My first-ever research paper

As part of my ITEC5204 course final assignment, I had the incredible opportunity to collaborate with my classmate Alicia Morgan, under the invaluable guidance of our instructor Shanel Wu. Together, we produced my first-ever academic-style paper! Receiving the final LaTeX draft from Alicia was incredibly exciting—I never imagined I could create something like this!

The Spark: Why Handwriting Still Matters (And How AI Can Help)

Our project, formally titled "Evaluating User Perceptions and Workflow Preferences in AI-Assisted Handwriting-to-Digital Note Transitions," dove headfirst into a topic I'm personally very passionate about: the evolving role of handwriting in our increasingly digital age. The central question we wrestled with was this: How can we leverage the power of context-aware AI to seamlessly transform our beloved handwritten notes into dynamic, truly useful digital content?

Like many, I have a deep appreciation for the tactile, cognitive experience of putting pen to paper. There’s a certain flow, a connection to thought, that typing doesn’t always replicate. Yet, the efficiency, searchability, and shareability of digital notes are undeniable conveniences in modern workflows. My core belief is that the future isn’t about an either/or choice between analog and digital; it's about intelligently bridging these two worlds. This assignment was our first significant step in exploring a more intuitive, integrated way to combine these input methods. We're envisioning a future of Human-Computer Interaction (HCI) that feels more natural, more seamless – where the information we need appears contextually, almost an extension of our own thoughts.

Our Approach: Building a Bridge, Iteration by Iteration

One of the most fascinating parts of this project was the iterative development of our prototype. We didn’t just design something in a vacuum. We actively sought out participant feedback through interviews, and this input was gold. It led us to evolve our initial pre-made prototype into a live, interactive version. Leveraging the capabilities of advanced AI tools like Claude and Gemini was instrumental here, significantly enriching our research process and outcomes.

The "Magic" of Context-Aware AI in Action

The feedback from our participants was incredibly validating. Many described the experience of our prototype as "almost magical." Here’s what they did:

1. They took handwritten notes in their usual style while watching a 6-minute introductory video about Dynamicland.
2. Immediately afterward, we simply took a photo of their notes.
3. Sent photo to Gemini by leveraging its Optical Character Recognition (OCR) capability to digitize the text.
4. Here’s the exciting part: within just five minutes, their handwritten notes were transformed into a structured website. This wasn't just a flat transcription; the digital version was enriched with context-aware information, automatically linking ideas and providing relevant supplementary data.

Seeing their physical notes almost instantly become an interactive digital asset was a powerful demonstration of what AI can enable.

Key Insights from Our Exploration

This project, while an academic exercise, yielded some compelling takeaways that I believe have real-world implications:

1. The Enduring Value of Handwriting & The Pull of Digital Efficiency: Our study underscored that people still deeply value handwriting for its cognitive benefits. Simultaneously, the efficiency of digital tools is indispensable. This creates a strong, inherent demand for solutions that seamlessly integrate the best of both.
2. Context-Aware AI as a Powerful Bridge: We found that context-aware AI offers a truly effective way to bridge this gap. It's not just about transcribing text; it's about understanding and preserving the inherent structure of handwritten notes and then automatically enriching them with relevant context.
3. Adoption Hinges on User Trust and Control: For any such tool to be truly adopted, it absolutely must align with user expectations. This means prioritizing accuracy in transcription, giving users meaningful control over the process and their data, ensuring seamless workflow integration, and rigorously addressing data privacy concerns.
4. AI as an Augmentation, Not a Replacement: A crucial insight was that AI should be positioned to enhance human agency, not replace it. The goal is to support the user, preserving the intentionality and unique qualities of their handwritten notes while adding digital superpowers.
5. The Path Forward: Transparency, Customization, and Fluidity: Looking ahead, future note-taking solutions in this hybrid space must champion transparency in how AI works, offer robust customization options to suit individual needs, and ensure fluid integration into existing ecosystems to create a truly user-centric experience.

Looking Ahead: The Future is Hybrid (and Maybe AR!)

While I don’t have immediate plans to build this specific prototype into a commercial product, this project has profoundly stoked my eagerness to continue exploring this fascinating space. I’m particularly intrigued by the potential of emerging technologies like AR glasses to create even more intuitive bridges between our physical and digital note-taking worlds. The core vision remains: to seamlessly link our physical and digital experiences.

As a fun side-effect, the thinking and problem-solving involved in this assignment also directly inspired another little project of mine: File Visualizer!

Explore Our Work

I invite you to take a look at the visualized version of our paper, which we created to make our findings more accessible: Explore the Visualized Paper Here

A Note of Thanks

This journey wouldn't have been possible without the invaluable time and insights from our research participants – a heartfelt thank you to each of you! We are also incredibly grateful for the modern tools that streamlined our process, including Limitless AI for helping us record interview transcripts and AI assistants like Gemini for their significant help in analyzing that transcript data.