[58x14]

Incredible. My favorite line is the last: "they're both tools that do nothing more than show me the results of my own decisions."

I've unsuccessfully attempted various implementations of life logging for years and I wonder, how much of those failures were due not to poor system design, but rather a poor relationship with myself throughout time? I have some theories to test starting in a few days.

Some questions I have for OP and anyone else with similar habits: Do you journal regularly alongside / in addition to your activity logging? How are you recording emotional variables like stress? Any plans to expand your data capture to encompass more signal?

[artnc]

OP here. Great question! I've never journaled or recorded emotional variables partly because that takes much more time and partly because they're harder to analyze.

For example, I could record stress on a 1-10 scale. Do I just put 11 for a day of unprecedented stress? Or should 10 always represent the global maximum thus far? More generally, how do I ensure consistency with myself?

I do vividly remember the stressful times in my life and can see their effects in the blog post's charts. I figured it's probably best to preserve at least some of my own privacy, although I did touch on this a bit when discussing music.

All that said, I have friends who journal and I might end up trying it too out of fear of losing data (even if it's squishy). Maybe once a week would be a sustainable cadence?

[58x14]

I've been attempting to solve this as well. I think stress is particularly difficult to measure as a simple number, as there are multiple forms of stress. For me, I've noticed urgency (stress on time), difficulty (stress on cognition), uncertainty (stress on confidence), and guilt (stress on emotions) make up my 'stress matrix'. I'm often able to identify the key strain during stream of consciousness, but only once I've made the conscious effort to backtrace.

I'm also a musician, and ironically, I will often increase my total stress levels by procrastinating certain tasks, using an instrument as my weapon of distraction. Sometimes this has instigated negative feedback loops, causing me to avoid playing for an extended period.

So my plan is to allocate a few minutes in the beginning and at the end of my day where I will reflect on the previous and upcoming cycle, including tasks and goals. Here I will have a -1, 0, 1 flag where -1 is "I am stressed," 0 is "I am relaxed," and 1 is "I am excited." I can either set these flags for an entire period (hour, day, week) or I can set them per object (task, goal). The default state is 0, as I'm more likely to set a flag if I'm experiencing emotion in either direction. Importantly, I will also set these flags retrospectively; so in the morning, I might flag a task as -1, but in the evening, it might be +1.

There are some useful data points here that I should be able to achieve with minimal friction. 1) how often am I stressed or excited? 2) how often can I easily identify the source(s) of emotion? 3) how consistent are my emotional flags over periods of time?

I also wonder what quantifiable data must exist in any form of extended journaling that can be unlocked via NLP or simpler heuristics. When I'm excited, do I tend to use more or less punctuation? When I'm stressed, does it take me longer to complete a journal entry?

Thanks for the inspiration to jot these thoughts down. They've been lingering in my head for awhile.

[artnc]

Interesting approach! It'd be great to hear how it ends up working out for you.

That's also a great point about NLP etc. I was kinda hoping that my "Average daily log entries" would show some kind of trend like that, but I guess it was too simplistic. It really might be worth recording journal data and just assuming that the tools for analysis will catch up later.

[iamacyborg]

There’s also physiological stress, which some smartwatches do record.

[pkrotich]

What inspired you to start? You’ve always been a data/metrics person?

[artnc]

You could say that. More concretely, it was a stressful first semester of college and I felt like adding more structure to my life would help.

[andreilys]

I built a real-time personal dashboard for my life which primarily relies on lazy tracking (requiring little to no manual data entry from me) - https://github.com/Andreilys/personal_dashboard

I do journal as well, using vimwiki (Where I implemented some simple tagging for things like happiness/productivity/stress/etc. ratings from 1-10).

However one of the side projects I'd like to work on at some point is a mood recognition project using a camera that I can put beside my door. The idea being to take a "photo" that then uses an ML algo to determine my mood (from 1-10) and record it whenever I enter/leave my home.