As a rule, user interface designers work iteratively. Over the course of a project, they repeatedly gather feedback, typically through in-person meetings, and update their designs accordingly. Through formative work, we find that design software tools do not support designers in managing meeting notes and previous design iterations as a cohesive whole. This causes designers to rely on ad-hoc practices for organizing work, which makes it hard for them to keep track of relevant feedback and explain their design decisions. To address this problem, we present Charrette, a system that allows designers to curate design iterations, attach meeting notes to the relevant content, and navigate sequences of design iterations with the associated notes to facilitate in-person discussions. In an exploratory user study, we evaluate how Charrette affects designers' self-reported ease in handling feedback during face-to-face discussions, compared with using their own tools. We find that using Charrette correlates with increased confidence and recall in discussing previous design decisions.

Interface designers often use screenshot images of example designs as building blocks for new designs. Since images are unstructured and hard to edit, designers typically reconstruct screenshots with vector graphics tools in order to reuse or edit parts of the design. Unfortunately, this reconstruction process is tedious and slow. We present Rewire, an interactive system that helps designers leverage example screenshots. Rewire automatically infers a vector representation of screenshots where each UI component is a separate object with editable shape and style properties. Based on this representation, the system provides three design assistance modes that help designers reuse or redraw components of the example design. The results from our quantitative and user evaluations demonstrate that Rewire can generate accurate vector representations of interface screenshots found in the wild and that design assistance enables users to reconstruct and edit example designs more efficiently compared to a baseline design tool.

Pixel-based methods offer unique potential for modifying existing interfaces independent of their underlying implementation. Prior work has demonstrated a variety of modifications to existing interfaces, including accessibility enhancements, interface language translation, testing frameworks, and interaction techniques. But pixel-based methods have also been limited in their understanding of the interface and therefore the complexity of modifications they can support. This work examines deeper pixel-level understanding of widgets and the resulting capabilities of pixel-based runtime enhancements. Specifically, we present three new sets of methods: methods for pixel-based modeling of widgets in multiple states, methods for managing the combinatorial complexity that arises in creating a multitude of runtime enhancements, and methods for styling runtime enhancements to preserve consistency with the design of an existing interface. We validate our methods through an implementation of Moscovich et al.'s Sliding Widgets, a novel runtime enhancement that could not have been implemented with prior pixel-based methods.

Pixel-based methods have the potential to fundamentally change how we build graphical interfaces, but remain difficult to implement. We introduce a new toolkit for pixel based enhancements, focused on two areas of support. Prefab Layers helps developers write interpretation logic that can be composed, reused, and shared to manage the multi-faceted nature of pixel-based interpretation. Prefab Annotations supports robustly annotating interface elements with metadata needed to enable runtime enhancements. Together, these help developers overcome subtle but critical dependencies between code and data. We validate our toolkit with (1) demonstrative applications and (2) a lab study that compares how developers build an enhancement using our toolkit versus state of the art methods. Our toolkit addresses core challenges faced by developers when building pixel based enhancements, potentially opening up pixel based systems to broader adoption.

We present a general-purpose implementation of a target aware pointing technique, functional across an entire desktop and independent of application implementations. Specifically, we implement Grossman and Balakrishnan's Bubble Cursor, the fastest general pointing facilitation technique in the literature. Our implementation obtains the necessary knowledge of interface targets using a combination of pixel-level analysis and social annotation. We discuss the most novel aspects of our implementation, including methods for interactive creation and correction of pixel-level prototypes of interface elements and methods for interactive annotation of how the cursor should select identified elements. We also report on limitations of the Bubble Cursor unearthed by examining our implementation in the complexity of real-world interfaces. We therefore contribute important progress toward real-world deployment of an important family of techniques and shed light on the gap between understanding techniques in controlled settings versus behavior with real-world interfaces.

The rigidity and fragmentation of GUI toolkits are fundamentally limiting the progress and impact of interaction research. Pixel-based methods offer unique potential for addressing these challenges independent of the implementation of any particular interface or toolkit. This work builds upon Prefab, which enables the modification of existing interfaces. We present new methods for hierarchical models of complex widgets, real-time interpretation of interface content, and real-time interpretation of content and hierarchy throughout an entire interface. We validate our new methods through implementations of four applications: stencil-based tutorials, ephemeral adaptation, interface translation, and end-user interface customization. We demonstrate these enhancements in complex existing applications created from different user interface toolkits running on different operating systems.

Current chasms between applications implemented with different user interface toolkits make it difficult to implement and explore potentially important interaction techniques in new and existing applications, limiting the progress and impact of human-computer interaction research. We examine an approach based in the single most common characteristic of all graphical user interface toolkits, that they ultimately paint pixels to a display. We present Prefab, a system for implementing advanced behaviors through the reverse engineering of the pixels in graphical interfaces. Informed by how user interface toolkits paint interfaces, Prefab features a separation of the modeling of widget layout from the recognition of widget appearance. We validate Prefab in implementations of three applications: target-aware pointing techniques, Phosphor transitions, and Side Views parameter spectrums. Working only from pixels, we demonstrate a single implementation of these enhancements in complex existing applications created in different user interface toolkits running on different windowing systems.

Paper prototyping offers unique affordances for interface design. However, due to its spontaneous nature and the limitations of paper, it is difficult to distill and communicate a paper prototype design and its user test findings to a wide audience. To address these issues, we created FrameWire, a computer vision-based system that automatically extracts interaction flows from the video recording of paper prototype user tests. Based on the extracted logic, FrameWire offers two distinct benefits for designers: a structural view of the video recording that allows a designer or a stakeholder to easily distill and understand the design concept and user interaction behaviors, and automatic generation of interactive HTML-based prototypes that can be easily tested with a larger group of users as well as "walked through" by other stakeholders. The extraction is achieved by automatically aggregating video frame sequences into an interaction flow graph based on frame similarities and a designer-guided clustering process. The results of evaluating FrameWire with realistic paper prototyping tests show that our extraction approach is feasible and FrameWire is a promising tool for enhancing existing prototyping practice.

Computer users with motor impairments face major challenges with conventional mouse pointing. These challenges are mostly due to fine pointing corrections at the final stages of target acquisition. To reduce the need for correction-phase pointing and to lessen the effects of small target size on acquisition difficulty, we introduce four enhanced area cursors, two of which rely on magnification and two of which use goal crossing. In a study with motor-impaired and able-bodied users, we compared the new designs to the point and Bubble cursors, the latter of which had not been evaluated for users with motor impairments. Two enhanced area cursors, the Visual-Motor-Magnifier and Click-and-Cross, were the most successful new designs for users with motor impairments, reducing selection time for small targets by 19%, corrective submovements by 45%, and error rate by up to 82% compared to the point cursor. Although the Bubble cursor also improved performance, participants with motor impairments unanimously preferred the enhanced area cursors.

The Codex is a dual-screen tablet computer, about the size of a 4"x 6 day planner, with a self-supporting binding and embedded sensors. The device can be oriented in a variety of postures to support different nuances of individual work, ambient display, or collaboration with another user. In the context of a pen-operated note taking application, we demonstrate interaction techniques that support a fluid division of labor for tasks and information across the two displays while minimizing disruption to the primary experience of authoring notes.

Existing e-book readers do not do a good job supporting many reading tasks that people perform, as ethnographers report that when reading, people frequently read from multiple display surfaces. In this paper we present our design of a dual display e-book reader and explore how it can be used to interact with electronic documents. Our design supports embodied interactions like folding, flipping, and fanning for local/lightweight navigation. We also show how mechanisms like Space Filling Thumbnails can use the increased display space to aid global navigation. Lastly, the detachable faces in our design can facilitate inter-document operations and flexible layout of documents in the workspace. Semi-directed interviews with seven users found that dual-displays have the potential to improve the reading experience by supporting several local navigation tasks better than a single display device. Users also identified many reading tasks for which the device would be valuable. Users did not find the embodied interface particularly useful when reading in our controlled lab setting, however.

We present an empirical analysis of crossing-based dialog boxes. First, we study the spatial constraints required for efficient crossing-based interactions in the case of a simple multi-parameter dialog box. Through a series of 3 tasks, we establish the minimal value of the landing margin, the takeoff margin, and the column width. We also offer an estimation of the role of stroke shape on user performance. After studying the reasons for errors during our experiment, we propose a relaxed crossing semantic that combines aspects of pointing and crossing-based interfaces.

We present ExperiScope, an analytical tool to help designers and experimenters explore the results of quantitative evaluations of interaction techniques. ExperiScope combines a new visualization incorporating aspects of the KLM and the three-state model with an interface helping users to rapidly cluster similar patterns of interactions. The tool makes it easy to identify and compare key patterns of use encountered during data collection. This promotes a deeper understanding of the results of a given evaluation.We illustrate the advantages of this tool by revisiting the data collected for an experiment conducted by Hinckley et al. [19] which compared different mode switching techniques. Our results show that our tool complements the previously reported results by offering insights about error behavior and the impact of mode switching on user performance.By providing a more fine-grained analysis of the data gathered during empirical evaluations, we hope that our tool will improve researchers' understanding of existing and newly developed interaction techniques.