Repositories 29

DOMS Project

Digital Repository – roles, functions, and directions Report to DOMS Project Steering Committee Prepared by DOMS Team: Dr Simon Pockley - Manager Ebe Kartus - Metadata coordinator Monica Brock - Business Process Coordinator Susan Russell - Training & Documentation Coordinator Version 1.1 August 20005

Contents Preface................................................................................................................... 1 Executive Summary................................................................................................2 What is a digital repository?....................................................................................2 Why are repositories important?.............................................................................3 What are the big challenges?..................................................................................6 International initiatives............................................................................................6 National initiatives...................................................................................................7 University repositories.............................................................................................7 The Deakin environment.........................................................................................7 Technical considerations.........................................................................................8 Key development areas- services...........................................................................9 References........................................................................................................... 10

Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 1 of 12

DOMS Project Preface The need to develop a digital repository at Deakin University was identified in August 2000. Following the development of an RFI, and the subsequent testing of a content management solution, this report marks the beginning of a more comprehensive approach to supporting the organisational changes required to lift the quality of teaching and learning at Deakin University.

Executive Summary What is a digital repository? The learning environment at Deakin University (like most universities) consists of a range of system components. Most of these components are fairly narrow in scope and aligned in some way with what the various divisions believe they need or what vendors have offered. The central position of a repository within the learning environment makes it important to clarify the difference between the various kinds of content management systems and a repository. Content management systems (CMS – e.g. Web Management Tool) typically focus on the creation, management, distribution of material targeted at either a website or intranet. Document and Records management systems (DMS & RMS e.g. Ken Mould’s use of Lotus Notes) typically focus on the creation, management, and flow of documents. Digital asset management systems (DAMS) often include rights management functions. Library management systems (LMS e.g. Innopac) support the daily administrative activities of the library. Learning management systems (LMS e.g. WebCT Vista) are aimed at the needs of professional educators. Although they share many common features, each in its own way, supports an implicit pedagogical approach. The limited flexibility of these systems can lead to the establishment of data islands and constrain innovative educational practice. In the wider university environment, the notion of an institutional repository for digital resources, as an assembly of components, needs to be considered not only from the context of these more process driven systems, but also from a higher vantage point – wider in scope and more durable in outlook. In order to be sustainable, the establishment of a central repository necessarily involves a long-term commitment. ‘...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 members. It is most essentially an organizational commitment to the stewardship of these digital materials…’ (Lynch 2003) 

 



Set of services: including receiving the documents into the repository; maintaining them (including who has rights to access them); preserving them; and delivering them to those who need them. Management: including establishment of policies at the institution level and the drafting and execution of procedures. Dissemination: search and delivery tools to deliver meaningful and functional data/objects/records; access and use rights (including copyright, etc.) Digital materials: including objects, publications, data, and records.

Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 2 of 12

DOMS Project 

Stewardship: institutional repository commitment, both long-term and shortterm, to define who is responsible for the repository and the relationship between the responsible party and the rest of the institution.

A simple definition of a repository has been employed by the Arrow project, where a repository is defined as ‘a managed collection of digital objects’ (Payne 2005). However such a definition needs to be augmented by a description of the services provided. In order to capture some of the experience gained so far through the DOMS project we can expand this definition as follows: A digital repository can be defined as a managed collection of digital material available for access by humans and machines. An effective repository is broadly available, widely accessible, simple to use, scaleable and sustainable over time. Essential services and features include:      

Ability to search for digital resources through query, browsing, and via software agents Adaptable ways to access and use resource content by providing read access, import and export of digital objects Respect for access constraints on content but controls only employed at the point of need. Standard and consistent metadata for all digital resources encoded in commonly used formats in forms that can be understood by people and machines Stable references to digital resources (eg. Persistent URI) and support for citations (in recognised scholarly formats) Access to documentation about the functions, technical profile, and policies of the repository.

Why are repositories important? The digital repository has become a visible manifestation of the emerging importance of knowledge management within higher education. The validity of the proposition to be tested is that the widespread use of an institutional repository can influence the quality of digital material by aligning the criteria for submitting or depositing material with content quality, pedagogical effectiveness and usability. The long-term impact of the values that underpin the development and use of interoperable repositories is likely to change many of the basic assumptions about how the conduct and expression of intellectual activity is managed by individuals, their colleagues, and the university. Different communities of interest place different levels of importance on the benefits that can arise from the widespread use of digital repositories.

Business case perspective: Although the full costs of developing and maintaining an effective electronic learning environment are difficult to quantify, they are far higher (average 76% increase) than traditional classroom environments (Gartner 2003).

Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 3 of 12

DOMS Project Scholarly publishing perspective: The Scholarly Publishing & Academic Coalition position paper, The Case for Institutional Repositories describes the institutional repository as a compelling response to two interrelated issues of strategic importance: 1. Repositories are the foundation of a new model of scholarly publishing. The repository expands access to research, reasserts control over scholarship, reduces the monopoly power of journals, and brings economic relief and heightened relevance to the institution that supports it. 2. Repositories have the potential to serve as tangible indicators of a university's quality and to demonstrate the scientific, societal, and economic relevance of its research activities, thus increasing the institution’s visibility, status, and public value. The position paper identifies other significant benefits of institutional repositories. They include:    

stimulating innovation in a new disaggregated publishing structure building on the grassroots of faculty practice of self-posting and self-archiving research online able to be implemented using existing resources without radically altering the status quo facilitating the development of intellectual property policies

The ITS perspective The IT community is beginning to acknowledge a trend towards using technology to support the individual to using technology to support relationships between individuals. The existence of a digital repository reflects traditionally distinct and seemingly contradictory social dimensions when individualism and community are brought together. Networked repositories of digital media become new and significant vehicles for efficiency, insight and conceptualization. These repositories require data structures that reflect the requirements for merging, linking and managing diverse, interdisciplinary digital assets. Five principles, referred to as a coordinated autonomy (Davis 2000), support a core social value and allows for individual pursuit to be harnessed within the context of a worldwide community. Mass Individualism: broadly disseminated information but with a focus on the needs of the individual who can be surrounded with the resources relevant to local inquiry or objective. Mass individualism also captures the implications of access, the notions of data as an institutional resource, portal and reporting strategies that integrate information services but offer flexibility for local configuration, ready collaboration, and personally relevant learning. Robust Flexibility: recognizes an infinite number of possible user applications. The front-end investment in infrastructure, commonly used tools, data and information, and standards that support modular deployment of applications is not only fiscally responsible, but critical to coordinated autonomy. This principle positions the university better for responding to opportunities. Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 4 of 12

DOMS Project Community Direction: better to invest less in big application, deterministic planning and more in planning of the form that includes rapid prototyping and multi-user feedback. This planning process involves the university community earlier and can produce more buy-in and earlier adoption. It widens the potential of drawing upon the wisdom of an engaged community. Persuasive Standardization: imposed standards philosophically, and sometimes in reality, cut at the core of autonomy. Yet, standards that are appropriately accepted, coordinated and managed are critical to autonomy. Managed Openess: shared digital resources assist people to interact intellectually on a very broad scale and pursue individual lines of inquiry. While serious security and access issues must be addressed and managed, there is enormous intellectual value in accessibility.

Teaching and learning perspective There has been considerable theoretical research within both the industrial and the academic and educational metadata communities (e.g. IEEE, LOM, SCORM) into the storage and reuse of digital teaching materials also known as reusable digital objects. While there has been a significant lack of consensus about how a digital learning object can be defined, there is also vigorous disagreement as to whether a learning object should be regarded as a discrete entity or whether it should contain embedded pedagogical information. For this community of interest, the existence of a digital repository underpins an explicit model for improving the overall learning environment. To date, many eLearning initiatives have been characterised by a heavy focus on units of content and metadata rather than on units of activity. This often leads to an inclination towards information transmission as the overriding yet implicit pedagogical model. The major problem with the learning objects model, is that it fails to provide a coherent framework for expressing semantic relationships between the learning objects in an educational context. Furthermore, the need to standardise content to fit a particular pedagogical approach is both restrictive and inflexible. A learning object that encompasses a single pedagogical concept (e.g. an image) will have a greater chance of being reused than a multipart object such as a course. Despite considerable interest in establishing digital repositories by institutions, there are reports of poor uptake by academics. Use by students is either not an option or ad-hoc and optional. Instead of focusing on developing staff skills, local tools, applications and learning activities to meet specific learning needs, many institutions have focused on centralising content and integrating with existing systems. Furthermore, implicit in the development of Standards for content reuse, such as the Sharable Content Object Reuse Model (SCORM), is the assumption that academics will actively participate in a collaborative culture of shared teaching and learning material. While teachers have long been reusing their own material and incorporating third-party sound, image and paper-based resources, there is little evidence of any willingness to participate in a culture of shared digital resources (Littlejohn 2003). Nevertheless, while both the pedagogical and collaborative assumptions should be challenged, the most commonly identified potential benefits of a digital repository identified by the teaching and learning community include:

Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 5 of 12

DOMS Project    

Minimisation of duplication where savings in both staff time and money are assumed to occur if material can be reused Increase in the quality of courseware is assumed to occur when academics cooperate in the creation of course material Integration of internal and external content sources resulting in a rich bank of resources from which academics can draw when creating course material Development of a collaborative culture

Here have been a number of cooperative ventures between universities to investigate the development of interoperable repositories and the reuse of learning objects. International initiatives include ?????. The most notable Australian initiative is the Arrow Project.

Social perspectives - expand: Information ecologies: Social connections inherent to an interconnected system of people, practices, tools and values.

What are the big challenges? Sustainability Preservation and collection management Persistent Identifiers Rights management – Deakin IP - Creative Commons Interoperability Operational silos, legacy systems and migration Point of creation tools for adaptable media to suit multiple deployments

International initiatives The Open Archives Initiative develops and promotes interoperability standards to facilitate the dissemination of content. Their Metadata Harvesting Protocol provides mechanisms for multiple disciplines to contribute to an institutional repository using common metadata. Participants include the Library of Congress, Harvard, Virginia Tech, Los Alamos, Cornell, CNI, NSF, and the Mellon Foundation. MIT’s DSpace is an open source software platform that enables capture and submission of works, distribution of those works, and long-term preservation of assets. DSpace' endeavors to create a federated collection of intellectual resources from the world's leading research institutions. Harvard’s Digital Repository Service (DRS) provides Harvard owners of digital material with a storage and retrieval system for their collections. Services and facilities include an electronic storage facility, management of administrative and structural metadata, preservation policies and procedures, and delivery of objects to front-end systems such as online catalogs.

Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 6 of 12

DOMS Project The Flexible Extensible Digital Object and Repository Architecture (Fedora) is a foundation for developing interoperable digital libraries and institutional repositories using XML and Web services. Jointly developed by the University of Virginia and Cornell University, and sponsored by the Mellon Foundation. The Canada ARL (CARL) Institutional Repositories Pilot Project is implementing institutional repositories at nine Canadian universities.9 The University of Southampton’s EPrints initiative is designed to manage disciplinary or institutional print collections, rather than digital collections. Eprints software is OAI compliant and freely available under a GNU license, and is in use at California Institute of Technology, the University of Queensland, University of Melbourne (planned).

National initiatives University repositories The Deakin environment A survey of Deakin University’s individual hard drives and shared drives would reveal a rich and heterogeneous locally-produced digital content such as images, maps, electronic texts, audio and video files, archives inventories, research data, departmental reports, scholarly literature, digital presentations and instructional tools. Many of these collections and initiatives are supported and used by particular user communities within specific disciplines. Unfortunately, many such distributed resources are less well known, and generally are less discoverable, to a wider audience. The historical absence of incentives and mechanisms for coordination and data-sharing has resulted in a low probability of interoperability among these distributed resources. Combined with the inability of campus ‘search engines’ to index content within databases, or to satisfactorily index anything but the simplest text encoding, this means that most of the unique and valuable digital content created on campus is essentially undiscoverable by the average online user. The long-term survival of much of this existing digital content is questionable. Over time, departmental and faculty interests shift, electronic formats and platforms change, and the long-term support and processes needed for digital archiving often are not part of the initial planning for specific projects. Though much of the digital scholarly output of Deakin University is of enduring value and interest, myriad competing demands on time and resources, a lack of adequate documentation, and other constraints are endangering the survival of much of this content. With new initiatives sprouting continually, and the Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 7 of 12

DOMS Project acceleration of digital content being produced, steps must be taken soon to prevent loss and to preserve the unique intellectual capital of our institution. An extended consideration of the campus digital initiatives landscape also produces the realization that many potential digital collections never materialize because of barriers to entry for campus units and faculty. Shifting technologies and standards, the high costs of developing and maintaining adequate systems, general uncertainty about best practices, and few guarantees of success can be very intimidating. As a result, many useful and valuable potential collections are not being created or shared with others at the university or with the wider world of scholars. Policy and risk management Code of good practice in research - procedure Code of good online practice - operational policy Research and training – operational policy Quality assurance and continuous quality improvement in respect of academic matters – operational policy Quality assurance enabling policy and procedure

Knowledge media risk register identifies the implementation of the DOMS as a significant control for the following risks: 

Teaching and learning materials not available to students on time Register reference: LS-3 Impact: high



Non compliance with legal requirements (copyright, licensing, accessibility) Register reference: LS-18 Impact: medium



Loss of knowledge and skills Register reference: LS-22 Impact: medium

Technical: [description of the Deakin environment from technical perspective]

Technical considerations Capacity planning exposed costs/hidden costs Trends: Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 8 of 12

DOMS Project  

storage capacity doubles every 12-18 months where the price of storage drops by 30% - 40% disk cost per gigabyte is now cheaper that tape storage

Servers on hand Technical specifications for the development server – Mintaka (doms-dev) [requires headings] Sun Fire 880 8 x SPARC III CPUs @ 1 Ghz 32 Gb RAM 6 x 72 Gb internal disks 2 x 2Gbit FC cards 64 Gb volume  /opt/doms  The location where all Interwoven software is installed (TeamSite, OpenDeploy, MetaTagger, etc). 128 Gb volume/  mnt/doms/iw-store/  The location where TeamSite manages its own file system, and stores all TeamSite content. This is the address for additional information about the server Mintaka's specifications from the Sun website, which I am including for your additional information:http://www.sun.com/servers/midrange/v880/spec.html

kajam - Domain on SUN Fire 15K - Solaris 9 - 8 x SPARC IV 1Ghz CPU's - 32 Gb RAM - 6 x 72Gb SCSI HDD's - 128 Gb LUN This is the link to the additional detail for the production server Kajam http://www.sun.com/servers/highend/sunfire15k/specs.xml

Key development areas- services

o o

 Policy  Unique identifiers http (separation of reference form resolution) anyone can play It's distributed (no single central source of name bindings)  It's scalable (no intrinsic upper bound on what can be named) 

Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 9 of 12

DOMS Project o o

Self certifying file system http://www.fs.net/sfswww/ Ergo anyone designing a persistent identifier system should start from  the assumption that http: URIs are sufficient for their technology  needs. Citation is a challenge for many existing schemes open url DOI/handle  Authentication and access systems need to be improved to support role based access to various renderings of digital assets.

References Crow, R. The Case for Institutional Repositories. A SPARC Position paper. 2002. http://www.arl.org/sparc/IR/IR_Final_Release_102.pdf Sandy Britain, Oleg Liber. A Framework for the Pedagogical Evaluation of eLearning Environments February 2004 http://www.elearning.ac.uk/resources/VLEFullReport08.doc

40% 35% 30% 25%

2001

20%

2003

15% 10% 5% 0% in tra ne ti n ho Bl ac use kb G oa ra n rd O ada We th b L er C e VL arn T w E - i ise n ho us e ot he Te r kn iC Fr Fi al et rs C we tCl as om ll m Dow s er ci nin al g in tra Bo ne Lo d in t gt tu o s D n om in o C o To se Lo p C tu la s ss Le ar M ni er ng lin Sp ac C ol e lo qu ia

% of institutions using VLE type

VLE use in institutions - a comparison between 2001 and 2003

one major reason …why there has been little pedagogical innovation using these tools to date is that .. they do not …support more radical or diverse learning activities. If the design of the software environment encourages a pattern of use that mimics traditional lecturer-student roles there is little incentive for lecturers to adopt new approaches… were designed to support a model of teaching and learning interactions that was strongly based around Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 10 of 12

DOMS Project information transmission via the provision of structured content from teacher to learners and the subsequent testing of learners on the content with little consideration given to the activities that the learners themselves might engage in. p5 It was quickly realised that student and course information already held in administrative systems was involved in implementing many of the functions of a VLE. Problems associated with replication of this information - not to mention time-consuming data-entry tasks - mean that some level of integration with student records systems is highly desirable if not essential. The need for data transfer between MIS systems and VLEs is one driver behind MLE development, but not the only one. In addition there was a more general demand by both students and staff for web access to institutional information and services coupled with a desire to join-up administrative systems (e.g. finance and student records). p7 Lynch, Clifford A. "Institutional Repositories: Essential Infrastructure for Scholarship in the Digital Age" ARL, no. 226 (February 2003): 1-7. at http://www.arl.org/newsltr/226/ir.html

Institutional Repositories: Capturing and Preserving Digital Collections http://www.acrlnec.org/sigs/itig/tc_mar18_05.htm

A guide to institutional repository software, 2nd ed. (Open Society Institute, 2004) [View 3rd edition at:http://www.soros.org/openaccess/pdf/OSI_Guide_to_IR_Software_v3.pdf]. Executive Summary Managed Learning Environment Activity in Further and Higher Education in the UK http://www.jisc.ac.uk/uploaded_documents/mle-study-exec-summary.doc Institutional Repositories: Enhancing Teaching, Learning, and Research EDUCAUSE Evolving Technologies Committee Alan McCord, Lawrence Technological University / University of Michigan October 16, 2003 http://sitemaker.umich.edu/dams/files/etcom-2003-repositories.pdf Architecture of the World Wide Web, Volume One http://www.w3.org/TR/webarch/#uri-benefits Lilia Efimova & Stephanie Hendrick In search for a virtual settlement: An exploration of weblog community boundaries https://doc.telin.nl/dscgi/ds.py/Get/File-46041 Davis, J. Strategic Principles Coordinated Autonomy – A Statement of Principles for Information Technology at UCLA

http://www.oit.ucla.edu/strategicPrinciples.htm

Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 11 of 12

DOMS Project

Case studies Linden,J. Martin. S, Masters. R, Parker. R. The British Library: The large-scale archival storage of digital objects DPC Technology Watch Series Report 04-03 February 2005 Robert H. McDonald and Chuck Thomas, "Building a FSU Digital Institutional Repository: A Vision Statement" (April 16, 2003) Florida State University DScholarship Repository, Article #1. http://dscholarship.lib.fsu.edu/general/1 Definitions: Resource: an item of interest that can be identified by a global identifier called Uniform Resource Identifiers (URI).

Prepared by DOMS Team 22/06/2005 Last updated 15/07/2015

page 12 of 12