Mar 2

I thought this week’s readings were very good examples of what good design could look like. I was particularly fascinated by 99DOTS, and this application will definitely serve as an inspiration for my projects. While the solution is extremely simple at the outset, reading about it made it clear the immense thought process that went into designing the product in a way that has resulted in the simplicity that’s visible. I particularly enjoyed reading the design evolution section.
“This caused us to redesign our envelopes from scratch, with over 25 graphic design iterations, more than a dozen focus groups, and field observations of over 100 patients. The result was a major simplification of the envelopes (and supporting algorithms) that allowed patients to dispense pills in any order.” (Cross et al., 2019, p. 5)
The above excerpt has reemphasized to me how good design takes time and meticulous refinement. What stood out for me in both projects is the depth at which both groups engaged with their respective stakeholder groups, learning from them and building together through continuous iterations. Having worked in a conservative rural school setting for several years, it resonated with the insights shared by Sorcar et al. Reading the article reminded me of an instance when I organized a menstrual health awareness program for middle and high school girl students, conducted by trained female facilitators. Some of the teachers in the school were scandalized to learn that the girls were being taught about bodily processes and protested against the program. This was despite the fact that the program was conducted only for female students, as we anticipated backlash for also including male students. This experience aligned well with the insight shared by Sorcar et al. on how it is important to design solutions while accounting for the taboos in communities.
A key takeaway for me from both readings is the indispensable role of designers, especially in the age of AI. I believe it will be good designers with the instincts of researchers who will determine whether a digital artifact is slop or useful. With AI, it has become incredibly easy to develop software applications quickly. While writing software code can be accelerated and automated, I am not sure the same can be applied to understanding human systems and how they interact with a solution over time. This requires open-minded observation and insight.
“99DOTS aims to preserve the benefits of prior approaches while reducing cost and implementation complexity. Its philosophy is to replace a fully automated solution (such as an electronic pillbox) with one that contains a manual human step (placing a phone call). In other words, by ‘undoing’ automation, it might be possible to offer transformative affordability without sacrificing user acceptance.” (Cross et al., 2019, p. 11)
I believe the above quote is well applicable to AI systems as well. Maybe the success of any intervention depends on integrating certain critical human steps rather than blindly relying on technology and automation.
In 99DOTS, the authors made it clear that their solution is only one of the many solution in addressing the overall adherence problem. Still the solution appears to be successfully in areas where its applicable. What are the design aspects of 99DOTS that has made it possible?