Large language models (LLMs) offer novel opportunities to support health behavior change, yet existing work has narrowly focused on text-only interactions. Building on decades of HCI research on effective behavior change interactions, we present Bloom, an application for physical activity promotion that integrates an LLM-based health coaching chatbot with existing design strategies and UI elements. As part of Bloom’s development, we conducted a redteaming evaluation and contribute a safety benchmark dataset. In a four-week randomized field study (N=54) comparing Bloom to a no-LLM control, we observed important shifts in psychological outcomes: participants in the LLM condition reported stronger beliefs that activity was beneficial, greater enjoyment, and more self-compassion. Both conditions significantly increased physical activity levels, doubling the proportion of participants meeting recommended weekly guidelines, though descriptively, we observed no advantage for the LLM condition in short-term physical activity levels. Instead, our findings suggest that LLMs may be more effective at shifting mindsets that precede longer-term behavior change.

Daily life presents several barriers for people with visual impairments.Modern computers and smartphones alleviate some of these barriers by providing assistive software that can magnify content, adjust visuals, and read text aloud.However, such software is limited to digital contexts; when it interacts with the physical world, it is often too general to support tasks such as reading a physical book, document, or sign.We hypothesize that modern spatial computing devices enable a new generation of assistive technologies.We introduce Spatial Continuity as a novel method for making physical reality accessible to users with visual impairments.Spatial Continuity allows visually impaired readers to engage with physical experiences by transferring physical objects into a virtual environment.We present a prototype in which a smartphone, functioning as a handheld magnifying glass, streams its camera feed to a head-worn augmented reality device.We evaluated the impact of Spatial Continuity in an informal user study, focusing on its usability and potential as an accessibility enabler.

Due to the ever-growing demand for computational resources, the environmental impact of data centers is continuously increasing. Recently, a great effort has been made to mitigate this impact, while the demand for computational resources has continued to grow. Current mitigation strategies focus on the infrastructure perspective, while the application perspective has been neglected. This paper aims to engage application designers and developers on the path to greener application design. Following the Sustainable Application Design Process (SADP) methodology, we introduce a Sustainable Application Design Architecture (SADA) for enriching cloud-native application components with sustainability features that can be exploited to adapt the application workflow to the environmental context. The architecture enables synergies from design to deployment between all stakeholders involved in the application management. The paper focuses on enriching the application with sustainability features in the design and development phases. We also present and discuss a prototype that can translate design-level sustainability features into development features.

The COVID-19 pandemic has challenged our day-today life, businesses, and educational institutions by changing the way we interact with each other in a very short space of time. In the context of teaching, lecturers had to rapidly develop concepts and teaching materials that enable distributed virtual and safe learning experiences. In this paper, we describe the challenges of reorganizing a two-week programming course that we have taught for more than ten years in a classroom setting using face-to-face communication. The course teaches students the basics of developing a software system using the Swift programming language and is a prerequisite for a single semester capstone course. We show how we reworked the course into a distributed format using online sessions and real-time feedback for about 80 students. We describe the remote supervision approaches we used to support students and the process of rethinking the course infrastructure by enabling remote access as well as by offering a semi-automated merge management and code review system. Based on this experience, we provide instructors with insights on how to set up and conduct a distributed software engineering course when face-to-face teaching is impossible.