Transcript hitchcock-sconul-workshop-jun06-updated

Digital and institutional repositories:
emerging architectures
A short workshop moderated by
Steve Hitchcock, School of Electronics and
Computer Science (ECS), Southampton University
On 21 and 22 June, 2006, at the
SCONUL Conference 2006, Newcastle-upon-Tyne
The workshop brief
The question for repositories is not which software you should
choose, but which applications and services you want to support.
With growing numbers of institutional repositories and increasing
commitment within institutions, alongside an active and growing
programme of development more broadly in digital repositories,
notably that sponsored by JISC in the UK, this is a picture that
could change significantly in the next few years. This workshop
will provide the opportunity to explore and discuss, from
institutional perspectives, some of these developments, with a
view to anticipating emerging architectures that could support
expanded repository capabilities.
For a brief report and follow-up, see this blog entry
http://www.eprints.org/community/blog/index.php?/archives/89Emerging-IR-architectures-investigation-by-workshop.html
What is an IR? Lynch 2003
“a university-based institutional repository is a set of services that a
university offers to the members of its community for the
management and dissemination of digital materials created by the
institution and its community members. It is most essentially an
organizational commitment to the stewardship of these digital
materials, including long-term preservation where appropriate, as
well as organization and access or distribution. ….. An institutional
repository is not simply a fixed set of software and hardware.”
Cliff Lynch, 2003 http://www.arl.org/newsltr/226/ir.htm
What the workshop is NOT about
Open access
Technical interoperability (OAI-PMH, METS,
Z39.50, SRW, AJAX, Web 2.0, etc., see
Augmenting interoperability across scholarly
repositories, New York, April 2006
http://msc.mellon.org/Meetings/Interop/ )
National repository services (Linking UK
Repositories)
Departmental repositories (e.g. Caltech)
Consortium repositories (e.g. White Rose)
Comments on following schematics
Commentary slide added post-workshop
The following three slides illustrate the sort of components that might
be found in an IR:
1 Schematic, Liz Lyon, UKOLN, for eBank UK project
2 OCLC 2003 environmental scan
These are two network-based examples, but don’t have an institutional
perspective
3 Chart, RepoMMan project
This is getting closer to the institutional view being investigated here,
coordinating data types with users
Schematic by Liz Lyon, UKOLN, for eBank UK project
Data creation /
capture /
gathering:
laboratory
experiments,
Grids,
fieldwork,
surveys, media
Resource
discovery, linking,
embedding
Data analysis,
transformation,
mining, modelling
Searching ,
harvesting,
embedding
Aggregator services:
eBank UK
Resource
discovery,
linking,
embedding
Learning object
creation, re-use
Harvesting
metadata
Research &
e-Science
workflows
Validation
Deposit / selfarchiving
Learning &
Teaching
workflows
Repositories :
institutional,
e-prints, subject,
data, learning objects
Validation
Publication
Resource
discovery, linking,
embedding
Linking
Data curation:
databases & databanks
Deposit / selfarchiving
Institutional
presentation
services: portals,
Learning
Management
Systems, u/g, p/g
courses, modules
Peer-reviewed
publications: journals,
conference proceedings
Validation
Quality
assurance
bodies
OCLC 2003 environmental scan
RepoMMan project, Hull
What this workshop IS about
The institutional repository
The institutional perspective
The multi-repository institution
Where does it exist? What does it look like?
What will it look like?
Your view
Edinburgh example: now
We have at present:
• A catalogue repository with records for ejournals and ebooks (and
some web sites), which we feel should be accommodated in
separate repositories;
• a single repository for 'research outputs' (eprints, research papers
and reports, and theses);
• a proto-research publications repository (currently serving as our
RAE publications repository);
• Separate repositories for image and museum collections;
• a learning objects repository;
• an archives repository;
• and a proto-repository for locally digitised research collections
(with little in it, but quite a lot of planning done).
Thanks to John MacColl for this example
Edinburgh example: next?
We wish to develop this into a more efficient architecture by:
• splitting out the catalogue records by their various types;
• splitting out the research outputs into separate repositories;
• introducing an image/musuem management system (currently being
implemented);
• introducing a licence management system (currently being
implemented);
• creating a digital records repository (not started);
• migrating the locally digitised research collections from one system
(Endeavor ENCompass) to another (probably Dspace).
We are also seeking to
• apply a federated search engine to the entire architecture (WebFeat,
currently being implemented).
Repository data types
Open Access (published papers. preprints, tech. reports, etc.)
Electronic Theses and Dissertations
Teaching & Learning
E-science, datasets
Research Information (CRIS)
Multimedia (audio, video, images, museum collections)
Digitisation
Publishing
Preservation
Administration
Open source software?
Open source software
DSpace, EPrints, Fedora, Moodle, Bodington, Sakai, Plone ….
Repository data types
Open Access
ETDs
T&L
Datasets
EPrints
DSpace
Moodle, Bodington, Sakai, Plone
EPrints
Multimedia
CRIS
Digitisation
Publishing
Preservation
Bepress
Fedora
Administration
Examination papers
Web pages
Structured databases
???
Open source or Services?
Open source software
DSpace, EPrints, Fedora, Moodle, Bodington, Sakai, Plone ….
IR Services
Open Repository (based on DSpace), EPrints Services
Multi-repository institution
???
???
Open Access
ETDs
T&L
Datasets
Multimedia
I
N
T
E
R
O
P
E
R
A
B
I
L
I
T
Y
CRIS
Digitisation
Publishing
Preservation
Administration
An interoperating network?
“Repository deployment is fragmented, and repositories tend to exist in
isolation rather than being embedded into an interoperating network of
services. “We've got bits and pieces but it doesn't operate as a whole
and there are big gaps in provision in some areas.” Within
institutions, repositories tend not to inter-work with other
applications. Nor are they well integrated with other institutional
repositories (although there are some examples of innovative
workflows, for example between laboratory repository and crossinstitutional repository in R4L/eBank).”
R4L: Repository for the Laboratory http://r4l.eprints.org/
eBank UK http://www.ukoln.ac.uk/projects/ebank-uk/
Rachel Heery and Andy Powell, Digital Repositories Roadmap: looking
forward, JISC, April 2006
http://www.jisc.ac.uk/uploaded_documents/rep-roadmap-v15.doc
The Institutional Repository?
???
Open Access
CRIS
ETDs
Digitisation
T&L
Publishing
Datasets
Preservation
Multimedia
Administration
???
Open source or Services or LMS?
Open source software
DSpace, EPrints, Fedora, Moodle, Bodington, Sakai, Plone ….
IR Services
Open Repository (based on DSpace), EPrints Services
IR Services + library management systems
VTLS VITAL (Fedora), Proquest (Digital Commons), Ex Libris
(DigiTools)
The Digital Library
Library Management System
Institutionally-generated content
Open Access
Library content
CRIS
ETDs
Digitisation
Library Services:
OPAC
OpenURL resolver
E-journals
Etc.
T&L
Publishing
Datasets
Preservation
Multimedia
???
???
Administration
What is an IR? Lynch 2006
“As I've looked more at various institutional
deployments and planned deployments, I think
that the distinction between digital libraries,
digital collection management systems, digital
archives, and institutional repositories is less
clear than I might have felt in 2003."
Cliff Lynch, 2006, quoted in
http://poynder.blogspot.com/2006/03/institutional-repositories-andlittle.html
Networkflows
"Historically, users have built their workflow around the services the
library provides. As we move forward, the reverse will increasingly
be the case. On the network, the library needs to build its
services around its users' work- and learn-flows
(networkflows).
"one of the discussion points around institutional repositories is about
which goals they support: open access, curation of institutional
intellectual assets, reputation management. And which processes?
Over time, it is clear that what we now call institutional repositories
will be part of wider research process support. What is currently the
institutional repository will be a component of the
workflow/curation/disclosure apparatus that develops to support
research activities."
Lorcan Dempsey, Networkflows, January 2006
http://orweblog.oclc.org/archives/000933.html
Integration, workflow, portals
The role of the repository will influence the level of integration and
interaction required.
Where a repository is being used for multiple content types and as an
everyday working tool then greater integration is required to allow it
to take on this role. Integration may also be focussed at the
presentation level for end-user interaction (e.g., presenting a search
or deposit screen within a portal) or can be at the data level for the
exchange of information between systems.
Alma Swan and Chris Awre, LINKING UK REPOSITORIES, A6.6
Repository integration in local infrastructure
http://www.jisc.ac.uk/uploaded_documents/Linking_UK_repositories_appendix.pdf
The Digital Library
Portal
Library Management System
Institutionally-generated content
Open Access
Library content
CRIS
ETDs
Library Services:
OPAC
OpenURL resolver
Digitisation
T&L
Publishing
Datasets
Preservation
Multimedia
???
???
Administration
Embedding IRs in institutional
strategy
“Often institutions are not clear as to their strategy for establishing
repositories. There are real benefits for institutions in effectively
managing their digital assets (promoting research outcomes, fulfilling
preservation responsibilities, facilitating added value services such as
overlay journals, data mining, etc). Such benefits can be assisted by
leveraging the open access agenda. Despite this, repositories are
not yet fully embedded in institutional strategy and there is
perhaps a misplaced confidence that institutions will take on the
full range of repository business functions. Interoperability
between institutional libraries, repositories, learning management
systems and MIS is still rare.“
Rachel Heery and Andy Powell, Digital Repositories Roadmap: looking
forward, JISC, April 2006
http://www.jisc.ac.uk/uploaded_documents/rep-roadmap-v15.doc
Assessing the costs
From a spreadsheet on the costs of setting up and
maintaining Open Source repository: "There was
a range, from about $6,886.62 for a set-up cost,
all the way to over $1 million."
Rebecca Kemp, list posting, November 2005
http://www.library.yale.edu/~llicense/ListArchives/0511/msg00030.html
Wider frameworks
Where do wider services – national
frameworks and services, Web search and
other Web services – fit into the
institutional agenda?
Need for IRs to be visible, searchable and
usable by local and distant users
Summary
•
•
•
•
•
One repository or multi?
OSS vs Services vs extended LMS
IR or DL
Build around workflows
Embed in institutional strategy