conceptarium / README.md
paulbricman's picture
Update README.md
9cd2e99 unverified

mockuper

Conceptarium

A conceptarium (noun. /knsɛptɛriəm/, plural: conceptaria) is a fluid medium for storing, relating, and surfacing thoughts based on a new representation of knowledge. It’s meant to provide a foundation for new tools for thought to build onto, a means to nurture a new tooling ecosystem for knowledge work – a cognitive infrastructure. It embodies a philosophy of knowledge which differs in important ways from the one held by the knowledge graph poster children (e.g. Roam Research, Obsidian, Logseq), and can be deployed today in a self-hosted regime, even on a modest Raspberry Pi.

Read more...

Installation

The conceptarium is a server app which can be deployed on managed hosting (e.g. DigitalOcean) or in a self-hosted regime (e.g. salvaged desktop, Raspberry Pi 4). The app itself takes up about 1GB RAM when running.

Docker

To deploy the conceptarium using Docker, first make sure to have it installed, then simply run the following.

Note: This installation method is not suitable for the Rasberry Pi or other ARM devices, as they need a special version of pytorch. To deploy on those, please follow the second installation option (from source).

docker run -P paulbricman/conceptarium 

Your conceptarium should be available at 127.0.0.1:8000.

From Source

To set up the conceptarium, clone the repository, and install the requirements using:

python3 -m pip install -r requirements.txt

Note: If you plan to deploy your conceptarium on a Raspberry Pi 4, you'll need a version of pytorch compiled for ARM devices, such as pytorch-rpi.

Note: Your system might be missing some prior dependencies. The missing packages show up during installation, and Googling one-liners for installing them is rather straightforward. However, for deploying on Ubuntu, you can also look into the CI/CD tests for details on setup.

Once the requirements have been installed, start the web server using in the root of the cloned repository:

python3 -m uvicorn main:app --reload

Your conceptarium should be available at 127.0.0.1:8000.

Usage

Note: If you want to access your conceptarium remotely, you might want to have a look at ngrok and expose local port 8000 to the world via a URL.

Out of the box, the conceptarium exposes endpoints for all relevant interactions which can be accessed through a browser (see mockups below). However, for power users, a host of endpoints are exposed which return JSONs, plain text, image files, and others, rather than web pages. Those can be used to set up the conceptarium as a browser search engine, for creating AutoKey / AutoHotKey scripts, for setting up IFTTT integrations, and others.

Please find the complete auto-generated API docs over at 127.0.0.1:8000/docs.

The style and color scheme of your conceptarium can be tweaked via assets/style.css.

Further Reading

Screenshots

mockuper(1) mockuper(2)

Demos

Note: Content listed below uses dummy data and an older design.

AutoKey "I'm feeling lucky" text-to-image

https://user-images.githubusercontent.com/20104026/134158253-1ff71d90-2f89-40f0-a074-2dd489bd1683.mp4

Browser search shortcut

https://user-images.githubusercontent.com/20104026/134158264-c8b7ecd8-b11c-4b56-9a42-4f4c3a36573f.mp4

Autokey custom text-to-text dialog

https://user-images.githubusercontent.com/20104026/134158276-cadae2cd-6991-41df-a09d-3e9dbc09ba29.mp4

Google Assistant + IFTTT

eef9e426-06cc-42ff-935b-a74ac64d5991