Thesis (M.App.Sc.) - Carleton University, 2002. / Includes bibliographical references (p. 180-187). Also available in electronic format on the Internet.

Thesis (M.App.Sc.) - Carleton University, 2005. / Includes bibliographical references (p. 95-103). Also available in electronic format on the Internet.

Thesis (M.Sc. (Computer Science)) - Rhodes University, 2008.

Thesis (Ph. D.)--University of Washington, 2004. / Vita. Includes bibliographical references (p. 113-124).

Thesis (Ph. D.)--State University of New York at Binghamton, Thomas J. Watson School of Engineering and Applied Science, Department of Computer Science, 2009. / Includes bibliographical references.

Thesis (M.S.)--The University of North Carolina at Greensboro, 2009. / Directed by Fereidoon Sadri; submitted to the Dept. of Computer Science. Title from PDF t.p. (viewed May 5, 2010). Includes bibliographical references (p. 32).

Thesis (M. Sc. (Computer Science))--Rhodes University, 2006.

Recent trends show an increase in the availability and functionality of handheld devices, wireless network technology, and electronic display networks. We propose the novel integration of these technologies to provide wireless access to content delivered to large-screen display systems. Content adaptation is used as a method of reformatting web pages to display more appropriately on handheld devices, and to remove unwanted content. A framework is presented that facilitates content adaptation, implemented as an adaptation layer, which is extended to provide personalization of adaptation settings and response to network conditions. The framework is implemented as a proxy server for a wireless network, and handles HTML and XML documents. Once a document has been requested by a user, the HTML/XML is retrieved and parsed, creating a Document Object Model tree representation. It is then altered according to the user’s personal settings or predefined settings, based on current network usage and the network resources available. Three adaptation techniques were implemented; spatial representation, which generates an image map of the document, text summarization, which creates a tree view representation of a document, and tag extraction, which replaces specific tags with links. Three proof-of-concept systems were developed in order to test the robustness of the framework. A system for use with digital slide shows, a digital signage system, and a generalized system for use with the internet were implemented. Testing was performed by accessing sample web pages through the content adaptation proxy server. Tag extraction works correctly for all HTML and XML document structures, whereas spatial representation and text summarization are limited to a controlled subset. Results indicate that the adaptive system has the ability to reduce average bandwidth usage, by decreasing the amount of data on the network, thereby allowing a greater number of users access to content. This suggests that responsive content adaptation has a positive influence on network performance metrics.

Applications that use directory services or relational databases operate in client-server mode where a client requests information from a server, and the server returns a response to the client. Communication between each client-server application is achieved by using separate custom built front-ends with non-portable data formats. A need exists to access information from different heterogeneous client-server systems in a standard message request-response format. This research proposes a generic XML document that presents a common request-response interface to the client from which they can access network protocol or database information. The XML component is easily adaptable to accessing any new client-server type protocol or database data that may be added to a server. The approach in determining the XML elements is, firstly review each systems command and data structure separately, and then determine if there are any commonalities within each protocol that would allow for a common representation of both the data and command structure. For the purposes of this project, three different data sources that are typically used in an Internet application were analysed, namely: -- a TCP based server program; -- a relational type database; and -- a directory service. The solution was implemented using Linux as the operating system, Java as the programming language, MySQL as the relational database, openLDAP as the directory server and a proprietary TCP based server application. Initially the complete system was developed for the proprietary TCP-based application. The other systems were added with minimum additional work. The result of the implementation was that it is relatively easy to add new protocols (for e.g. LDAP) on an as needed basis with minimal changes required on the server side. A client will receive XML responses that the client can either adapt (typically using a separate style-sheet) to their specific needs or use the existing front-ends if they are suitable. After the design was implemented and tested, the performance of XML and non-XML messages was evaluated. As expected the increased verbosity of XML results in a larger footprint that requires more processing time and resources. This means that any implementation using XML has to carefully weight the benefits of flexibility, extensibility and standard message formats against reduced performance. After evaluating XML type messages in an Internet type environment that involved human-computer interaction, it was concluded that the slower response times is not that significant to negate the benefits of a common message interface provided by using XML. / Dissertation (M Eng (Computer Engineering))--University of Pretoria, 2006. / Electrical, Electronic and Computer Engineering / unrestricted

This was a project to develop a Telephone Directory Web service (TDWS) to provide convenient and cost-effective access to public telephone directory data.