A memex (memory index) is a conceptual data storage and retrieval system outlined by Vannevar Bush. With the rise of the Internet, we increasingly use our memory to remember where to find information (i.e. a Google search) rather than remembering the actual information. This project is an attempt to allow users to search and analyze the digital data that Tech companies collect.
Some example memex use cases:
- Trying to remember when you first encountered a topic
- Search trends across time
- Seeing a heatmap of location history
Currently, the project only uses data from parts of Google Takeout. I would like to extend it to provide more insights from Google Takeout, as well as insights on data from other digital services.
Memex does not send your data to off-device locations. Everything is processed and handled on your device.
When downloading your data export as the largest download size (50 GB) as this keeps your data from being divided.
Icons created with Gradientify by Iconshock
This project uses ElectronJS and Bulma.io for the frontend, and Golang and SQLite3 as the backend. The Google Takeout data is parsed using https://github.com/dylan-mitchell/ParseTakeout.
From source:
- Clone the repo
npm installnpm startin project root
Check the Releases for builds. Currently I am unable to build for Mac as I do not own one :( .
There are .deb, .snap, .pacman for Linux and .exe for Windows.
Note: I have only been able to test the .pacman build
