Languages
Pages
Legal
Rusbridge, C. and Burnhill, P. and Ross, S. and Buneman, P. and Giaretta, D. and Lyon, L. and Atkinson, M. (2005) The Digital Curation Centre: a vision for digital curation. In: Local to Global Data Interoperability - Challenges and Technologies: 20th-24th June 2005, Sardinia, Italy. IEEE, Piscataway, N.J., USA, pp. 31-41. ISBN 0780392280
http://eprints.gla.ac.uk/33612/ Deposited on: 26 July 2010
Enlighten – Research publications by members of the University of Glasgow http://eprints.gla.ac.uk
The Digital Curation Centre: A Vision for Digital Curation
Chris Rusbridge Peter Burnhill Seamus RossDigital Curation Centre
(DCC), University ofEdinburgh
Digital Curation Centre(DCC), & EDINA, University
Digital Curation Centre(DCC), & HATII, University of
Peter Buneman David Giaretta Liz LyonDigital Curation Centre
(DCC) & School of InfomaticsUniversity of Edinburgh
Digital Curation Centre(DCC), & Council for theCentral Laboratory of the
Research Councils (CCLRC)[email protected]
Digital Curation Centre(DCC), & [email protected]
Malcolm AtkinsonNational e-Science Centre,
Department of Computing Science, University of Glasgow &School of Informatics, University of Edinburgh
AbstractWe describe the aims and aspirations for the Digital
Curation Centre (DCC), the UK response to therealisation that digital information is both essentialand fragile. We recognise the equivalence ofpreservation as "interoperability with the future",asserting that digital curation is concerned with'communication across time'. We see the DCC ashaving relevance for present day data curation and forcontinuing data access for generations to come. Wedescribe the structure and plans of the DCC, designedto support these aspirations and based on a view ofworld class research being developed into curationservices, all of which are underpinned by outreach tothe broadest community.
Introduction
The foundation of the Digital Curation Centre(DCC) reflects the belief that long term stewardship ofdigital assets is the responsibility of everyone in thedigital information value chain [15]. The maintenance,usability and survival of digital resources depends onregular planned interventions; care needs to be taken at
conception, at creation, during use, and as usetransitions to lower levels. It may be tempting to viewe-Science or Cyber-infrastructure as concerning huge,often distributed databases in current use, quite distinctfrom viewing digital preservation as concerned withmore common, much smaller files, ensuring they areavailable for the future. The Open ArchivalInformation System (OAIS) model's [5] emphasis oningest, followed by preservation management, and thenlater acts of dissemination, does tend to suggest amodel where resources are put away in safe keeping forthe future. In reality, the DCC views digital curation asa continuum of activities, supporting the requirementsfor both current and future use [21]. Huge datasets incurrent use need action now to ensure their futureutility, while digital preservation has always deprecated'dark' inaccessible archives.
The long term value of data rests in their potential asevidence, their reuse possibilities, and their role infacilitating compliance and in ameliorating risk.Curation, the active management and care of data is thekey to realising that potential. As scholarly researchand scientific study becomes increasingly driven by theanalysis of data, long term access to these data is
1
crucial in enabling the verification of scientificdiscovery and to providing a data platform for futureresearch.
This view of curation embraces and goes beyondthat of enhanced present-day re-use, and of archivalresponsibility, to embrace stewardship that adds valuethrough the provision of context and linkage: placingemphasis on publishing data in ways that ease re-useand promoting accountability and integration. Contextand linkage are terms rich in intention, withimplications for metadata and interoperability.Resource discovery and retrieval requires mark-up withtime/place referencing as well as subject descriptionand linkage to discipline-based ontology.
Only by promoting the ideas that underpin digitalcuration from the conception and creation of our digitalassets until long after they have passed out of theirprimary usefulness can we claim to have succeeded.The Digital Curation Centre, through its organisation,emphases and practical activities closely reflects theseideals and it aims to catalyse action in innovativeresearch, development, service delivery, and outreach.Its primary aims are:• to promote an understanding of the need for digital
curation among the communities of scientists andscholars;
• to provide services to facilitate digital curation; • to share knowledge of digital curation among the
many disciplines for which it is essential; • to develop technology in support of digital curation;
and,• to conduct long-term research into all aspects of
digital curation The DCC has been funded for three years in the first
instance by the UK Joint Information SystemsCommittee (JISC) [24] and the UK e-Science CoreProgramme of the Engineering and Physical SciencesResearch Council (EPSRC) [19]. Three quarters of theproject's funding was awarded by the JISC from March2004 with the remaining 25% coming on stream at thebeginning of September later that year. By workingwith other practitioners and agencies the DCC willsupport UK institutions to store, manage and preservedigital assets to ensure their enhancement and theirusability over the long term [26]. The DCC is thenational focus for research into curation issues and topromote expertise and good practice, both nationallyand internationally, in the management of all researchoutputs in digital format. The drivers which led to theestablishment of the DCC are an increasing awarenesswithin the scientific and research community thatdigital assets are reusable, that access to them isessential if contemporary scholarship is to be verifiable
and reproducible, and that for many reasons rangingfrom degradation of media to metadata drift and totechnological obsolescence they are fragile.
Engaging the community
The DCC represents a collaborative effort led by aconsortium of four institutions, each bringing diverseexperience in a range of Digital Curation areas. Itbrings together organisations across three Universitiesand a research council. Led by the University ofEdinburgh [35], which hosts the School of Informatics[7,31], the National eScience Centre (NeSC) [27], theEDINA national data centre [17], the AHRC Centre forthe Studies in Intellectual Property and TechnologyLaw [2], the DCC consortium includes HATII [22] atthe University of Glasgow [36], UKOLN [32] at theUniversity of Bath [34], and the Council for the CentralLaboratory of the Research Councils (CCLRC) [6].Given the broadness and pervasiveness of the digitalcuration challenge the core partners recognise that asustainable contribution can only be made ifwidespread activity can be leveraged. To ensure thatthis happens the partners have established mechanismsto support the development of a network of associates.
Engaging the community and creating conduitsbetween our different activities with other research,development, services and user communities isessential if the DCC is to deliver the catalytic impactits funders envisaged in the call for expression ofinterest in JISC Circular 6/03, issued in September2003 [25]. It is crucial for the Centre's success that itbenefits from the diversity of expertise available withinthe UK digital curation community, and that thebroadest range of communities needing digital curationsupport and services are engaged in its activities. Thesedifferent communities should help steer the DCC'sresearch programme, contribute to its communitybuilding events, and benefit from the advisory servicesthat it operates. Science and scholarship areinternational in nature, and best practice in digitalcuration must reflect that. This is immediately obviousin such fields as astronomy, physics andbioinformatics, where data are increasingly created andcurated by multi-national consortia. This requiresinternational collaboration for shared development andconsensus on methodology and standards. The demandfor digital preservation technologies comes from andwill be met by a wider community than academia;therefore, the DCC must engage with the industrial andbusiness communities. In building such relationshipsthe DCC will create outlets for the technologies andmethods it develops and benefit from access tocommercial developments.
2
DCC activity domains
The DCC activities have been separated into fourkey task areas, with an umbrella management groupoverseeing and coordinating the work of each. The fourmain areas are Research, Development, Services, andOutreach. Each activity domain has its own workprogramme. Many elements of these programmes canbe delivered independently of the work being done inother domains, but the DCC will have proven itselfmost successful if it proves possible for us to movefrom concept through research and development toservices and outreach. We hope that the outputs ofindividual task areas will benefit from inputs fromcontributions by members of the Associates Network.These contributions may come as the sharing ofexpertise, donations of technology, methodologies, orapplications, and collaboration in research,development and service provision. Inter-task-arearelationships are facilitated by the DCC's distributednature, with representatives of each activity domaindistributed across the four partner institutions. Thecommunication infrastructure required to enable suchcollaboration and interaction is well established, andthe Consortium relies upon a range of technologies (e.g.online fora, email, conference calls, and the AccessGrid [1]) to allow the straightforward exchange of ideasand results. By driving our activities from a solidresearch footing into innovative and service-leddevelopment work we can meet our numerous serviceobjectives and ensure an effective and dynamicoutreach programme. Since the relationships betweenactivities are non-linear and bidirectional in natureevery task area can benefit from experiences acquiredthroughout the project's full extent and beyond.
Task Area Inter-relationships
Curation is the key
Since its conception, the DCC has assumed astructure representative of an overall vision of DigitalCuration. In order for one's digital curation endeavoursto be successful, and consequently for a DigitalCuration Centre to be successful, it is essential that aset of core activities are undertaken. The task areainfrastructure and their associated work areas aredesigned to facilitate the completion of these activities.The DCC is perceived as a potentially long termproject. Research being undertaken now might notcome into fruition for some years, by which time it willbe possible to trace a clear development, service andoutreach path through which to disseminate the results.The DCC has a key role to play in developing andpromulgating the changes in community practice thatare needed if our research communities are to avoid thedecay of our digital research and scholarship heritage.
Digital Curation itself is the active management ofdata over the life-cycle of scholarly and scientificinterest; it is the key to reproducibility and re-use.Metadata for resource discovery and retrieval areimportant, with mark-up on time/place referencing aswell as subject description and linkage to discipline-based ontologies providing key research foci. Specialemphasis is required on the descriptive informationthat allows effective re-analysis of datasets of scientificand scholarly significance, and re-use in new andunexpected contexts, e.g. e-Learning or history ofscience. The demands for linkage to the two furtherdomains of scholarly communication and e-Learningmust also be understood. The Centre is pluralistic: itvalues and works with the many and diverse culturesthat span the scholarly and scientific communities, andit will seek to value and understand the differentparadigms and methodologies. It aims to address bothgeneric and disciplinary perspectives.
Science and scholarship, more generally, cut acrossdisciplinary boundaries. So too does digital curation:appreciation of differences between disciplines is anessential aspect for understanding and consensusbuilding. An open and creative culture is necessary tofoster the flow of ideas between research and practice,provoking research with practical challenges andinforming practice from theory and experiment. Thislays the grounds for leadership and advocacy, forcontinuing professional development, for mattersrequiring mediation, for building consensus, and forsuccessfully enabling the adoption of commonstandards. Use of open standards enables considerablythe re-use and re-purposing by avoiding dependencies
3
of platform and application software.Fundamental to the success of digital curation is an
understanding of the ubiquity of its relationship withdigital information. Even before digital content isconceived, and even after it has fulfilled its primaryusefulness, digital curation activities must beundertaken and promoted if digital materials are toremain viable. Curation is not a box to be ticked on amanifest or a single process through which data passes.It is an ubiquitous endeavour that should characteriseall other interactions with and manipulations of digitalcontent.
The necessity to share access to digital researchresources has long been a driver within the sciences butpressure for action is now keenest within the physicaland life sciences. As e-Science has become increasinglydata driven such problems as scale, complexity,distribution, provenance and integration of data havebecome increasingly prominent. The 'document' (libraryand archive) tradition has an even longer claim onshared access, now augmented by pervasive internetaccess to digital objects; a foundation for globalcollaboration in the provision and use of digitalresources as evidence in research. The DCC willevaluate and select best practices adopted by datalibraries, data archives and data centres, promotingthose it finds most effective and efficient. It will drawupon and enhance well-tested principles developed bythe research library, records management and archive
traditions as well.
Research
The DCC is research led. It has three main goals:
• To draw together the various functions ofcuration, from the traditional archival functionsto the maintenance and publication of evolvingknowledge as seen in scientific databases.
• To conduct research in areas already identifiedby the partners as crucial to digital curation.
• To institute two-way conduits between researchand service in which real problems can be drawnto the attention of researchers and the products ofresearch can be tested in practice.
A range of topics has been chosen, reflecting thesegoals. Each has in common the fundamental nature oftheir implicit problems, offering a likelihood that theresearch generated will be highly visible. In addition,every topic within the agenda will yield results that canbe immediately exploited and investigated furtherthrough the subsequent work by the development andservices teams. Finally, each of the initial researchtopics is relevant to the partner institutions and a broadcross section of the network of associates[9]. Aresearch committee meets regularly with the task of
4
identifying emerging research opportunities. Thiscommittee is outward-looking, and partners andmembers of the network of associates feed ideas andneeds into its discussions. This should keep ourresearch vital and relevant.
There is much preservation research going onoutside the UK with which we are engaged. There areresearchers from the industrial sector who wouldbenefit from taking time out and participating in theDCC research programme. These individuals bringfresh perspectives and create a conduit between theresearch laboratories and theDCC research community. To enable collaboration andcommunication with these communities the DCC hasdeveloped a visiting scholars programme.
The Research Agenda currently focuses on:
Data integration and publishingSpecial emphasis is given here on integration
techniques in the context of digital preservationmetadata. Current work investigates publishing curateddatabases or parts of curated databases that otherresearchers or scientists may bring into or integratewith their own "research databases". This includesdeveloping techniques to pull records out of differentlystructured relational databases and translate them intoXML documents that share the same structure and canbe more easily integrated.
AnnotationAnnotation plays a fundamental role in scientific and
scholarly work. Researchers use annotation as amechanism to enhance the description or interpretationfrom trusted sources that may inform furtherinterpretation of data drawn or research with differentcommunities wishing to annotate different types ofdigital data (structured text, audio, images, video) in avariety of ways. Our understanding of where and howto "attach" annotations of various types to base data,and how to let others search (query), view, or trackannotations across applications, researchers, time, andmigrations remains sketchy. A near complete scopingreport provides an exploratory literature review ofprevious research concerning databases and annotationand defines a range of research topics that the DCCmight pursue in this area. Among the problems worthyof investigating is the extent to which forms ofannotation can be predicted when metadata formats ordatabases are designed.
Archiving and appraisal Since digital curation is an active and costly process
the long term preservation of most digital objects isunlikely to happen. Objects which are to be preserved
will need to be selected for investment and care.Traditional archiving methods include some form ofappraisal to determine what data to preserve and whento preserve it; these methods must also be in place fordigital data. Current research also concerns techniquesto appraise and archive efficiently large databases thatundergo regular change, such as genome or proteindatabases that grow rapidly as biological researchmoves forward.
Provenance and data quality The value of digital curation is lessened if the
evidence as to the origin and integrity of the data is lostor unknown. Current DCC work in this area includesdeveloping formal models to support the description ofthe state of the problem and to extend ourunderstanding of the problems that arise when data arecopied from one database to another or when trackingwhere the data has come from that has not beenadequately documented (its provenance not suitablyclear). As these formal models emerge they will assistthose developing software tools and standards fortracking, exchanging and managing the provenance ofdata as it is transferred between databases.
Metadata extraction Current digital preservation processes require
extensive human intervention in selection, validation,description, assigning unique identifiers, datamanagement, migration, and delivery of desiredcontent. The investment of human labour currentlyrequired does not scale to the number or complexity ofthe digital content that needs to be curated. A case inpoint is preservation metadata, which is an essentialpart of the information infrastructure necessary tosupport all the processes in digital preservation. Tomake the preservation of digital objects more viable,automatic or semi-automated creation and authoring ofthe metadata must be addressed. The challenge is thedevelopment of methods and approaches for creationof metadata supporting the understandability of digitalobjects.
Legal issues Legislation and the variations in legal frameworks
across national boundaries pose obstacles to digitalcuration, but also create opportunities. DCC ledinvestigations include examination of the impact oflicensing, intellectual property rights, issuessurrounding digital rights in databases, and theimplications for how courts view digital informationand the ways these views can be shaped by how digitalmaterials are curated.
5
Networks of trusted repositories With the diversity in repository implementations
currently available the issues involved in theircoordination and integration and applicability toparticular resource classes require investigation.Attempts have been made to identify the necessaryattributes of trusted repositories, and this has promotedthe recognition that research is needed to address howmultiple repositories might co-operate within a globalcuration network. Co-operation offers many potentialbenefits. For example, successful networks may enablerepositories to co-operate on the selection of contentand on the development of technical approaches tocuration, thereby helping to reduce costs andduplication of effort. They may also be able to helpreduce the risks of institutional impermanence bysupporting the distribution and replication of dataacross multiple institutions.
Economic cost-benefit analysisCuration activities appear to be expensive. This view
must be balanced, however, with the economic andsocial costs of losing digital assets, some of which maybe unique and impossible to recreate. Research in thisarea will use economic analysis and modellingtechniques to explore the cost-benefits of digitalcuration.
Performance and optimisationThe data avalanche hitting many scientific
disciplines imposes new responsibilities on datacreators, users and curators. When data volumes are toolarge for researchers to download data sets to analyseon their desktops, the data centre is not only the placewhere the data are stored but where they must beanalysed. Data curators must therefore provide theenvironment within which users can upload and runanalysis code safely, while ensuring that such analysisdoes not have an impact on the integrity of the datacollection itself. For an increasing number of studiesthis analysis of code will require access to data locatedelsewhere. If this is to be possible the data centre mustbe capable of storing intermediate data for users. Thisis an area where data curation research is likely tointerface directly with Grid computing.
Services
The digital information creating, using, and curatingcommunities require a range of practical services ifthey are to have any chance of securing, using, andenabling the longevity of digital information. Theprovision of effective, practical services for thestakeholder community is a high priority for the DCC.
Work in each of the other task areas reflects this driveto deliver tools and services which will enablecommunities to curate their digital resourceseffectively. The project has created a conduit fromresearch and development into the establishment anddelivery of world-class services and products.
The DCC is not itself a digital repository. Theintention is to provide services and guidance that helpsdata centres and repositories be more productive in theselection, acquisition, ingest and exploitation of data intheir care by current and future users, both withexisting computing and the developing e-science/Gridinfrastructure. Initial tasks are directed towardsintelligence gathering across our communities aboutneeds, practices and provision, and emergent standards.In subsequent work the intention is to provideinformation on tools for automating the ingest andcuration process, and on mechanisms that supportcontinuing use of digital resources, data interchangeand support preservation.
Help desk and FAQs Edinburgh's experience through EDINA provided
the consortium with evidence of the merits of a singlepoint of contact for queries, staffed throughout officehours and skilled at logging, tracking, filtering andpassing queries onto the appropriate expert staff. Asthe DCC develops such information resources asFrequently Asked Questions (FAQs) and itsprogramme of site visits many queries will either beanswered through our website or by the help desk staffpointing callers to the FAQ in which the answer maybe found. The latter is greatly facilitated by thedevelopment of FAQs. The help desk also providessupport for online bookings for training events and sitevisits.
Advisory Service Prior to the DCC's conception there was no single
central source of expertise on digital curation andpreservation matters serving the United Kingdom'sfurther education (FE) and higher education (HE)communities from research to learning and teaching toadministration. Scientists, researchers, and librariansneed to search across many sources of information ifthey wish to identify best practice, find the lateststandards information, or locate a recommendation onappropriate file formats. One of the major benefits ofthe Advisory Service in partnership with the help deskand Web portal, is that it delivers a one-stop-shopwhere relevant information is gathered, evaluated, andpresented to the community. Consortium partners andtheir DCC staff are responsible for contributing to thisactivity to ensure that the wide range of expertise based
6
at each partner site is mobilised for the maximumbenefit of the community. The service is co-ordinatedfrom HATII at the University of Glasgow, which hasgained through ERPANET [18] extensive experience inleading advisory services in digital preservation. TheDCC's professional advisors offer advice on strategic,technical, practical and legal aspects of data curationand preservation. The service provides authoritativeanswers to digital curation and preservation questions,and makes accessible examples of best practice. Theadvisory service works closely with every part of theDCC, responding to queries that cannot be dealt withby the help desk or extant FAQs. The advisory servicetracks DCC (and other) research and developmentactivities in order to keep up-to-date with innovativeresearch, it monitors standards and technology watchoutputs, ensures that the community is informed aboutleading edge activities, and repackages R&D materialsappropriately for practitioners. Using site visits andfocus groups the advisory service engages with thewider community in order to gather user requirementsand feedback on services which will inform work onset-up and gap analysis, tailors services to the differentsectoral/disciplinary groups, develops consensus onbest practice, disseminates information to the widestpossible audience (e.g. Current Awareness Bulletin),and works towards the formation of a curationcommunity that pro-actively shares expertise. More in-depth or tailored support is available where required,through a range of additional services such asinstitutional site visits, so that the DCC canaccommodate the varied requirements of its users in acost effective and efficient manner.
StandardsThe DCC operates a standard watch covering
relevant existing and emerging standards, identifiedthrough interaction with users and researchers. TheDCC staff actively contribute to the definition ofemerging standards by working with the communityand standards establishing bodies (e.g. ISO), organisingassociates groups around new standard developments,and initiating standardisation definition groups wheregaps have been identified.
Audit and certification Institutions are increasingly recognising the need to
provide their user communities with access to trustedrepositories. Irrespective of whether they construct andmanage these themselves or rely on outsourced servicesinstitutions need mechanisms to validate the trustedstatus of repositories. The Online Computer LibraryCenter (OCLC) [28] and the Research Libraries Group(RLG) [29] in their Attributes of a Trusted Digital
Repository [30] paper, have proposed a high levelmodel for the design, delivery, and maintenance of adigital repository, and RLG and NARA [33] areprogressing towards certification requirements forestablishing and selecting reliable digital informationrepositories. The DCC partnership will continuecollaboration with RLG, (CCLRC is alreadyrepresented on this working group), to ensure that inthe development of audit and certification standardsinternational consensus is favoured. The DCC will useits testbeds to support the development of these auditand certification standards. Although there is growingawareness of the certifiable characteristics (e.g.activities, attributes, functions, processes) ofrepositories the mechanisms for audit and the processby which certificates would be issued (and revoked)remain to be agreed. The Centre offers guidance onself-audit and self-certification and will conductindependent audits and issue certificates to repositorieswhich meet its guidelines.
Developmental workshops and professionaldevelopment
Digital curation is an immature discipline. There aresome areas in which our knowledge is secure enough toprovide training but in others the community is stilldeveloping its expertise. The DCC has planned a seriesof workshops and training events to reflect existingknowledge and practices, and to enable the communityto work together to build new understanding in otherareas.
Our programme of training workshops and seminarson best practice, strategy and applications is designedto promote participation across the diverse range ofcuration and preservation need communities in FE andHE institutions. This provides a network of learning fordemand-driven training and continuing professionaldevelopment (CPD) related to preservation andcuration. This work has begun by identifying whereand what further practitioner training and staffdevelopment is required. Identification and discussionof key digital curation issues highlighted by Researchand Development will follow, with subsequentcollaboration fostered between private industry andpublic organisations. Our initial workshops coverPersistent Identifiers [12], Digital Repositories [11],and Costing Models [8] for Digital Curation.
Digital Curation Manual The DCC is committed to the publication of a range
of resources to further assist institutions, data centresand repositories with their digital curation efforts. Atthe forefront of these is the Digital Curation Manual[10], a world-class resource. The Digital Curation
7
Manual is being constructed from fascicules written byinternational experts overseen by leading researchersand practitioners in the area of digital curation. Amongthe forty-five initial topics are Appraisal and Selection,Costs, Freedom of Information, Interoperability, theOAIS Model, Preservation Strategies, and OpenSource. A full list of areas that the curation manualaims to cover can be found at the DCC web site. Toensure that this manual reflects new developments,discoveries, and emerging practices authors will have achance to update their contributions annually. Initially,we anticipate that the manual will be composed of somefifty fascicules, but as new topics emerge and oldertopics require more detailed coverage more might beadded.
The DCC recognises that the manual is very muchfocused on the needs of technical staff. For manytopics, a less in-depth insight is being offered by theDCC briefing papers, which have been designed tomeet the needs of senior managers. These supply quickand high level overviews of the topics that are exploredin technical detail in the Curation Manual.
Development
A central, and demanding, task is to transformresearch products into services for those who areresponsible for digital preservation and data curationactivities within institutions. This requires bothattention to the definition of desirable services and aconcerted programme of development activities. Thereis need here to look to, and gain leverage from extantresearch product as much as from research that is nowin progress, and to build on work successfully carriedout by others as well as by staff within the DCC. In linewith concern for the longevity of information content aswell as the longevity of digital media, early prioritywithin the development programme has been given to'representation information', as defined within the OAISReference Model. The DCC anticipates that the resultsof research suitable for transformation into products orservices will be some years away. Therefore, in the firstinstance the development team is looking outside thecore DCC for research results which can be deliveredas products, methods, or services. These developmentactivities are focussed on making available tools andservices which support various aspects of the OAISReference Model, and is, wherever possible, standardsbased in its approach.
Representation Information Registry and RepositoryArchitecture
Long term access to Representation Information isessential for the curation and preservation of digital
information as it documents the structure and semanticsof the ways in which digital data are stored andprovides a method for accessing the content of digitalobjects. Representation Information is used here in thecontext of the OAIS model, meaning any and allinformation that may be required to access theinformation content of a data object. RepresentationInformation is not necessarily the original or officialsoftware access method or format specification, but cantake the form of anything that allows the informationcontent of a digital object to be interpreted. The currentfocus of development within this area lies with thedesign and implementation of a distributed system ofRepresentation Information Registry/Repositories(DCC-RR). As automation of process is essential ifdigital curation is to prove cost effective efforts tofacilitate interoperability and automated processing ofrepresentation information lies at the core of this work.
In order to implement the DCC-RR, the open-sourcefreebXML [20] registry software is being extended tomeet the functional and architectural needs. Clientsoftware is under development to provide a variety ofmethods to interface with the repository contents.Descriptive and preservation metadata schemas areunder investigation, focussed on enabling thecontextual identification of objects and maintaining arecord of their authenticity and provenance. Tofacilitate interoperability, standardised packagingstructures are under development, these will supportdistributed and federated registry/repositories.Although the current focus of development lies withthe technical infrastructure, there remain manystructural, organisational, and rights issues associatedwith building such a repository that must be addressed.These will enable the selection of RepresentationInformation for inclusion in the registry, profiling ofusage constraints, mechanisms to ensure the contentsare maintained and up-to-date, and adhere tointernationally agreed standards.
Project Integration and Community InvolvementTo broaden the scope of development and the
comprehensiveness of the services provided, thedevelopment team intend to build upon the resourcesand findings of existing projects. For example,interoperability is essential for Digital Curation, withvalue identified in making DCC services as easy tointegrate as possible into existing projects which use orservice digital information. Such interoperability willaid the development of a global network ofRepresentation Information repositories. Investigationsare being undertaken into how RepresentationInformation may fit into the architecture of digitalrepositories in order to provide such a global,
8
distributed infrastructure. The dynamic nature of the contents of the DCC-RR
means that maintaining and updating the informationresource will require continuous investment of effortand resource. The development team will look at waysto foster and encourage community interaction andcontribution in order to improve the implementation,the functionality provided, and the informationcontained within.
Tool development In order to best exploit the DCC-RR's potential
services that will sit on top of the softwareimplementation are needed. These services will beprimarily software driven, taking the form of externaltools and extension modules providing functionalitythat builds upon the core registry utilities. The DCC-RR requires that a label is associated with every digitalobject, which acts as a pointer towards relevantRepresentation Information stored within therepository. Tools must be selected or developed thatidentifies both the requirements of a digital object andthe Representation Information that satisfies theserequirements. A label containing persistent identifierspointing into the DCC-RR should be produced fromthis resultant set of information. Additional utilities willprovide functionality to browse, search, view, andupdate contents. Tools to provide data description arein active development, and the development team planto investigate methods to identify the format ofunknown files, verify whether a file is the format itpurports to be, assess the viability and success oftransformation, and identify the risk that particularrepresentations may pose for a special digital object.
Community Development and PopulationThe dynamic nature of the contents of the DCC-RR
means that maintaining and updating the informationcollection will require significant work. Communityinteraction is welcomed and actively encouraged toassist in the improvement of the implementation and thefunctionality it provides.
Standards The development work is based on international
standards such as OAIS and others in the OAISroadmap, and there are ongoing efforts to influence thedevelopment of several other international standards.One developing standard of particular interest dealswith OAIS certification.
TestbedsA number of factors influence the suitability of a
curation software solution for a particular purpose.
These factors include the cost, the effectiveness oftechnical solutions such as migration or emulation,techniques for metadata creation and management, andverification of the authenticity of digital objects. A testbed framework is under development which willoperate on a variety of hardware and softwareconfigurations. The test beds are expected to enablethe assessment of testing of these tools and processesto a high standard of QA to enable their distributionand usage by designated user communities. Whereverpossible, this work will be carried out in collaborationwith the original tool developers.
Outreach & Associates Network
At the forefront of all the DCC's interactions with itsstakeholder communities is our Outreach team, whothrough a variety of activities are committed to thepromotion of the DCC's message, its services and itsresearch and development successes. The Centre mustachieve high visibility if it is to have a measurableimpact on the communities of practice and of policythat it aims to serve. Like NeSC, the DCC is acommunity forum that serves as an environment forinformed discussion and debate on key issues, and aimsto facilitate the sharing of expertise and skills.Outreach and dissemination work provides the DCCwith the opportunity to listen and by capturing detailsabout individual user and community needs, to reflectthem in our research, development, and servicesactivities. Close contact is anticipated with other keyorganisations including the Digital PreservationCoalition (DPC) [16] and the DELOS Network ofExcellence [13,14] in order to co-ordinate timings andto run joint events.
User requirementsThe DCC has also commissioned a study to
investigate user requirements for digital curation aspart of the Outreach activity. The study has used a mixof methodologies such as focus groups and interviewswith key individuals who are working with data sets,researchers drawn from various disciplines, and policyand funder representatives. The outcomes from thestudy include an outline taxonomy of curation usersand organisations and an overview of requirements,which will inform the planning of services and researchand development activities of the DCC. The resultssuggest that there is a clear need for further informationand expertise in a wide range of areas, includingprofessional development and training, technicaladvice and guidance on best practice, research ontopics such as annotation, economic cost-benefitsanalyses and pragmatic case studies.
9
Web portalThis 'virtual point of presence' is the main access
route for users to discover information about the DCCand UK-led work in this area, including best practiceguides, and digital curation resources. The DCC Website showcases the wide variety of user communitieswhose work and challenges the DCC addresses. TheWeb portal represents views of different disciplines,with language about digital curation and preservationtailored to their needs and circumstances. The portal isthe first to establish a cross disciplinary set of tools andadvice that encompasses the full range of scholarlyenquiry. Reciprocal links to and from related sites arebeing established, thus leveraging the ways we canreach out to our target user communities. The Webteam is working closely with the Advisory Service, andthis collaboration is yielding online documentationincluding best practice guides and other deliverables,that show the expertise gained through interaction withthose in the field and provide the community withaccess to preservation information.
International Journal of Digital Curation There is currently no high-quality peer reviewed
journal in digital curation. If digital curation is to havea significant intellectual focus it must have a journal.As a result the production and delivery of an electronicjournal dedicated to digital curation and preservationresearch, service development issues is seen as a keyelement of the outreach activity [23]. This e-journalwill not be a dissemination and promotion channel forthe DCC and its services. It will instead aim to providea publishing focal point for researchers to presentoutstanding contributions to the field; of course, wehope our independent peer reviewers will consider theoutputs of the DCC research team suitable forpublication in the journal, but it will be their call.
Associates NetworkThe DCC Associates Network [9] makes the DCC
partnership more pervasive; it brings togetherprominent members from UK data creating andmanaging organisations, leading data curators overseas,supranational standards agencies, and representatives ofsectors of UK industry and commerce involved indigital curation. Members for the Associates Networkcontinue to be sought. The benefits of membershipinclude: financial savings for DCC events, early accessto outputs from leading-edge research anddevelopment, support from Advisory Services,tutorials, training courses, input into user needsdefinition and service design, access to a range ofcollaborative tools, including a web-based forum, and
the ability to post news and events information on theDCC's web site.
Way Forward
Our vision of a successful DCC is one that iscoherent and international in its outlook whichpromotes world-class research and ensures that thefruits of that research will serve the purposes ofscholarship and science in the UK. By combiningacknowledged research strengths with proven serviceorganisations, we seek to achieve the virtuous circle(see figure 2 above). By engaging varied communitiesof practice through our network of associates, and bygaining leverage from their expertise throughmechanisms such as secondment, fellowships and part-time assignment, we ensure the relevance of ourresearch, and inform the advice we can offer our users.This lays the grounds for: leadership and advocacy,continuing professional development, and promotingdigital curation and preservation.
The consortium partners recognise that the viabilityof the DCC depends upon it gathering the right level ofexpertise, making that expertise available to the widestcommunity, and demonstrating long-term commitmentto the provision of research, development, services,and outreach. The DCC business plan sets incrementalstage targets towards achieving long-term sustainabilityand it accommodates review of the DCC's progresstowards achieving these targets. Encompassing vision,expertise and an acute awareness of the essential roleof effective curation in all our digital activities, theDCC aims to be the embodiment of its vision fordigital curation, and to succeed in its goals to be astandard-bearer for best practice in an area that isrelevant to every individual, institution andorganisation that relies upon and uses digitalinformation.
AcknowledgementsThe authors wish to thank all their colleagues in the
DCC and on this occasion to acknowledge particularlyAndrew McHugh (HATII, University of Glasgow),Robin Rice (Edinburgh University Data Library), andAdam Rusbridge (HATII, University of Glasgow) fortheir contributions. Many individuals played a crucialrole in the design of the Digital Curation Centre andare active in helping it to deliver on its mission, amongthese are Charlotte Waelde (Co-Director of the AHRCResearch Centre for Studies in Intellectual Propertyand Technology Law), and Bob Mann (University ofEdinburgh and the National eScience Centre). Theauthors wish to thank Professor Tony Hey (Director of
10
the UK e-Science Core Programme of the EPSRC), andNeil Beagrie (JISC), for their support and work toconstruct the funding platform that has made the DCCpossible.
References[1] Access Grid, http://www.accessgrid.org/ [Accessed:
10 May 2005, 16:44][2] AHRC Research Centre for Studies in Intellectual
Property and Technology Law,http://www.law.ed.ac.uk/ahrb/index.asp [Accessed:9 May 2005, 11:40]
[3] Ariadne Magazine, http://www.ariadne.ac.uk/[Accessed: 9 May 2005, 12:00]
[4] Biotechnology and Biological Sciences ResearchCouncil, http://www.bbsrc.ac.uk/ [Accessed: 9 May2005, 12:14]
[5] CCSDS, 2002, Reference Model for an OpenArchival Information System (OAIS),http://www.ccsds.org/documents/650x0b1.pdf
[6] Council for the Central Laboratory of ResearchCouncils, http://www.cclrc.ac.uk [Accessed: 9 May2005, 11:29]
[7] Database Group at Edinburgh,http://www.lfcs.inf.ed.ac.uk/research/database/dbs.html [Accessed: 9 May 2005, 11:39]
[8] DCC and DPC Joint Workshop on Digital CurationCost Models,http://www.dcc.ac.uk/cmworkshop.html [Accessed:11 May 2005, 13:31]
[9] DCC Associates Network,http://www.dcc.ac.uk/network.html [Accessed: 9May 2005, 12:26]
[10] DCC Digital Curation Manual,http://www.dcc.ac.uk/
[11] DCC Workshop on Long-Term Curation withinDigital Repositories Workshop,http://www.dcc.ac.uk/drworkshop.html [Accessed:11 May 2005, 13:30]
[12] DCC Workshop Workshop on PersistentIdentifiers, http://www.dcc.ac.uk/piworkshop.html[Accessed: 11 May 2005, 13:30]
[13] Delos Network of Excellence on Digital Libraries,http://www.delos.info [Accessed: 9 May 2005
[14] DELOS Network of Excellence on DigitalLibraries, Digital Preservation Cluster,http://www.dpc.delos.info [Accessed: 10 May2005, 9:25]
[15] Digital Curation Centre, http://www.dcc.ac.uk[Accessed: 9 May 2005, 11:20]
[16] Digital Preservation Coalition,http://www.dpconline.org [Accessed: 11 May 2005
[17] EDINA, http://edina.ac.uk [Accessed: 9 May2005, 11:47]
[18] Electronic Resource Preservation and AccessNetwork (ERPANET), http://www.erpanet.org[Accessed: 9 May 2005, 11:53]
[19] Engineering and Physical Sciences ResearchCouncil, http://www.epsrc.ac.uk/ [Accessed: 9 May2005, 11:25]
[20] Freebxml, http://www.freebxml.org [Accessed: 11May 2005, 16:40]
[21] Giaretta, D, et.al. 2005, Draft DCC Approach toDigital Curation,http://dev.dcc.rl.ac.uk/twiki/bin/view/Main/DCCApproachToCuration?rev=1.1.9
[22] Humanities Advanced Technology andInformation Institute (HATII),http://www.hatii.arts.gla.ac.uk [Accessed: 9 May2005, 11:25]
[23] International Journal of Digital Curation,http://www.ijdc.net [Accessed: 9 May 2005, 12:05]
[24] Joint Information Systems Committee (JISC),http://www.jisc.ac.uk [Accessed: 9 May 2005]
[25] JISC Circular 06/03 (Revised) Digital CurationCentrehttp://www.jisc.ac.uk/index.cfm?name=funding_digcentre [Accessed: 9 May 2005, 12:24]
[26] JISC Press Release, 2004, New UK Centre toCommunicate Across Time,http://www.jisc.ac.uk/index.cfm?name=pr_dcc_051104
[27] National e-Science Centre, http://www.nesc.ac.uk[Accessed: 9 May 2005, 11:29]
[28] Online Computer Library Center,http://www.oclc.org/ [Accessed: 9 May 2005,12:28]
[29] Research Libraries Group, http://www.rlg.org[Accessed: 9 May 2005, 12:29]
[30] RLG-OCLC, 2002, Attributes of a TrustedRepository, http://www.rlg.org/longterm/repositories.pdf[Accessed: 9 May 2005, 12:30]
[31] School of Informatics at the University ofEdinburgh, http://www.inf.ed.ac.uk [Accessed: 9May 2005, 11:30]
[32] UKOLN, http://www.ukoln.ac.uk [Accessed: 9May 2005, 11:29]
[33] US National Archives and RecordsAdministration,http://www.archives.gov [Accessed:9 May 2005, 12:31]
[34] University of Bath, http://www.bath.ac.uk[Accessed: 9 May 2005, 11:29]
[35] University of Edinburgh, http://www.ed.ac.uk[Accessed: 9 May 2005, 11:25]
[36] University of Glasgow, http://www.gla.ac.uk[Accessed: 9 May 2005, 11:30]
11
Top Related