188宝金博页面版

Window Shopping: A Study of Desktop Window Switching Andrew Warr, Ed H. Chi, Helen Harris, Alexander Kuscher, Jenn Chen, Robert Flack, Nicholas Jitkoff Google Inc. 1600 Amphitheatre Parkway, Mountain View, CA 94043 {andywarr, edchi, helenharris, kuscher, jennschen, flackr, alcor}@google.com ABSTRACT Desktop users frequently open and switch between multiple windows. Here we present an experiment comparing 3 window switching interfaces: the Cards interface spreads windows out like a vertical stack of cards with the most recent window at the front; the Exposé interface provides an map-like overview based on the relative size and position of windows; and, the Mosaic interface places each window in a grid ordered by recency. Experimental results suggest that the Mosaic interface scales, enabling faster window selection than the Cards interface and less erroneous window selection than the Exposé interface. Author Keywords Window Management, User Interfaces ACM Classification Keywords H.5.2 User Interfaces INTRODUCTION AND RELATED WORK Multiple application windows are a ubiquitous feature of modern desktop operating systems, yet users often find switching between windows problematic. Since windows frequently obstruct other windows, and users often forget the location of windows, users often end up hunting for a specific window and the problem exacerbates as the number of windows increases. For example, Hutchings et al. [3] showed that participants in their study had 8 or more windows open 78.1% of the time, 3.5 of which were visible at any time on a single display. Highlighting the frequency of window switching, their results show windows were active for an average of 20.9 seconds and the median activation time took 3.8 seconds. The 1980s was a golden age researching windowing systems in HCI. Early work showed that tiled windowing systems are quicker than overlapping windows [1]. Later research showed tiled windowing systems were also faster when there are large numbers of windows [4]. This is intuitive since there are no overlapping windows to manage. Despite this, users generally prefer overlapping windows [1]. Therefore, it is imperative to create windowing systems that are both efficient for the user and meet their preferences. In one early work, Card & Henderson [2] argued for designs that reduce the time and cognitive load of switching tasks and context. In their system called Rooms, a virtual window manager keeps track of active sets of windows relating to tasks and users could easily switch between different ‘rooms’ to complete their associated tasks. Later researchers investigated providing the user with an overview of open windows, as well as user-defined sets of windows that might be associated with some task [5, 6, 7]. The Task Gallery [6] allowed windows to be grouped as virtual artworks representing tasks and arranged them in a virtual gallery, thereby leveraging human spatial cognition and perception. Scalable Fabric [5] used the periphery of the screen to hold scaled down groups of live windows that represented tasks. Adhering to the ‘window sets’ idea, a system called GroupBar [7], allowed users to organize windows into groups in Windows' TaskBar. More recently, SCOTZ [8] presents windows in spatially consistent layouts, such as a spiral treemap, squarified treemap and a grid. Inspired by these prior works, here we investigate the efficiency of three window switching interfaces (two of which we were considering for a new operating system). It is not the purpose of this paper to be a comprehensive design exploratory of window switching interfaces. Instead, our focus is to evaluate the window switching interfaces being considered for a new operating system, as well as a popular window switching interface, with focus on understanding the switching costs [2]. Cards: Warr and Chi [9] examined the Cards interface for mobile browser tab switching. Cards show a vertical stack of tabs/windows for users to touch and browse through before selecting one as the main window. We considered this design for desktop window management, where windows would be spread out like a stack of cards with the most recent window at the front, as depicted in Figure 1a. Since it performed well on mobile devices [9], a consistent experience across two devices might be preferable to users.Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Permissions@acm.org. CHI'16, May 07-12, 2016, San Jose, CA, USA © 2016 ACM. ISBN 978-1-4503-3362-7/16/05…$15.00 DOI: http://dx.doi.org/10.1145/2858036.2858526 Designing for Attention and Multitasking #chi4good, CHI 2016, San Jose, CA, USA3335