MY ROLES
Research
Interview
Wireframing
Prototyping
UI Design
Development
TITLE
Project Lead
Product Designer
Design Researcher
DATE
2023.04 - 2024.06
OUTCOME
I tested my new web application that gamifies the way users engage with their e-book highlights through keyword-based interactions with generative AI with 10 participants. Their feedback shows Quologue was successful in making their interactions with e-book highlights more fun, reflection-inducing, and personal, compared to their usual e-reading services.
After reading an e-book, what happens?
E-book users create highlights, but they don't really re-read them. Most e-reading applications don't pay attention to this data either. In this project, I explored e-book highlights as potential resources to make e-reading support self-reflection and entertainment.
I designed Quologue, a GPT-integrated web application that allows users to reconnect with their e-book highlights through ongoing dialogue.
MY ROLE
As my Master’s thesis project, I designed the user experience and the interactive prototype of Quologue while building code components for the data pipeline. I developed a design system for Quologue and led a field study with 10 participants over 8 weeks, conducted 29 interviews, and analyzed the data. This project was in collaboration with two software engineers.
DESIGN PROBLEM
Why highlights data go to waste?
Users create highlights on e-books, but they rarely re-read them [1]. These highlights data are valuable resources since they tell interesting stories of their past thoughts and emotions [2].
AI for creating meaningful stories with data
Think of Spotify Wrapped. It uses people’s music listening histories data to create personalized summaries of their year in music, including top artists, songs, and genres. The success of this feature was due to how it created a personally meaningful and evocative experience that emotionally appealed to the users.
For business, this kind of feature would be a valuable add-on, though not for an everyday experience.
DESIGN PROCESS
Autobiographical design
Instead of opting for more traditional UX methods, I chose an autobiographical design approach. The design space for this kind of infrastructure is too new, and it was essential to gather rich and deep personal insights on designing and living with the prototype system and make quick design decisions.
Diary entries on a FigJam board
Wireframes
I used wireframing to ideate and communicate with the team members.
Interactive post e-reading experience through keywords
After months of rigorously experimenting with generative AI and personal e-book data, I arrived at this user flow, involving a set of keywords, a user's response, and a remix.
FINAL DESIGN
Keyword-based interaction
From the early explorations, I discovered that keywords provide a great format to stimulate curiosity and creative engagement with the highlights data.
Revealing the highlight after the user's response
Testing Quologue with power e-book users
We conducted an 8-week user testing to see if other avid e-book readers also have engaging and meaningful experiences on Quologue.
More fun and personal e-reading with generative AI
Participants generally reported that through my keyword-based design, they had more fun, reflection-inducing, and personalized interactions with their e-book highlights on Quologue compared to their current e-reading apps.
REFERENCE
[1] Jane Gruning. 2018. Displaying Invisible Objects: Why People Rarely Re-read E-books. In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems, April 19, 2018. ACM, Montreal QC Canada, 1–12.
[2] Bold, M. R., & Wagstaff, K. L. (2017). Marginalia in the digital age: Are digital reading devices meeting the needs of today's readers?. Library & Information Science Research, 39(1), 16-22.