The Keystroke-Level Model (KLM) has been studied in many areas of user interaction with computers because of its validity and predictive value, but it has not been applied to estimate user performance time in mobile environment such as handheld platforms . This paper investigates and verifies the applicability of KLM on handheld tasks. The KLMs and prediction time are created using a suite of cognitive tools called CogTool. A user study on 10 participants has shown that KLM can accurately predict task execution time on handheld devices with less than 8% prediction error.

power management, displays, energy, low-power, OLED, user-interface, mobilityThe utility of a mobile computer, such as a laptop, is largely constrained by battery life. The display stands out as a major consumer of battery energy, so reducing that consumption is desirable. In this paper, we motivate and study energy-adaptive display sub-systems that match display energy consumption to the functionality required by the workload/user. Through a detailed characterization of display usage patterns, we show that screen usage of a typical user is primarily associated with content that could be d isplayed in smaller and simpler displays with significantly lower energy use. We propose example energyadaptive designs that use emerging OLED displays and software optimizations that we call dark windows. Modeling the power benefits from this approach shows significant, though user-specific, energy benefits. Prototype implementations also show acceptability of the new user interfaces among users. AbstractThe utility of a mobile computer, such as a laptop, is largely constrained by battery life. The display stands out as a major consumer of battery energy, so reducing that consumption is desirable. In this paper, we motivate and study energy-adaptive display sub-systems that match display energy consumption to the functionality required by the workload/user. Through a detailed characterization of display usage patterns, we show that screen usage of a typical user is primarily associated with content that could be displayed in smaller and simpler displays with significantly lower energy use. We propose example energy-adaptive designs that use emerging OLED displays and software optimizations that we call dark windows. Modeling the power benefits from this approach shows significant, though user-specific, energy benefits. Prototype implementations also show acceptability of the new user interfaces among users.

The relative power consumed in the WLAN interface of a mobile device is rising due to significant improvements in the energy efficiency of the other device components. The unpredictability of the incoming WLAN traffic limits the effectiveness of existing power saving techniques. This paper introduces a Power Aware Web Proxy (PAWP) architecture designed to schedule incoming web traffic into intervals of high and no communication. This traffic pattern allows WLAN interfaces to switch to a low power state after very short idle intervals. PAWP uses a collection of HTTP-level techniques to compensate any negative impact that traffic scheduling may have. PAWP does not require any client or web server modifications.In this paper, we describe our initial experiences with a PAWP implementation for 802.11b WLANs. Our experiments show savings of more than 50% in the energy consumed by the WLAN interface. Finally, our experiences give us insights into possible browser improvements when power consumption is taken into account.

No abstract

Abstract