Quantified Self

Last modified:
State: draft

Value Proposition

Personal level

There is a lot of discussion in the Quantified Self movement about the actual value of self-tracking/logging and it seems many have grown skeptical about its value over time.

I will probably write about my own thoughts on this someday, but to summarize: I find it fun and rewarding, keeping a history of what I’ve been doing is important to me so that I can look back and have a less biased view of my past. I think there is a lot of value to be extracted from some of the data. I’m interested in several ideas with high chance of failure/stagnation yet potentially high-impact projects. I reach out to people about these to share ideas and get feedback from time to time, if that brought you here: Hello!

Societal level

A lot of data is collected about us daily, but most of it does not end up in our hands.

Perhaps one of the most important data out there is medical data. This topic has received a lot of discussion, for good reason. We want to protect the patients but we also want the data available for analysis. Today there are very strict regulations on the use of medical data in research, effectively preventing large scale exploratory data analysis (“big data”) of patient records.

One writer noted that giving the patient ownership of their data would make it even harder to do large scale data analysis as there would be no information silo one could aggregate from. He instead advocates that patients should have data control, where the user is not the sole owner but still controls his/her copy. This seems to be the direction that medical data is moving in right now, at least in Sweden.

But data silos can be dangerous for both security and privacy reasons (no transparency about which data is kept) I can however imagine a different world: where patients can give away their data (all or just some) on an opt-in/out basis to some regulated entity after it has been filtered through an anonymization system. This would still lead to information silos, but a voluntary and less intrusive one.

My efforts

Automatic

Computer Activity

  • Active window/website
    • RescueTime
    • ActivityWatch
  • Active file in text editor
    • WakaTime
    • ActivityWatch extension possible using existing WakaTime plugins
  • Input activity
    • ActivityWatch (currently only for AFK detection)
    • ulogme (keypresses)
    • Selfspy (full keylogger)

Location

  • Google Maps Timeline / Google Location History
  • Moves

Fitness

  • Fitness trackers
    • Mi Band 2 (cheap, recommended, my current one)
    • Mi Band (even cheaper, much better 30-60 day battery)
    • Fitbit (Flex was my first one, the 5 day battery was annoying after a while)

Sleep

  • Currently tracking with Mi Band 2, acceptable accuracy but not great.

Manual logging

During my first year of university (2013) I did extensive manual logging in a spreadsheet. It worked really well and it was pretty detailed (including mood + productivity 3x daily, sleep, exercise, etc.). I kept it up for months and it’s one of the most important logging experiences I’ve had, the spreadsheet format is great for most manual logging (apart from things like diaries).

Issues

  • Takes effort

Benefits

  • Enables logging of a lot of data otherwise unrecordable.

Subjective measures

  • Mood
  • Productivity
    • Could be worth more than objective measurements in some aspects.
    • Highly correlated with mood in my findings (as expected, but interesting nonetheless).

Drugs/Pharmaceuticals

  • Recreational drugs (alcohol, etc.)
    • A great way to keep track of consumption patterns and tolerance.
    • I believe this greatly improves self-control of consumption (due to the increased awareness of consumption patterns).
    • Definitely recommended.
  • Pharmaceuticals and supplements

Privacy

The best argument for not logging everything is the risk of compromising very personal information. I therefore don’t use things like keyloggers but plan to use keycounters.