Presented my paper in the Alt.chi strand at CHI2014 in Toronto last week. CHI is a massive machine and annual ritual, it’s where the heavy theoretical and practical work in human computer interaction is presented with all the big hitters (Bannon, Dourish, Schneiderman, Rogers, Kuutti etc.) in attendance. My presentation was in graphic form and was relatively well attended considering the six parallel sessions going on. I got some interesting questions and an approach by Google who have been thinking about a similar way of eliciting stories from people. I was between two other papers. One proposing a way of generating automatic software tutorials using semantic matching, and the other a detailed eye-tracking study by the BBC of how people watch TV with 2nd and 3rd screens. Both highly quantitative lab-based studies and therefore very different to what I was talking about – the political and moral case for a narrative orientation to digital experiences and interfaces.
Later in the week there was a whole session devoted to ‘Narratives and Storytelling’ which contained some interesting work. Dan Andrews, a PhD candidate at Birmingham University (where Russell Beale is Director of HCI) showed his work on branching comics. His research compared how participants navigated a branching interactive comic against a linear paper one. He tracked where participants clicked and looked, then mapped how the regions compared. Results were shown as heat maps. Analysis of the resulting maps is then intended to inform the design of other interactive and narrative media. The work was limited to a very constrained set of pages but did shed some light on how people ‘read’ interactive visual narratives. This paper introduced me to Neil Cohn‘s work on narrative visual structure, which looks like it might be relevant to my research.
The other paper in this session that appealed to me was by Elizabeth Bonsignore from the University of Maryland (home of Bens Schneiderman and Bederson) who presented results from StoryKit, a smartphone app designed for multimodal storytelling. You can take photos or video, annotate with text and voice or other sound recordings then save and share. It’s a very simple application, they emphasised that the interface is simple but well integrated – definitely not designed to be visually appealing or with graphic elegance in mind. It has been downloaded a massive 800,000 times. This has resulted in a huge dataset the researchers have been mining with Google Analytics to get an understanding of what users have done and what features they value the most. Elizabeth is interested in the implications for ‘in-the-wild’ HCI research. Intriguingly, they claim not to be interested in quantitative insights such as causality or treatment effects from particular designs, but in user behaviour.
Overall my impression was that these two papers and projects were rigorous, qualitative, human-centred research about design but not displaying any particular interest or applicability to design processes.
Given my interest in browser behaviour there was another session I found interesting; Desktop Search and History that I’ll blog about next week.
My presentation is here alt.chi presentation 2014.