The topic of my two previous blog post were papers that we submitted to the upcoming ICT for Sustainability (ICT4S) conference. This blog post is about yet another paper I submitted to that same conference (the third and last). The paper "Quantified Self for Sustainability: Limitations and Possibilities" is written by Björn Hedin, Daniel Pargman and Miriam Börjesson Rivera. We are all at the KTH Royal Institute of Technology but while me and Björn work at the School of Computer Science and Communication, Miriam work at the School of Architecture and the Built Environment. Me and Miriam have on the other hand sat next to each other at the premises of the Center for Sustainable Communication (CESC) for the better part of two years while Björn sits at the Department of Media Technology and Interaction Design (MID) - one floor above from me and Miriam.
The paper primarily builds on the first two authors' work in the project "Improved energy counseling and energy habits by Quantified Self Assisted Advisory". This paper does however not report on the results of any particular project but is rather a synthesis of us thinking about quantified self (QS) for a long time and in may sub-projects. The main contribution of the paper is the presentation of a framework "which can be used to develop and assess QS solutions". We really think we are on to something but it could be that we could sharpen the discussion and the conclusions further (it was a bit hectic in the end).
Quantified Self for Sustainability: Limitations and Possibilities
Quantified self has been described as “any individual engaged in the self-tracking of any kind of biological, physical, behavioral, or environmental information” in order to obtain information and act upon it. Some aspects of our lives such as walking is easy to track, for example by using a step counter. Other aspects such as greenhouse gas emissions (CO2e) generated by different decisions in our everyday lives are much harder to track. Such aspects must instead be derived from other data sources, such as when calculating CO2e emitted when driving a car, or from eating a meal. The transformations that are required are in most cases based on templates, simplifications and assumptions that all introduce uncertainties, making the results deceptively precise. In this paper, we present a framework that can be used to understand and express these uncertainties. The framework highlights limitations of how derived data can be used and what can and cannot be framed as “facts” with a reasonable degree of certainty. However, by explicitly acknowledging and presenting the reasons for the uncertainty, we argue that Quantified Self can - with proper knowledge of its strengths and weaknesses - be a powerful tool for reflective learning and pro-environmental behavior change.
Keywords—Sustainable HCI, Quantified Self, Behavior Change