Skip to content

Data Source: Google Location History

Matt Holt edited this page Dec 4, 2020 · 2 revisions

This data source adds your location history from the Google Maps Timeline to your own timeline.

If enough information is there, Timeliner will associate two coordinates at adjacent times with the activity that got you between them. For example, it will link two locations by BICYCLING, which can add interesting context to your timeline.

Altitude, velocity, and direction metadata -- when available -- is also preserved.

Google Location History
Site https://www.google.com/maps/timeline
Listing Mode ⬜️ Download via API
✅ Import from file
Data Source ID google_location
Authentication None

Obtaining the data file from Google

  1. Go to https://takeout.google.com/
  2. Click "Deselect All" then scroll down to Location History.
  3. Enable Location History (JSON format).
  4. Scroll down and click Next.
  5. The other options are up to you. Click "Create export".

After you download your archive, extract it so the JSON file is on your disk.

Importing location history

If you haven't done so already, add your Location History account:

$ timeliner add-account google_location/you@gmail.com

Make sure the archive is extracted first; Timeliner needs the .json file, not a .zip or .tgz file.

$ timeliner import /path/to/location_history.json google_location/you@gmail.com

Caveats

  • There's a LOT of location history. It is sometimes very detailed, and much of it is redundant. Timeliner skips data points that do not add much new information. For example, several coordinates adjacent in time are often the same or within a few meters of each other. In this case, only one will be imported to save space.

  • For many data points, Google provides the full list of guesses of possible activities at that particular time and place, like WALKING, ON_FOOT, or ROTATE (device picked up or put down). Timeliner evaluates each group of activities and, based on reported confidence levels, keeps only the most likely and useful activities as metadata. For example, it will prefer to keep WALKING as a relationship between two coordinates adjacent in time if that has high confidence, even if ON_FOOT is actually the highest confidence, because WALKING is more specific.


Note: Timeliner does not duplicate the functionality of Google Maps.