Ten Years of the Computer

In 2011, when I was 19 years old, I made a prescient decision. I was an aspiring data visualization designer working at my first startup, and when I discovered then 4-year-old app RescueTime - a sort of self-development introspective spyware, I immediately installed it on all of my computers. (This is ../robots/quantified-self-as-archaeology|tracking as archaeology, not optimization]] - capturing data before knowing what questions I'd ask.)

Rescuetime would quietly watch me for the next 14 or so years, tracking most of my personal computer usage, with a small gap in 2020 when I bought a new computer a briefly forgot to reinstall RescueTime.

What accumulated over that time was an incredibly high-resolution dataset of how much time I spend doing different things on the computer.

I was particularly interested to look at the data surrounding the few years of AI-assisted coding. For me personally, it has felt like a ground-shift in terms of how I use the computer to make things, and luckily, I have the data to support that now.

Getting the data

On RescueTime's settings page you can ask them to "Download your data archive" - which took a few hours to complete for me.

The data that was most useful to me was a 448.9MB rescuetime_data.json file, and I promptly broke it down into .csvs for each year in the dataset, for more intricate analysis. I also converted the data to a parquet file so it can easily be loaded into the browser with DuckDB for on-the-fly analysis and visualization.

22,380 hours tracked over 14 years (2.6 years of straight computer usage)
31,516 different apps and websites tracked
3,827 days of data across 5,115 total days
932 days worth of pure screen time

What the Data Revealed

The YouTube Learning Explosion

My YouTube consumption tells a story about how I learn. I started small in 2012 with just 1 hour, peaked at 315 hours in 2014, then settled into a pattern. But 2021 was different:

2014: 315 hours
2020: 299 hours
2021: 589 hours (97% increase - nearly doubling)
2022: 244 hours (crashed back to earth)

With nowhere to go and endless time, I accidentally conducted the largest educational experiment of my life. While others binged Netflix, I was deep in programming tutorials, design theory, and maker content.

From Software to Hardware: The Music Story

The data captured my transition from digital music production to physical instruments and analog gear:

2014 - Peak Digital Era:

Logic Pro X: 118 hours (building production skills)
Traktor: 197 hours (digital DJing obsession)
Total: 393 hours making music on computers

2016 - The Turning Point:

Logic Pro X peaked at 169 hours
Traktor crashed to 26 hours (the hardware bug was biting)

2017-2020 - The Great Migration: The software hours disappeared as I moved to hardware synthesizers, drum machines, and analog mixers that aren't trackable with RescueTime. What looks like "giving up music" in the data was actually me getting more serious about it - just away from the computer.

If there's one thing the data reveals about my relationship with social platforms, it's that I'm apparently incapable of moderation. I either binge or quit cold turkey.

The Facebook Years (2013-2016) My early twenties were peak Facebook addiction. In 2014, I spent 169 hours on the platform - that's 28 minutes every single day, including a legendary 17-hour binge on June 1st, 2015. (What was I even doing for 17 hours? The mind boggles.)

The Great Migration (2017-2019) Then something interesting happened: I just… stopped. Facebook usage collapsed from 116 hours to 15 hours over two years. This wasn't a conscious decision - I simply found myself opening Twitter instead. The data shows a near-perfect inverse relationship: as Facebook died, Twitter grew.

The Pandemic Peak (2020-2021) COVID broke my brain in measurable ways. Twitter usage exploded from 50 hours in 2019 to 327 hours in 2021 - nearly an hour every day. I was mainlining political discourse, pandemic updates, and tech Twitter drama with 0.6-minute average sessions. Pure dopamine farming.

The Musk Exodus (2022-2025) Then Elon bought Twitter and I noped out entirely. 327 hours to zero overnight. By 2025, my total social media usage across all platforms: 9 hours for the entire year. I went from social media addict to digital hermit in the span of a few years.

Hybrid Design Engineer Evolution

2014: The Creative Supernova

Design peak: 508.5 hours (1.4 hours every single day)

Traktor: 197h (digital DJing mastery)
Logic Pro X: 118h (music production peak)
Adobe After Effects: 24h (motion graphics)
Illustrator CS6: 21h (classic design work)

2014 wasn't the end of my design career - it was the absolute zenith.

The Terminal Renaissance

2021: 42.3 hours (breakthrough year)

2024: 62.9 hours (peak command-line mastery)

But here's the twist: terminal usage peaked the same year as my design renaissance. 2024 saw:

Engineering: 776.5h (career high)
Design: 132.4h (strongest since 2017)

I didn't choose engineering over design. I chose both, simultaneously.

The Hybrid Reality

Recent years (2022-2025):

Engineering: 1,603 hours
Design: 372 hours
Ratio: 4.3:1

Even in my "full engineering" phase, I maintained 93 hours of design work annually. That's nearly 2 hours per week of

creative work that's invisible to the "engineer" narrative.

The Real Story

The data reveals something unprecedented: a professional shape-shifter who refused to pick a lane.

2014: Creative peak (508h design, 136h engineering)
2019: Engineering peak (179h engineering, 5h design)
2024: Hybrid peak (777h engineering, 132h design)

Ten Years of the Computer