As students from the University of Washington, we want the best for our campus. However, we’ve noticed that a lot of the information about campus events, organizations and other services are all over the place and difficult to find in one centralized location. Also, it’s incredibly difficult to find things that match up to your interests if you don’t already have friends or connections within existing organizations. That led us to wonder, "how can we aggregate and visualize information about events and services on campus so that students can make friends with common interests and learn more about the resources the UW has to offer?" We conducted many interviews and user surveys and ultimately decided to create a responsive web application that targets three main audiences: first, incoming and prospective students; secondly, students with access needs; and lastly, registered student organizations and campus services that want to advertise and become more well-known.

So far, we've conducted several user interviews to access the viability of our project, and we're currently finishing up user testing on our wireframes and high-fidelity prototypes. Results from our testing as well as a beta-stage application will be shared.

Whiteboards fulfill a variety of purposes within a software development organization. My research project is how the materiality and artifacts of a whiteboard wall mediate work of a self-organizing team. It is trying to understand relationships between the whiteboard wall and cognitive fabrics of the team. The research project has collected a sequence of images since October 2012 of a large whiteboard wall used in a local software development organization to mediate the interactions among the software developers, and interactions with the rest of the organization. Our data shows a much wider variety of usage patterns than just design discussions. This research project focuses on analyzing the role that the whiteboard wall plays in distributed cognition. It involves analyzing the images to identify different types of information that have appeared on different sections of this whiteboard wall, and using the interviews of people from that organization to understand the particular lifecycles and usage patterns of these sections of information.

The twenty-first century has seen English establish itself as the dominant language of international business and global communication. Though English is arguably the most useful language to know, a shortage of qualified teachers has caused the education of English in public curricula to be lacking in some developing countries. This reality raises the question: ‘How can we improve the quality of English education in developing countries?’ The goal of this project is to develop conversational robotic agents that can carry out foreign language lessons autonomously or with partial human supervision, for individuals or small groups. Although such robots could be used for teaching any language to any age group anywhere in the world, this project focuses on teaching English to children in countries where access to English-language education is limited, as part of their primary education (K-6). We are motivated by recent results demonstrating the effectiveness of robotic instruction in comparison to other electronic media and the recent increase in the availability of reliable, low-cost robotic platforms. Our project will explore several frameworks that allow English teachers around the globe to remotely program, generate content for, and control an instructional robot to deliver language lessons. We will leverage our relationship with a non-profit Moroccan foundation that builds schools across Africa to deploy our system and test different frameworks developed as part of this project. Our project will impact children in these schools by facilitating English learning supervised by qualified teachers, and consequently giving them access to the world’s largest information resources for self-education. In the current state, we are designing a course in which students have meaningful interactions with a tabletop robot. A pre-test/post-test will be designed to measure the effectiveness of robotic instruction.

Robots have high potential to serve as information gatherers. The DUB-E autonomous robot being developed at the UW Paul Allen Center for Computer Science and Engineering is one such example. It will ultimately answer human inquiries about physical locations, such as "Is Mike Chung in his office?" by navigating to a specific location and gathering data to answer the question at hand. To expand the potential of robots like DUB-E to answer questions regarding locations beyond a single floor, it is important that they have multi-floor navigation capabilities via an elevator. We therefore present a generic elevator controller implemented within the Robot Operating System (ROS), a popular framework for building complex robotic systems. We developed an implementation that uses a hardware-independent client-server model within ROS to process a robot's requests for the elevator and provide generic state feedback about the elevator's current floor, motion state, and door state. We also created a separate message definition to communicate non-generic information about a variety of elevators depending on what information can be extracted from them (e.g. floors requested, absolute position, etc.) and what is needed for other robots. The server within ROS acts as a secure robotic interface for the physical, hardware-dependent elevator controller. With this system in place, the DUB-E robot will be able to navigate across multiple floors with minimal human intervention, gather additional data, and ultimately answer more questions. In addition, this implementation provides an encapsulated elevator interface that can be ported to existing and future ROS-based robots despite differences in form factor and purpose. Our contribution is expected to significantly reduce development time in future research projects exploiting multi-floor navigation, such as delivery or telepresence.

Sensory objects are representations of sensory phenomena interpreted and used by people and software. Given that digital self-tracking activities have become increasingly shared, we wonder what form sensory objects may take within everyday contexts of communication. How might we use sensory objects to articulate the body in new ways? What are the consequences of these new forms of articulation for understanding and engagement? Using design research, we speculate on these questions: exploring and intermingling how people see through the eyes of a vision algorithm and hear through the ears of sound processing. Within the field of HCI (Human-Computer Interaction), a recent turn toward self-tracking has implicated smartphones in the tracing, codification, and analysis of myriad human activities, from exercise to sleep. With this new attention to sensing, researchers rely on algorithmic reconstructions of sensory experiences that otherwise remain tacit, such as sights and sounds. As people move toward sharing sensory data, it becomes increasingly crucial to acknowledge and question the limitations of sensory reconstruction. Our recent work explored emergent forms of sensory perception with an Android application we developed called SynKu. SynKu processes images and sounds to create prompts for cross-sensory media recording (e.g., a “black and white” audio clip, a “high pitch” image). SynKu interweaves the algorithmic processing of five such media recording activities using ten media attributes (e.g. light / dark, loud / soft) with users’ cross-sensory interpretation of those attributes. People can then share the resulting media assemblage as a short video narrative (or “Ku”). By creating and sharing Ku’s, they produce new sensory objects interpretable by people and software. Following on HCI’s tradition of using artifacts as conversational prompts, we focus on the mobile application’s role in social inquiry: enabling us to investigate how algorithms influence and legitimate the reconstruction of sensory phenomena across self-tracking platforms.

In analyzing the random movement of a particle in two dimensions, we find there is a simple standard model for the probability of the particle being a set distance from the origin, at a given time. This random motion, which is known as reflected Brownian motion, becomes more complex when it interects with set boundries. To support a proposed class of discrete analogues for the boundary local time of reflected Brownian motions, we developed simple computer models using R and MATLAB. These standard random motion models include a constructed algorithm to measure the boundary interactions in discrete local time. Using these models and observing the discrete local time, we see similarities to modeled 2-dimensional Neumann heat equations, as expected from our proposed theorem. This in turn supports the viability for the developed discrete analogues.