The successor to https://github.com/kovasap/autojournal.
Autojournal is meant to be a personal timeline information management tool that can aggregate data from many sources and then visualize/analyze it. It tries to do this as automatically as possible.
This is useful to aid memory, and to help spot delayed consequences/effects of behavior and environment on mood and future behavior.
See also https://kovasap.github.io/docs/lifelogging/what-and-why/.
Visually, autojournal could be used to generate a widely zoomable life timeline. This timeline could be zoomed out to view from birth to present day, or zoomed in all the way to a daily resolution.
To facilitate this, events in the underlying data would need to be tagged with a "significance" level to dictate what is filtered out when zooming out all the way. Additionally, recurring events should be grouped together; this would give a view like "in 2021, I went climbing on average twice a week".
The fully zoomed out timeline (or the entire timeline) would be a target for sharing, either on social media or in a database somewhere. It would be interesting to view a sampling of how people live their lives this way.
Throughout: wear CGM, make sure ActivityWatch is running on all platforms, and make sure GPS tracking is enabled.
Upon waking up, log perceived sleep quality via mementodb.
Whenever going to a bathroom throughout the day, log pee/poop and freeform emotions text via mementodb.
When eating, take a picture of the food and log what it is via the custom food tracker built into autojournal using mementodb (still experimental).
Every Saturday: Go through location data and create map from coordinates to locations (that are new) + cooresponding activities.
sudo npm install -g @google/claspnpm i shadow-cljssudo apt install joker- Clone this repository and cd into the base directory
- Set up a
claspproject as instructed - If a project already exists, just
clasp loginandclasp clone.
Go to project at https://script.google.com/home/my to run.
See https://lambdaisland.com/blog/2016-10-01-clojurescript-and-google-apps-script for background details.
Compile your ClojureScript and push it to your Apps Script project with this command:
./build.joke
In one terminal, run:
npx shadow-cljs watch autojournal-for-node
In another, run:
node dev-Code.js
export FILENAME=vega.html; gdrive download query "name='$FILENAME' and trashed=false" --force; xdg-open $FILENAME
Use https://github.com/jrmycanady/cronometer-export/pull/2/files to get food data.
Use https://mementodatabase.com/ with a library like "Activity Journal.mlt2" in this repo.
Download https://gpslogger.app/. Make sure in the settings it is set to log to csv format only (creating a zip file for each day), then set it up to write the files automatically to Google Drive.
Manually "export all buckets" in the android app, then use DriveSync to push the json file to Google Drive.
To run every hour, run crontab -e and add this snippet (assuming you cloned
autojournal into your home directory ~/):
0 * * * * ~/autojournal-on-gas/sync-activitywatch.clj
You may need to specify the path to your bb like this:
0 * * * * /home/linuxbrew/.linuxbrew/bin/bb autojournal-on-gas/sync-activitywatch.clj &> awexport.log
- Install activitywatch for windows and have it run continuously
- Download sync-activitywatch.clj from this repo
- Install babashka for windows: https://github.com/babashka/babashka/releases
- Create a scheduled task to run babashka with sync-activitywatch.clj as an argument on a schedule.
- Have your program be
cmd.exeand your arguments/c C:\Users\kovas\babashka-1.3.182-windows-amd64\bb.exe C:\Users\kovas\sync-activitywatch.clj
- Have your program be
- Sync the export file to google drive by adding the dir to the google drive desktop client's watched dirs.
- Send emails with insights (perhaps with biomarker correlator links)
- Use notification emails from services (e.g. GitHub) as a way to get data into the system
- Try tailwind css for report styling.