ABSTRACT

With an explosive increase in the number of mobile devices in use and the development of communication infrastructure, the demand for getting an online service through the wireless network is growing rapidly. As most online services are available through the Web these days, accessing the Web contents using the mobile devices is one of the most frequent activities for the mobile users. However, existing Web contents created for the desktop PCs or workstations in mind may not be suitable for many mobile devices like cellular phones or PDAs due to some constraints such as the low bandwidth, small memory, small size of the display screen and low resolutions. Creating separate Web contents for mobile devices using the wireless internet will require additional time and efforts. Therefore, it would be worthwhile to develop an automated tool that can intelligently convert the existing Web contents in HTML to the contents in some markup language like WML[1] which is appropriate for mobile devices. WML is a public standard markup language developed by the Wireless Application Protocol Forum to represent applications that will be sent to the mobile devices. In addition to WML, the markup languages supported by mobile devices include mHTML, HDML, cHTML, and XHTML which is proposed by WAP 2.0. The HTML converter presented in this paper is designed to overcome some limitations with existing approaches or systems, such as an inappropriate handling of the Web contents with frames and image maps. In many Web pages, menus are placed in the left frame, and HTML documents linked from the menus are placed in the right frame. But, traditional HTML converters show the values of attributes in HTML frame tags instead of the contents on the pages. Users can access the desired contents when they select the attribute values regardless of the context of the contents. These converters do not produce an appropriate user interface and interoperability that the original HTML contents with the frames provide for the users. Thus, our conversion scheme has been

designed to provide a user interface and interoperability similar to the original frames by displaying the contents in the first page. In addition, our HTML converter can deal with the image maps in the Web page, which is one of the features not currently supported by other existing converters. Plain text contents are implemented by the tag conversion method.