Quantified self: useful or useless?

… critical design whose goal is not to provide clear answers, but to provoke reflection …

This week’s paper proposes a fictional, exaggerated pervasive technological design that: “encourages individuals to address the larger goal of reducing obesity in society by promoting individual healthy behaviors” [1]. A fictional design (what I found to be a massive plot-twist in this paper) is used therein as a constructed mechanism to provoke critical discourse and reflection within the HCI community surrounding the social, conceptual and ethical limits of pervasive computing, including: (1) How persuasion can shade into coercion; (2) How persuasive computing reinforces cultural trends to control and rationalize human behavior, and (3) How a modernistic approach to quantified self prevents mindfulness and reflection.

The authors argue that it is unethical to change an individual’s attitude, belief or behavior, and that system designers viewpoints influence the behaviors being modified in pervasive computing. Instead, the authors propose that systems should be “simpler” by eliminating system complexity (options) that often aims to enforce sublimated (by designers) social goals. I disagree with this as I do not consider it unethical to elicit behavioral change from end-users with their consent. It is often the users who want help to change, and see the pervasive system as a means of achieving this.

The authors also argue that common practice in persuasive technology is the emphasis on modifying behaviors, rather than attitudes. To achieve this, they suggest for system designers to provide tools for reflection that allow users to understand how they feel regarding the issues and to understand their own subjective “meanings and values” [1], rather than relying on “an illusion of an impeachable, scientifically objective source of measurement”  as a system. This encourages mindfulness, provides control to end-users and ensures that they think about their attitudes, rather than behaviors.

I am critical of the practical benefits for design of both critical and fictional design as the arguments put forward critique a none-existing design experience. If the purpose of critical design is to understand its possible consequences, can a design experience or its consequences be articulated without measuring or observing user-interactions?

I have chosen a blog post by Stephen Wolfram, CEO, and founder of Wolfram Research. He analyzes his personal data (from emails to letters typed), and discusses the outcomes this knowledge can bring. This real-world example of quantified self demonstrates that individuals have been collecting data and quantifying their behaviors for as long as systems has existed, but again, does the data provide useful information?

[1] Purpura, S., Schwanda, V., Williams, K., Stubler, W., and Sengers, P. 2011. Fit4life: the design of a persuasive technology promoting healthy behavior and ideal weight. Proc CHI ’11. ACM, 423-432.