H6637 Digital Libraries & Information Portals: Wee Kim Wee School of Communication and Information
H6637 Digital Libraries & Information Portals: Wee Kim Wee School of Communication and Information
Submitted By:
Elaya Perumal Jayasakthi (G1001230J)
Peringalath Rajan Rajasree (G1001253G)
Sahu Renu (G1001258C)
Lecturer
Associate Professor Jin-Cheon Na
4. Conclusion ............................................................................................................................. 11
5. References ............................................................................................................................. 12
1. INTRODUCTION
The digital library present in the world serves as an archive for the history of data. It provides the
access to digital data’s irrespective of location. One of the major reason for the improvement and
extension of digital libraries is the wide spread internet and it holds the success of digital
libraries. Digital libraries have to manage the enormous amount of data and it has to provide the
data to the users in a much more suitable way that the users feel better than the previous
versions. But the earlier infrastructure of the digital libraries was not sufficient to support all
these activities, as a result the introduction of FEDORA has appeared. FEDORA handles the
massive data’s efficiently [2].Flexible Extensible Digital Object Repository Architecture (or
FEDORA) is an architecture over which different digital libraries, digital archives, repositories
etc. can be built. Fedora was developed by Cornell University Information Science and the
University of Virginia Library. It is a digital object and repository architecture which provides a
secure and reliable way to store and retrieve digital content. The current released version of
Fedora repository is 3.4.2. This version was released as a fix for the bugs found related to
memory and stability in the core repository services of the later versions.
2. LITERATURE REVIEW
2.1.Overview of FEDORA
Fedora is an ongoing research project. The Fedora repository system is an open source system
which uses public APIs. The digital object model of the fedora enables good management of the
metadata and the digitalized content regardless of the format of the content. Fedora supports
import and export of XML digital objects. Hence, data and information can be exchanged
between fedora and XML based applications. Fedora supports a wide variety of complex objects,
which includes all types of documents, images, electronic books, multi-media learning objects,
datasets, computer programs, and other compound information entities [4].
FEDORA is an architecture that is designed in the way that it enhances the feature of
interoperability and extensibility by perform the combination of contents efficiently. According
to staples, the FEDORA not only handles the large amount of data but it also manages the
storage and dissemination of data’s efficiently [2]. The version 1 of FEDORA system is an open
source digital repository system. According to staples, the version 1 is seems to be digital library
and absolutely not. But the later versions of the FEDORA provided lots of facility and due to
these enhanced versions so many projects have been doing in various fields by keeping the
FEDORA as the basic context.
FEDORA plays an important role in Institutional repository. During 1999, the researchers felt
worried that they could not find multiple repositories. But after the introduction of open source
repository like FEDORA the worries had been solved. Institutional repository are become more
famous now-a-days because they are involved in research and they support many scholars work
and thereby they are gaining credit and their rating also increased among other institutions. As
well as companies also supports research and development, so they are also having the digital
libraries. The open source repository FEDORA plays an efficient role to support all these above
activities [3].
The contents in the repository are managed in the form of data objects. Each of the data object
constitutes of components known as data streams which contain the content or the metadata
about the content [7]. The data objects are represented in XML files. Using Fedora the digital
data can be viewed in a variety of ways. Objects can be created in two different ways in the
fedora repository. One way is to create objects by creating workflows and the second way is to
use the interactive web client.
Fedora is usually implemented in CORBA and Java. The current version of Fedora is 3.4.
Currently the fedora project is directed by Sandy Payette and Thornton Staples and the project s
supported by grants from the Andrew W. Mellon Foundation and the Gordon and Betty Moore
Foundation.
2.2.Architecture
The base of Fedora architecture is object models. Object models are templates for data objects.
The data objects include digital resources, metadata of these resources, and an association to
software tools and services that have been designed to deliver the data and information as desired
by the users. The software connections are provided as methods encoded into two kinds of inter-
related behavior objects. The Fedora repository gives the access to the data objects. The metadata
is stored by the behavior objects and the service binding is described by using the Web Services
Description Language [1].
Object Model of Fedora
Abstraction: Data Object can represent any kind of data like text, images, audio, video, maps,
geospatial data etc. For all different kind of data, object model is same, irrespective of what
kind of data a digital object is representing.
Flexibility: Fedora is flexible due to its service oriented architecture (SOA). SOA provides
concept for construction and integration of information system and application [6], which
provides high level of flexibility of content.
Generic: Digital object are highly linked with metadata and content.
Aggregation: Objects are able to refer the data stored locally as well as from web server.
Extensibility: Core principle behind the fedora is its extensibility, Objects and behaviors are
separated from each other, and required behavior can change without changing the objects.
The Fedora has a powerful object model that can express a variety of contents and their relations.
There are two perspectives to understand the object model of the fedora, the representational and
functional perspectives. From the representational perspective the digital object’s internal
structures are not visible but the relationships of these objects can be viewed. The functional
perspective helps in knowing the way the object models are related to the management services
[7].
FEDORA is centralized in the large framework and it is separated as the set of services. The
majorly vital services are as follows:
Repository services: This service provides the functionalities like giving access to
information, making the storage process, depositing the contents.
Indexing services: This service provides the functionalities like finding the needed
information by using the indexing process.
Collection services: This service provides the way of grouping the digital objects request and
responses into meaningful groups.
Naming services: This service provides the functionality of inbuilt library facility that it
checks the digital object names and solve the problem if any exist in the name matching
process.
User interface services: This service provides the interface and their by exchange the request
and responses between the users and the machine.
There are some main components existing in the FEDORA. They are:
Repository component: This performs the function to deal with management and accessing
data objects.
Disseminators: It just transforms the content data packets into the meaningful data.
Content type: It provides the same structure to the data objects which are from different
forms and it helps to abstract the structure of the data objects.
The following are the overall important layers existing in the FEDORA architecture [3].
Structural layer
Interface layer
Right management for the digital objects
Management layer
Accessing the digital objects.
Structural layer
This is the lower layer of the architecture. This layer is mentioned as digital object structural
kernel. It manages the structural collection of digital objects. It performs the functionality of
getting the request, provide the accessing right, storage, depositing. The digital objects are the
collection of disparate data’s into an single unit and it store the data as the data stream and the
structure for these data streams are provided by the set of services called primitive disseminators.
There are three basic requests used by the primitive disseminators to give access to the structural
layer, they are:
Interface layer
This layer is mentioned as the digital object content-type disseminators. It provides the request to
access the digital object in the context of structure base. But the access to structure base is not an
safe process because it may sometime leads to change the content so the use of content type is
rendered in this layer. It separated the data into allowable and not allowable so it will give access
to only allowable data and not for the raw streams. The content creators are used to collectively
represent the content type disseminators because data are having the different streams and in turn
this would have the different content type [3].
This provides the functionality of maintaining the property of collection of data streams with no
change. Access managers are the responsible component which provide access management
control to each disseminator and give different type of access to different services according to
their content [3].
Management layer
This layer deals with the digital object life cycle and the management of these processes. The
Repository provides the access and management control [3].
Accessing the digital object
All of the above processes are ended at this level. The digital objects are created and it can be
provided to the client according to their request [3].
Client applications, the internet browsers, the batch programs and the applications in the servers
enable the users to interact with repository contents. The API’s of Fedora help these applications
to access the data and contents in the repository effectively. These API’s help in the
management, retrieval, search, access of the repository contents. The below diagram shows a
simplified form of the entire structure of the fedora repository:
2.3.Software
The fedora repository software can manage and maintain different types of contents and
metadata of the contents in all types of format. It has different storage options like databases and
file systems. It has an option known as rebuilder utility which enables disaster recovery and data
migration. The software has full text search service named as GSearch.
The fedora repository consists of three layers, the web services exposure layer, the core
subsystem layer, and the storage layer [1]. The web services exposure layer consists of
management service, access service, and access-lite service, which are web services described
using the web services definition language. The management service defines an open interface
for doing various functions related to digital objects or its components like, creating, editing and
deleting it. The access service defines an open interface for accessing and disseminating the
digital objects according to the requests from the clients. The access-lite service is implemented
as HTTP service in fedora. The core subsystem consists of management subsystem which
implements all the operations necessary for administering the digital objects and an access
subsystem which implements operations necessary to disseminate these digital object contents.
The storage layer does the reading, writing and deletion of the data from the fedora repository.
Repository Access Protocol (RAP): This protocol is responsible for accessing the repositories
and its corresponding digital object. It includes all the operations related to the digital object like
create, delete, accessing metadata, dissemination etc. It is has been defined using CORBA IDL
[Obj99] and as a binary protocol on top of TCP/IP connections [6].
The digital libraries have not provided much facility to access the database images, contents
efficiently. On the introduction of FEDORA these facilities were extended up to the needed
level.
The digital libraries lacked to support the interoperability between the contents. But the
Fedora repository provides the suitable platform to support interoperability features.
The digital libraries were also lacked in supporting the interrelationship between the
contents. But the Fedora provides the proper interrelationship between the contents.
The digital libraries had not provided the proper management functions. But the Fedora
provides excellent management features.
The earlier digital libraries were not providing the flexible platform to support the addition of
tools and concepts but the Fedora repository supports all these features.
2.5.Examples
The US National Library of Medicine that provides books and videos related to biomedical
freely over the internet implements Fedora as the core system for their digital repository.
Public Library of Science has a communication tool PLoS ONE which is an open publishing
platform for science and medicine. It is built over technologies of Fedora, Topaz and
Mulgara.
The eSciDoc project which is intended for ensuring a permanent access to research materials
has Fedora as the critical and central component in its framework.
The National Library of Singapore uses Fedora with the VITAL product from VTLS.
Rutgers University Library Community Repository is a fedora based cyber infrastructure
which is flexible and user oriented.
Data archiving and networked service is a national organization which provides access to
research materials of humanities and social sciences is built upon Fedora.
WGBH which is a public service media of New England use the Fedora prototypes for their
WGBH Media Library and Archives (MLA). With help of Fedora, the library has preserved
and catalogued all the productions of WGBH and has provided access to them.
3. FUTURE WORK
The fedora is an architecture that is being refined continuously. The current work going on is on
accommodating community defined content types and rights management mechanisms. As the
architecture of Fedora is a distributed one there are many reliability and security related issues.
These are the key issues of fedora which will be refined in the next versions of fedora. The
processes that handle the data streams need to be improved [4]. The critical areas that will be
refined are mobile code security, rights management policy definitions, fail-safe component
design, and reliable replication mechanisms.
4. CONCLUSION
This paper has described the Flexible Extensible Digital Object Repository Architecture (or
FEDORA). Fedora has made many contributions for the extensibility and interoperability of the
repositories. Fedora is designed to support extensibility. As Fedora has an open architecture, it is
easy to develop it according to new requirements. Various features of Fedora make it applicable
to various application domains. The Flexible Extensible Digital Object Repository Architecture
is an excellent conceptual framework that can effectively manage the digital contents. Fedora
provides basis for information that will last for long time and the content can be used in various
ways.
5. REFERENCES
[1] Staples T, Wayland R, Payette S. (2003), The Fedora Project An Open-source Digital Object
Repository Management System, D-Lib Magazine, 9 (4).
[2] Manafy, M. (2003). This FEDORA’s big enough for any DAM project. Econtens . 26(10), 9-
13.
[4] Payette, S., Lagoze, C. (1998). Flexible and extensible digital object and repository
architecture (FEDORA). Springer-verlag berlin Heidelberg. 41-59
[5] www.dcc.ac.uk
[6] Christophe Blanchi, Jason Petrone Corporation for National Research Initiatives Reston,
Virginia 20191{cblanchi, jpetrone}@cnri.reston.va.us
[7] https://wiki.duraspace.org/display/FCR30/Fedora+Repository