Building

Flogger

A minimalist and unobtrusive life improvement tool.

I’ve always liked the idea of recording all the little everyday events in my life, and somehow having a digital fortune teller sift through it all and tell me something that I didn’t know about myself. The problem, of course, is that this requires adopting ten new apps, with ten new habits to form. The result? All that work is fragmented into ten different places. How do you make use of it? Sounds like a lot of work… maybe I could write my own app, and add some more useful data types? Maybe run some data analytics, learn a little of the data side?

Like everybody else, I’ve been playing with LLMs for the last little while, and as a software developer, I’ve been looking for an idea where they can be applied in a way that’s not just another chatbot (no disrespect). My favorite thing about LLMs is that they can serve as a natural language interface, so that people can speak naturally and have the machine “understand” it. While VR and AR have stolen the spotlight for new experiences, I think there’s a great opportunity in a much less flashy package: wearables. Specifically, spoken voice.

Flogger is my attempt to connect the two: A “One Life Logger App to Rule Them All,” and wearable/voice UI, made possible by LLMs.

Read the Dev Blog
Product Shot
Frog speaking to a smart watch

Voice and Wearables: First-Class Citizens

Stopping to open your phone, find the right app, and record a data point is a high friction experience.

  • WatchOS App: Easy data logging with simple UI, right on your wrist
  • SiriKit and Shortcuts Integration: Avoid drilling down in menus and hunting for the right parameters. Just use your voice and tell the us what you want to log.
  • Full-featured iOS and Web App: Collect data and view dashboards on your phone or laptop

Know Thyself: Actionable Insights, not Gamification

Custom datatypes combine with Apple Healthkit and Google Health Connect for increased self-understanding

  • Utilize correlation coefficient for early indicators: Immediately draw correlations with the data that the user already has already recorded/collected
  • Encourage frequent data collection to improve data analytics
  • Advanced Data Analytics: Use tools like Scikit-Learn and Pandas to analyze data. The more you record, the better the understanding
Frogs doing self improvement stuff

Built in Public

Why build in public? What does that mean, anyway? I like all the different parts of software development - thinking about system architecture, picking a tech stack, breaking work out into logical chunks, grooming tickets, building a good test harness, drawing up database relationships, you get the point. When I’ve done side projects in the past, I’ve just made the repo public and attempted to write about the project, continually referring back to the codebase. The problem, to me anyway, is that I put too much of my focus on the code. Is it properly structured? Does it look like I know what I’m doing (I do know what I’m doing)? It just feels limiting. I know it’s silly, but it keeps me from quickly moving forward. I end up yak shaving.

For this project, I’m keeping the repo private, but the process public. I’m keeping a public Project Board on GitHub, so you can see what I’m working on. The source of truth will all be in the blog, but I’ll syndicate some of it on my Substack. Maybe even LinkedIn? Side note: I want to like LinkedIn. Maybe some day. I have plans to make some stuff for YouTube - I really like DevAsLife’s style, and I might try to imitate that at some point.

If you’d like to talk about any aspect of software development, probably the best place to hit me up is on Reddit.

Get notified

Receive updates about new content: blog posts, videos, etc.

No spam!