28 Digital Library Case Study: DSpace

Yatrik Patel


I.  Objectives


Objectives  of  this  module  is  to  impart  knowledge  on  the  following  aspects  of  DSpace Institutional Repository Software:


•   Key factors to Dspace adaptation;

•   DSpace: Major features and functionalities;

•   DSpace system architecture; and

•   Installation, configuration and customization.



II.  Learning Outcomes


After going through this lesson, users would attain knowledge about the major features and functionalities ofDSpace Institutional Repository Software, its installation, configuration and customization process. Learners would be equipped with the knowledge of collection building process using DSpace.



III.   Structure


1.     Introduction

2.    Key Factors to DSpace’s Adaption

3.    DSpace Information Model

3.1   DSpace System Architecture

4.    Major Features of DSpace

4.1   Metadata Registry

4.2  File Format Registry

4.3  E-Persons

4.4  Authorization

4.5  Ingestion Process and Work Flow

4.6  Search and Browse

4.7  Handle System

4.8  OAI-PMH Support

4.9  Statistics

4.10  SWORD and Open URL Support

5.   Customization in Dspace

6.   Some Live Examples

7.    Summary





1. Introduction 


Digital libraries are nothing but services related to management and organization of available digital information and its retrieval with proper user interfacing. It also includes archiving and preservation of digital material, social issues attached to the same along with its application and evaluation to specific focused areas. To achieve all there are certain expectation from the software or solution which helps to create digital libraries.


For a proper digital library solution, primarily it is being expected that


• It should be cost effective in terms of hardware and software platform to be procured and management there after.

• The digital library software is also expected to be technically simple, easy to install and manage, so that a layman with working knowledge of information technology can install and administer the same.

• The solution should be robust and scalable in such a way that, it can handle large volume of data seamlessly along with inter-operable modular open architecture so that the necessary customization can be done easily without being dependent on software specialists.

• Digital library software should have user friendly, multi user interface so that multiple entities can use and administer the software simultaneously.

• It is also desirable that it should be platform independent so it can run on any popular software and hardware platform.

• Last but not least, it must have the capability to handle all types of digital object, the object can be data set, document, multimedia or say any digital format.


There is one stop solution, which meets all the expectations described above, that is DSpace, It is a platform that;


•   Captures items in any format and distributes it over the web,

•   Indexes digital items so user can easily search and retrieve them,

•   Preserve the digital content over the long term.


DSpace is typically being used to create a digital library with three major roles; First, it facilitates the capture and ingestion of material with associated metadata; Second, DSpace provides easy access to the material with user friendly searching and listing mechanisms; Third, it facilitates long term preservation of digital material.


When initiated (in year 2000), DSpace was a joint project of the Massachusetts Institute of Technology and Hewllet-Packard , DSpace project is now being handled by DuraSpace, a non-for-profit organization.


2. Key Factors to DSpace’s Adaption


DSpace has become quite popular among digital library implementers because it is open source and freely available software; it is being backed by very large worldwide user community who are ready to help.


DSpace software has been packaged in a way that it is very easy to use, It handles content in number of digital formats, and the major advantage is; contents in DSpace can be made searchable through search engines like Google scholar, thus one can increase outreach of digital library without much effort.


DSpace can be used to store any type of digital material, it can store journal papers, Data Sets, Electronic Theses, Reports, Conference posters, Video’s, images. Logically speaking DSpace can be used to store any material which is available in digital format.


DSpace is basically an open source software available under Berkeley Software Distribution (BSD)licence , where in one can use and redistribute source as well as binaries or executable programmes. DSpace software can be obtained from www.dspace.org or SourceForge’s dspace project site.


DSpace is having a community based development model, where in there is SVN (subversion) based common source code control repository, which is having dedicated committers and contributors. This developer community welcomes every one to submit bug reports, patches, feature requests and other related things. There are a number of active discussion groups and email lists are available for dspace support.


3. DSpace Information Model 


Information Model of dspace is broadly divided in four components, that is Communities, Collections, Items and Bitstreams.


Community reflects the unit of an organization, collection in each community is distinct grouping of items, Items are logical content objects where as bitstreams are individual files.


The way data is organized in DSpace is supposed to reflect the structure of the organization and its digital collection.


Each DSpace site is divided into communities, which can be further divided into sub- communities reflecting the typical university structure of college, department, research centre, or laboratory. Communities contain collections, which are groupings of related content. A collection may appear in more than one community. Each collection is composed of items, which are the basic archival elements of the archive. Each item is owned by one collection. Additionally, an item may appear in additional collections; however every item has one and only one owning collection.


Items are further subdivided into named bundles of bitstreams. Bitstreams are, as the name suggests, streams of bits, usually ordinary computer files.


As discussed earlier, communities and collections are used to organize digital content or items in to a hierarchical form, It also contain limited set of descriptive metadata like name, description, licence and availability for that particular community and collection etc.digital library implementer can create communities based on logical grouping of digital items and then further subdivide in to the collections.


Items are logical units of content which consist Dublin core based metadata, as well as other metadata which has been encoded as bitstream. The item can be an electronic thesis, an e- book, photographs, a complete web page which can include images and style sheets associated with HTML page.Each item can contain one or multiple files along with metadata.


DSpace facilitates all three types of metadata that is Descriptive, Administrative and Structural. Descriptive metadata can by any thing which describes the item; it takes care of all elements of DublinCore metadata set. Dspace also supports non Dublin Core metadata; these elements may not be searchable. Administrative metadata associated with an item can be access restrictions, means who can access, remove or modify an item, there is no standard format available for these type of metadata.Structural metadata describes very basic attributes about the item, for example what are the bitstream that contained in an item, or under which community and collection that item belongs to.


Bitstreams are individual digital files, which are having limited set of descriptive metadata like name, size of the file, format of the file etc. A pdf file, word doument, jpeg or bmp picture, executable program etc. can be considered as bitstream.


Bundles are basically nothing but a group of related files, for example, when you have an HTML page, it may also contain link to other HTML documents, images, flash object etc; therefore to view that html page, you also need to have all associated files, therefore html file along with other associated files makes a bundle. Dspace doesn’t support any metadata for bundles.


3.1  DSpace System Architecture 


The architecture of DSpace has been divided in to three major parts, that is DSpace public API on the top, business logic layer in the middle and Storage API at the bottom.

(Image Courtesy: http://www.dspace.org)


DSpace public API takes care of user interfacing and services, it contains components for web user interfacing, Federation services, Metadata providing services as defined by Open Archive Initiative’s protocol, interfaces for web services like SWORD which is Simple Web- service Offering Repository Deposit etc.


These public API components interacts with business logic layer in the middle, which provides search and browse components, Handle manager, History manager which takes care of logs and statistics. Business logic layer also contains components to manage ingestion process and workflow, components to manage e-persons, groups and their authorization along with content management API and administrative toolkit.


The bottom layer, the storage api has been two major components, one is Relational Database Management System wrapper which connects to RDBMS like Postgres or Oracle through Java Database Connectivity and the Bitstream storage manager which directly interact with the file system to store bitstreams.


4.   Major Features of DSpace


4.1  Metadata Registry


Dspace provides facility to create a new metadata registry or manage existing dublin core metadata registry, where in digital library implementer can manage and customize metadata elements. Metadata registry has three major components that is Schema, Element and Qualifier.


4.2  File Format Registry


In addition to metadata registry dspace also features registry to handle file formats, this file formats can be managed based on three levels that is supported, known and unknown. Here dspace administrator can specify MIME type, Name, Long Description and support level of the file along with file extension.


4.3  E-Persons


Persons or users who interact with dspace are called e-people, it is basically dspace user accounts, dspace provides facility to permit these e-people to login to the site, sign up to receive notification changes to subscription, submit new items to the collections,Administer collection / communities or entire dspace site.These e-people can also be managed by forming  groups.


4.4  Authorization


Authorization system in dspace enables administrators to give e-people the ability to perform add and remove operation by which an e-person can remove or add any community, collection or item.


As a collection administrator e-person can edit an item’s metadata, withdraw items or can map the items in to the collectionWrite permission enables e-person to add or remove bitstreams, where as read permission enables only reading of bitstreams.


4.5  Ingestion Process and Work Flow


Ingestion process is nothing but getting or putting contents in to Dspace. Dspace facilitates batch import as well as web based submission.


In batch import, multiple items can be submitted to dspace in one shot, this requires item to be a in specific format along with metadata encoded in XML.Whereas in web based submission only one item can go each time, the item being submitted has to go through a work flow process defined for that collection. Assume that there are three steps




1: May reject the submission Step

2: Edit metadata or reject Step

3: Edit Metadata

Image Courtesy http://www.dspace.org


A collection’s workflow can have up to three steps as shown in the figure above. Each collection may have an associated e-person group for performing each step; if no group is associated with a certain step, that step is skipped. If a collection has no e-person groups associated with any step, submissions to that collection are ingested straight into the main archive.


In other words, the sequence is this: The collection receives a submission. If the collection has a group assigned for workflow step 1, that step is invoked, and the group is notified. Otherwise, workflow step 1 is skipped. Likewise, workflow steps 2 and 3 are performed if and only if the collection has a group assigned to those steps.


When a step is invoked, the submission is put into the ‘task pool’ of the step’s associated group. One member of that group takes the task from the pool, and it is then removed from the task pool, to avoid the situation where several people in the group may be performing the same task without realizing it.


4.6    Search and Browse


DSpace allows end-users to discover content in a number of ways, that is via external reference, such as a handle searching for one or more keywords in metadata or extracted full- text. It also provides an option for browsing through title, author, date or subject indices, with optional image thumbnails etc.


4.7    Handle System


In present web world Universal Resource Location – URL of a digital content may change due to the changes in hardware or software, change in network or because of political change. This can be handled by creating a permanent URL independent of the repository. Handle system in DSpace provides a persistent handle for each item, if configured properly.


4.8    OAI-PMH Support


The Open Archives Initiative has developed a protocol for metadata harvesting. This allows sites to programmatically retrieve or ‘harvest’ the metadata from several sources, and offer services using that metadata, such as indexing or linking services. Such a service could allow users to access information from a large number of sites from one place. DSpace exposes the Dublin Core metadata for items that are publicly accessible. Additionally, the collection structure is also exposed via the OAI protocol’s ‘sets’ mechanism.


4.9    Statistics


DSpace offers system statistics for administrator usage, as well as usage statistics on the level of items, communities and collections. Dspace also provide customizable general overview of activities in the archive, by default including:


•   Number of items archived

•   Number of bitstream views

•   Number of item page views

•   Number of collection page views

•   Number of community page views

•   Number of user logins

•   Number of searches performed

•   Number of license rejections

•   Number of OAI Requests


4.10  SWORD and Open URL Support


SWORD (Simple Web-service Offering Repository Deposit) is a protocol that allows the remote deposit of items into repositories. SWORD was further developed in SWORD version 2 to add the ability to retrieve, update, or delete deposits. DSpace supports the SWORD protocol via the ‘sword’ web application and SWord v2 via the swordv2 web application.


DSpace supports the OpenURL protocol from SFX, in a rather simple fashion. If your institution has an SFX server, DSpace will display an OpenURL link on every item page, automatically using the Dublin Core metadata. Additionally, DSpace can respond to incoming OpenURLs too.


Customization in DSpaceDspace provides various flexibility and customization options, the areas that can be customized are;


•   Submission  process,  in  which  one  can  configure  the  submission  steps  to  suit  the organization.

•   One can also customize Browse and search terms in which fields and files can be chosen to index and display in the browse interface.

•   Dspace also provides flexibility to choose Database; one can choose Postgres or Oracle.

•   DSpace  can  be  customized  to  work  with  other  web  services-  using  Light  Network Interface one can pull or push content to or from DSpace

•   One can create an own user interface for interacting with DSpace.


Some Live Examples




In this module, we have seen how DSpace can be helpful in building digital libraries, along with Key factors to Dspace adaptation; its major features and functionalities. We have also studied DSpace system architecture; and its installation, configuration and customization. Along with some live examples.