Board and SCONUL Executive Board.

(45)

Has your library engaged with the

e-Research agenda?



_{Management of the large datasets is likely}

area of involvement.



_{Why are libraries not already involved:}

–

_{Lack of foresight by librarians?}

–

_{No e-Science funds for development of data}

management in libraries – no call for projects

–

_{No Customer demand for data curation}

(46)

WP 1 - Information and awareness

• Recruit network of e-research liaison contacts in HE library & information services; establish JISCmail list

• Survey of research activity, research support requirements, and e-research support work within HEIs (coordinate with WP2 needs analysis)

• Survey of the policy and practice of research funders in relation to data curation

• Disciplinary mapping of existing data curation services and gaps in provision

• Training & development needs analysis (link with WP1 survey activity on researcher support requirements)

• Design, commissioning and delivery of training and development events for HEI library & information services staff

• Maintain awareness of funding and bidding opportunities for the eRTF • Lead on bid drafting

• (with WP1) identify potential case studies/exemplar projects for development with DCC

WP 2 - Workforce development

(47)



_{Engaging ARL members in the development of new roles for}

libraries as

e- Science infrastructure and service needs emerge at research

institutions and

promoting the contributions of research libraries in this arena.



_{Identify the skills needed as information professionals move into}

the emerging

(48)

ARL Workshop Recommendations

 NSF should fund projects in which university research libraries develop

deep archives of irreplaceable data, assuring descriptions of these data

at a minimal level (floor, not ceiling) and facilitating discovery and access to these data, according to prevailing community standards

 NSF should partner with IMLS to train information and library professionals (extant and future) to work more credibly and knowledgably on data curation as members of research teams

 NSF should foster the training and development of a new workforce in data science

 Promote new curricula

 Develop new programs

 Link to training of domain scientists and information/library scientists

ARL Workshop on New

(49)

Shared Goals and Responsibilities

NSF –

(50)

NSF REPORT NSB-05-40, Long-Lived Digital Data

Collections Enabling Research and

Education in the 21st Century

(51)

Combining cultures, connecting people

 Librarian

 Documentalist

 Information Specialist  Information Scientist  Information Manager • Information Advisor

 Data Librarian

 Computer Specialist

 Data Scientist

 Data Manager

 Data Technician

Anoraks

What is a data scientist?

(52)

Data Scientist

• New skills requirements:

• interdisciplinary

• quantitative

• data curation

• Integrate data management within the LIS curriculum

• Various approaches to develop and obtain digital curation skills

• Skills are there but often in discrete communities: we need to bring communities together

• Integration within the

curriculum: undergraduate students, library &

information science, archival studies, computer science

• Provide recognition and a career path for emerging ‘data scientists’



There must be a blurring of the boundaries between

previously well

defined silos that existed between information

managers and

(53)

Connecting People – how to

 Encouraging partnerships - Inter-institutional partnerships

 Institutional management support

 Competencies – shared and developed

 Common issues – authentication, metadata, ontologies, standards, IPR, licenses

 New curricula - CPE

 Share experience of Institutional repositories- libraries

fundamentally transforming research publication practices and scholarly communication forerunner of e-Science

 Libraries involved in the research underlying the design of e-Science. Is there new research on Digital Libraries over the Grid

 Funding Agencies – JISC, NSF, EU etc

– define project members from library and data communities

– promote the necessary international dialog between

(54)

Role for Libraries in digital data

universe

• Data as primary source material – Libraries :

– Will not be primary providers of large scale storage infrastructure required

– Will not provide the specialized tools to work with data

– Will not provide the detailed information about the data

– Unlikely to provide the solutions to digital preservation because of cost

• _{Can contribute library practices}

:

• Collection policies (appraisal,

selection, weeding, destruction etc)

• Data clean up, normalisation, description and submission to repositories

• Data Citation

• Curation and Preservation

• Collaboration with researchers re scholarly communication ,deposit, education and training

• Innovative discovery and presentation mechanisms

• Data as part of ‘enhanced

publications’ – Libraries:

– Well positioned to define standards for

• Taxonomies and ontologies

(for complex publications that include data)

• Persistent identifiers • Consistent description

practices

• Data structuring conventions • Interoperability protocols for

searching and retrieval

– Well positioned to exploit IR experiences

(55)

Role of Digital Libraries - IRs



_{Institutional repository is a key component of}

e-infrastructure

– Mostly in library domain

– Access and preservation

– Digitization – data archaeology

– Interoperable with departmental, national, subject repositories



_{Data Curation}

– Creation, metadata, preservation institutional intellectual assets

• but disparate data types and ontologies



Training Provision

– Research methods training for researchers

• Data creation, documentation, management



_{Advocacy, policy setting}

– Cross disciplinary approach to key issues

• Expand OA agenda

(56)

Roles for Libraries

 Institutional repositories accept ‘small’ datasets (size or

subject outside remit of Data Repositories). Data deposited in IR until accepted by data repository

 Development of Regional or Discipline Repositories alongside IRs (singly or consortia) . Research libraries a natural home for content curation, (with funding)

 Mapping of commonalities (eg metadata) across disciplines, maintaining ready interoperability

 Management of metadata throughout a research project

 Address conditional and role-based access requirements for scientific data

 Support e-Science interface functions for local users

 Adding Value: linking, annotation, visualization

 Libraries and researchers can add value by creating ‘e-Science

Mashups’ - data needs to be re-used in multiple ways, on

multiple occasions and at multiple locations (reuse, remix)

(57)

The “mash-up”

Data from

FAO, WHO

+

Google

Earth

(58)

Role for Libraries – build on the

strengths

 Serving the needs of the scientific community

 Systematically managing and making accessible information from heterogeneous sources

– Metadata, discovery mechanisms, portals, VRE , “Science World”

– Publication and Citation

– Selection and use of tools and resources

– Digitization of legacy content

– access management, copyright, IPR, Licenses

– Curation and Preservation advice

 Provide specialist assistance to end-users

– expertise in user services and training

Exploit strengths in designing and implementing

innovative and useful e-Science information infrastructure.

 Reduce the risk of "re-inventing the wheel".

(59)

S

W

O

T

(60)

Facing the future



Build Institutional Repositories



Develop leadership & vision for e-Research engagement

– Web 3.0?, Semantic web - publishing?



Review organisational structures

– Extend & re-profile the Faculty/Subject/Reference Librarian role? – Closer collaboration with Computing Services, and Data Services?