Information Technology Reference
In-Depth Information
tor interface. With this in mind, our framework
can provide the basic administration interface to
manage the digital resources including add, delete,
update and search functions. Other administration
functions and interfaces have to be customized to
be implemented.
Ajax (shorthand for asynchronous JavaScript
and XML) is a group of interrelated web develop-
ment techniques used on the client-side to create
interactive web applications. With Ajax, web
applications can retrieve data from the server
asynchronously in the background without inter-
fering with the display and behavior of the existing
page. In our framework, we import two common
libraries to support Ajax: Prototype and JQuery.
In this architecture, standards are very impor-
tant for the interoperability. In consideration, we
follow the OSOA standards during the building
and maintenance of our DRSC model. Based on
these standards, we provide the specification of
digital resources, which are used to describe the
atomic service components and the composition
rules. For each service component, we give out the
specification of the properties, service interfaces
and references. From these standards and speci-
fications, the DRSC model can be implemented
in different languages and environments and still
easily communicate with each other.
Besides the support of standards and specifica-
tions, we also consider the security, testing and
maintenance in the architecture. Our framework
provides two kinds of security methods: role-based
authentication and encryption-based content secu-
rity. In the system, users are assigned one or more
predefined roles. These roles then determine the
user's privileges, the information they can view
and access, and items they are able to change.
We provide an interface for the administrator to
define different roles. To enable the authentication,
we add a filter so as to invoke the Struts actions.
Therefore, every request is checked according to
the privilege list. On the web pages, fine-grained
access control is applied for the display of the page
content. In addition, we also afford administrators
with the ability to drag and drop privileges from
one role in order to create a new role.
Users may worry about the security of the
content host server because if the host server is
attacked, the content on the server may be accessed
by some unauthorized person. The encryption-
based method solves this problem by encrypting
the content as it is uploaded to the clouds. Even
if the attacker obtains the content, they should not
be able to decrypt the content. After the content is
downloaded from the server, using the key stored
locally, the content is decrypted automatically.
For specific content, one can choose whether to
encrypt it or not. Currently, we only support the
encryption using Winzipaes from Apache.
In order to build and deploy the application, we
use Apache Ant. Apache Ant is a software tool used
to automate the software build processes. More
so, it uses XML to describe the build process and
its dependencies. Ant also includes a task JUnit,
which is a simple, open source framework to write
and run repeatable tests. Therefore, the framework
supports test-driven programming. Based on JU-
nit, Mule ESB extends JUnit test case in order to
test the services without embedding them into a
J2EE container. This set of automatic testing tools
support a smoking test that can keep the software
usable and qualified for release.
After the release of an Institutional Reposi-
tory, we continue to maintain it by correcting
coding errors, improving the performance and
extending the functions of the software. We have
a specification of maintenance which defines the
detailed processes, such as how to release the
patches, and how to add a new function. Releasing
an Institutional Repository is only the first step
of its lifecycle, as the continuous support is very
important. We investigated how to set up a wiki
to help maintenance.
After the institutional repository runs, there
may be a large amount of legacy data that needs
to be imported into the new system. Thusly, we
provide a tool that generates the code needed to
import the legacy data into the new system from
Search WWH ::
Custom Search