Paper Abstract - Stina Nylander & Annika Waern

This page is part of the CRIC web archive

PLAYING WITH THE FUTURE:
DEVELOPMENT AND DIRECTIONS
IN COMPUTER GAMING

HOME | PRESS RELEASE | ABSTRACTS | PROGRAMME | PHOTOS

ABSTRACT

Platform-Independent Interaction for Online Game Services

Stina Nylander & Annika Waern

The computer games of today are often 'multi-platform' games. PC games are ported to consoles and vice versa, and games on different platforms can interact with each other. The problem with this approach is that development and maintenance gets multiplied - the game must partly be reimplemented for each new platform.

Internet-based applications typically take another approach. Here, applications run at servers, distributing the user interface to the device in a way that is device independent. Web browsers and most multimedia technology use this approach. The problem with this approach is that applications are constrained to a 'lowest common denominator'. The interactivity is limited to what can be expressed in the presentation language (e.g. HTML or WML) and response times are limited both by browser processing speed and communication lag.

We have sought a way to combine the two approaches. We separate the application from the presentation by a language that expresses user interaction, rather than application presentation.

The interaction language consists of a basic set of interaction acts, each of which is independent both of device type, type of presentation system, and application. At run-time, applications offer interaction acts of different types and in different groupings, which are mapped to presentation templates, and used to generate a user interface.

The basic set contains seven interaction acts: create, destroy, move, input, output, selection and modification. Create, destroy and move are operations on an object space (creating new objects, destroying objects, move objects around), while input, output, selection and modification are operations on single objects, and offered by the object space for particular objects. The set of basic interaction acts does not contain navigation acts, although navigation is fundamental in interaction. There are two very different models for how a virtual space can be navigated. The user can be seen as having a position in space and moving through it. But the user can also be seen as residing entirely outside the virtual space, using tools to manipulate the representation of the space. A game interface will often mix the two models of control. The meta-level game controls (quit, high score information etc) are manipulated entirely from the 'outside', while the user's view of the game world is created from a position in the virtual space.

We are currently working on an implementation of interaction acts. The implementation consists of a presentation system that is configurable by the application, and dynamically controlled by the application that offers interaction acts. The implementation builds upon a previous version that implemented a remote browser for a subset of the interaction acts. To extend the approach to games, it also has been necessary to distinguish between the browser-application interaction and the client-server distribution. The presentation system under development will be able to work integrated with a local application as well as function as a browser for a server-based application.

Back to Abstracts page

Top

CRIC is now proud to be part of the Manchester Institute of Innovation Research (MIoIR)
Centre for Research on Innovation and Competition (CRIC), The University of Manchester,
Harold Hankins Building, Booth Street West, Manchester M13 9QH, England
Phone +44 (0)161 275 7365 Fax: +44 (0) 161 275 7361
Site maintained by: Ishty Hussain