Ericsson Position Paper - Mobile Web Initiative Workshop

Ericsson is a strong believer in the mobile web - as we are in all types of mobile services. The web on mobiles is one of several services, contrary to the Internet where it is the dominant service, rivalled only by peer-to-peer file transfer and email. PoC (Push to Talk over Cellular), mobile games, ringtone downloads, and a number of other services make the mobile usage different and richer than the Internet environment. The different environment creates a different user experience. The mobile web and mobile Internet content must be designed with a mobile user experience in mind.

The key to understanding the mobile web is to understand that the usage situation is different. Data supporting this go back to the earliest days of iMode, when Japanese college students - at that time the only users to have both Internet and mobile access - were shown to have very different expectations and usage patterns of mobile and fixed information services. While the different usages can be realized using the same technologies, this does not imply that you automatically can push one solution to another system. Just imagine trying to surf the web on a laptop when you are standing in a busy street corner, waiting for a green light.

The key is not the adaption of the formatting and providing the full content from existing web pages on mobile devices, but the selection and adaptation of the information. Mobile web usage will always appear disabled, if you start by trying to show pages developed for devices with very different capabilities, and users in very different situations. If the fixed web page is like a Word document, the mobile page is like a Powerpoint slide. Both can, as a matter of fact, be realized using XHTML and CSS, but you have to adapt the information and the formatting for it to work in the mobile environment. And especially, you have to adapt the content. This is where the Semantic Web may finally come into its own.

That said, the web as it is today, based on HTML at worst, at best on XHTML and CSS, is posed for major changes as the text of web pages interact with other information types, such as SVG diagrams, SMIL multimedia shows, PNG pictures, AMR sounds, and so on. To create a rich, satisfying mobile user experience, we need to go beyond the current text-and-image-display paradigm, into a multimodal presentation and interaction.

Much remains to be done to realize this. These include important things which affect the user interaction, such as an event model which is lightweight, device-, and modality-independent; a model for presentation of content which truly leverages multimodality - and which includes the existing data types. It also includes authoring and service creation issues, such as a way of including the users' context (location, presence, etc); and standardized charging and DRM functions.

The OMA is very well on the way to realize many of these functions, and has an excellent model to capture requirements and test interoperability of implemented specifications. It would make sense to leverage some of the W3C:s well-known specification prowess and development savvy, and leverage the requirements and interoperability processes of OMA, to enable the two organizations together to create the true mobile interaction environment of the future.

It would be of great detriment if the overlap was to increase, and functions in both organizations were to be duplicated. As members of both the W3C and OMA, that would be of concern to us. We applaud any initiatives directed at cooperation, but would be concerned if the result was otherwise.