Network lag is a fact of life for networked games. Lag can cause game states to diverge at different nodes in the network, making it difficult to maintain the illusion of a single shared space. Traditional lag compensation techniques help reduce inconsistency in networked games; however, these techniques do not address what to do when states actually have diverged. Traditional consistency maintenance (CM) does not specify how to make game- critical decisions when players' views of the shared state are different, nor does it indicate how to repair inconsistencies. These two issues -- decision-making and error repair -- can have substantial effects on players' gaming experience. To address this shortcoming, we have characterized a range of algorithmic choices for decision- making and error repair. We report on a study confirming that these algorithms can have significant effects on player experience and performance, and showing that they are often more important than degree of consistency itself.

Lag compensation algorithms used in networked games require programmers to manage the complexities of dealing with both time and shared state. This can make implementing lag compensation techniques challenging. The difficulties in expressing these algorithms limit experimentation with different algorithms and inhibit programmers from exploring the space of the algorithms and testing their effects. The solution is to have a programming model that is better able to deal with time. In this paper, we present such a programming model, timelines . Timelines dramatically reduce the time and effort required to implement lag compensation techniques by allowing for the explicit treatment of time. The timelines model has been implemented as part of the Janus toolkit.

We report on the design of a novel station supporting the play of exercise video games (exergames) by children with cerebral palsy (CP). The station combines a physical platform allowing children with CP to provide pedaling input into a game, a standard Xbox 360 controller, and algorithms for interpreting the cycling input to improve smoothness and accuracy of gameplay. The station was designed through an iterative and incremental participatory design process involving medical professionals, game designers, computer scientists, kinesiologists, physical therapists, and eight children with CP. It has been tested through observation of its use, through gathering opinions from the children, and through small experimental studies. With our initial design, only three of eight children were capable of playing a cycling-based game; with the final design, seven of eight could cycle effectively, and six reached energy expenditure levels recommended by the American College of Sports Medicine while pedaling unassisted.

The presence of network latency leads to usability problems in distributed groupware applications. Example problems include difficulty synchronizing tightly-coupled collaboration, jarring changes in the user interface following the repair of conflicting operations, and confusion when participants discuss state that appears differently to each of them. Techniques exist that can help mitigate the effects of latency, both in the user interface and the groupware application. However, as these techniques necessitate the manipulation of state over time, the effort required to implement them can be significant. In this paper, we present timelines, a programming model allowing the explicit treatment of time in groupware applications. The model has been implemented as part of the Janus toolkit.

Consistency maintenance (CM) techniques are a crucial part of many distributed systems, and are particularly important in networked games. In this paper we describe a framework of the human factors of CM, to help designers of networked games make better decisions about its use. The framework shows that there is wide variance in the CM requirements of different game situations, identifies the types of requirements that can be considered, and analyses the effects of several consistency schemes on user experience factors. To further explore these issues, we carried out a simulation study that compared four CM algorithms. The experiment confirms many of the predictions of the framework, and reveals additional subtleties of the algorithms. Our work is the first to look comprehensively at the tradeoffs and costs of CM, and our results are a strong starting point that will help designers improve on the user's quality of experience in distributed shared environments.