Methodology. I placed a stopwatch on the desk every time I worked, and paused it when I was conciously making the decision to think about something else. When I suddenly caught my mind drifting, I lapped the stopwatch (meaning these were not factored into uptime, but my intuition is that their contribution is miniscule anyways).
I staged each measurement in my notes app and "batch" logged them around every week or so.
My sample size was ~50 measurements. Not every single work session was logged. I.e., if I knew I would be continuously interrupted like when working with friends, I would not log the session.
Observations
- The action of logging did not improve my uptime over the past 2 months.
- Uptime was mostly uncorrelated with every variable I tracked.
- The most significant correlation I found was with total amount of time sitting down, which was roughly negative.
- Surprisingly, there was no correlation with the time of day I sat down to work. I strongly expected higher productivity in the morning.
- Uptime was mostly normally distributed, centered around ~55% uptime.
Now that I have a baseline, some ablation tests I would like to run involve meditation, medication, and willpower... or something...