26 February 2025 in i-com
In crisis situations, citizens’ situational awareness is paramount for effective response. While warning apps offer location-based alerts, their usage is relatively low. We propose a personalised messaging app channel as an alternative, presenting a warning bot that may lower adoption barriers. We employ the design science research process to define user requirements and iteratively evaluate and improve the bot’s usability and usefulness. The results showcase high usability, with over 40 % expressing an interest in utilising such a warning channel, stressing as reasons the added value of proactive warnings for personalised locations while not requiring a separate app. The derived requirements and design solutions, such as graphically enhanced user interface elements as guardrails for effective and error-free communication, demonstrate that a suitable warning chatbot does not necessarily require complex language processing capabilities. Additionally, our findings facilitate further research on accessibility via conversational design in the realm of crisis warnings.
3 March 2025 in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
As smart home technology becomes integral to modern living, researchers must consider safety aspects. While single-purpose devices alert users to specific dangers, integrating them within comprehensive smart home warning systems (SHWSs) offers new safety potentials by allowing actuators to respond to threats based on predefined protocols. Key questions include whether user preferences for automation levels in smart homes are affected by different warning scenarios, and how unwanted automation or false positives influence acceptance. To explore this, we conduct two studies: (1) A lab study in a smart home with various actuators, where participants (N = 48) encounter warnings across three automation levels. (2) A follow-up interview study (N = 16) further evaluating our prototype and unwanted automation situations. Results show that participants preferred higher automation during warnings and were more receptive to smart technology in dangerous situations, though customization remains essential to ensure acceptance. While higher automation levels reduced perceived interruption, some still preferred less intense warnings. Others preferred not receiving warnings of mild dangers, fully relying on automation. Finally, we find that specific safety protocols and handling of false positive alarms must be chosen carefully to avoid mistrust, users feeling a loss of control, and damage through unwanted executions.
23 September 2026 in Proceedings of the ACM on Human-Computer Interaction
During crises, reliable communication is essential for seeking help. Peer-to-peer messengers using disruption-tolerant networking (DTN) can improve crisis communication by enabling data transmission even during Internet outages. However, understanding DTN is complex due to its unintuitive behavior compared to server-based messaging. While DTN research has primarily focused on technical aspects, our work investigates users’ mental models and comprehension of DTN. We use a baseline messenger app prototype, adjust its user interface (UI) by introducing new UI elements that visualize relevant metadata, and evaluate the UI in a mixed-design lab study, where participants ( N = 42) engage in chats with partners at varying proximities. Through a mixed-methods approach, including a novel evaluation methodology for understanding users’ mental models of DTN, we find that the redesigned UI improves users’ comprehension of network topology and reduces uncertainty about message delivery. Although the UI does not increase users’ tolerance of DTN latency, our study offers key insights into designing more transparent and supportive DTN-based messengers. Based on these results, we present an updated UI and highlight the implications for crisis communication.