Skip to content
This repository has been archived by the owner on May 5, 2023. It is now read-only.

Commit

Permalink
feat: improve README
Browse files Browse the repository at this point in the history
  • Loading branch information
paulbricman authored Sep 18, 2021
1 parent 0f12565 commit f4c5e65
Showing 1 changed file with 23 additions and 8 deletions.
31 changes: 23 additions & 8 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,21 +1,36 @@
# Conceptarium
A fluid medium for storing, relating, and surfacing thoughts. [Read more...](https://psionica.org/tools/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.

# Instructions
The conceptarium takes up about **1GB RAM** when running. Clone the repository, and install the requirements using:
[Read more...](https://psionica.org/tools/conceptarium/)

# 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.

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](https://github.com/ljk53/pytorch-rpi/blob/master/torch-1.9.0a0%2Bgitd69c22d-cp39-cp39-linux_aarch64.whl).
**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](https://github.com/ljk53/pytorch-rpi/blob/master/torch-1.9.0a0%2Bgitd69c22d-cp39-cp39-linux_aarch64.whl).

Start the web server using:
**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](https://github.com/Psionica/conceptarium/blob/main/.github/workflows/pytest.yml) 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
```

This will make your conceptarium available at `127.0.0.1:8000`. Go to `127.0.0.1:8000/docs` to get an overview of the available endpoints, and read the [architecture section](https://psionica.org/tools/conceptarium/#architecture) of the write-up.
This will make your conceptarium 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](https://ngrok.com/) 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. 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](https://github.com/autokey/autokey) / [AutoHotKey](https://www.autohotkey.com/) scripts, for setting up [IFTTT](https://ifttt.com/) integrations, and others.

**Note**: If you want to access your conceptarium remotely, have a look at [ngrok](https://ngrok.com/).
**Please find the complete auto-generated API docs over at `127.0.0.1:8000/docs`.**

Simple ways of getting the most out of your conceptarium are by configuring adding it as a search engine via [browser extensions](https://addons.mozilla.org/en-US/firefox/addon/swift-selection-search/), or by creating some [AutoKey](https://github.com/autokey/autokey) / [AutoHotKey](https://www.autohotkey.com/) scripts to interact with it via hotkeys.
![mockuper](https://user-images.githubusercontent.com/20104026/133883441-0faae359-9335-46bf-b10c-27ebb8c274b3.png)
![mockuper(1)](https://user-images.githubusercontent.com/20104026/133883445-b26de5d2-cd71-4a7f-8c0d-99ae25da2865.png)
![mockuper(2)](https://user-images.githubusercontent.com/20104026/133883515-37e7853f-171e-4760-bf03-a2fc557dc364.png)

0 comments on commit f4c5e65

Please sign in to comment.