GaspardP / Paradone

Peer-to-peer assisted media diffusion in the browser with WebRTC and MediaSource

Paradone

Paradone is an Open-Source, peer-to-peer powered overlay network for media diffusion in the browser. Its aim is to reduce bandwidth cost of media file diffusion for the server and ISPs while providing a better service quality to the end user.

• It uses a mesh overlay network to share media directly between the users
• It uses the new WebRTC API to provide full P2P capabilities directly inside the web browser
• It uses the new Media Source Extension API to play video with HTML5
• It's a full JavaScript solution which needs no plugin and is invisible to end users
• It's Open-Source and free to use, share and modify

The project was created for and is laureate of the Boost Your Code 2014 contest.

Get more information about the project on:

• the website
• the wiki
• the mailing list
• the issue tracker

License

Paradone is licensed under the AGPLv3. See the LICENSE file for more information.

Getting Started

Install

You can either download one of the releases or clone the project and build the script from source with git and npm.

# Get the project
git clone https://github.com/Paradone/Paradone.git
cd Paradone
# Install dependencies
npm install
# Build the script
npm run build

Copy the file ./dist/paradone.js in your project directory and add it to your website.

<script src="./some/path/to/paradone.js"></script>
<script>
  paradone.start(options)
</script>

Requirements

User side

The user will need an up-to-date web browser supporting both WebRTC and Media Source Extension. It can either be Firefox, Google Chrome (or Chromium) or Opera.

The video served to the client must be played within a HTML5 Video tag. You will need to check the codec of your files.

Firefox users need to turn the porperty media.mediasource.webm.enabled to true in about:config

Server side

The system needs a signaling server allowing users to initiate the communication between them. WebRTC lets you choose your preferred technology for signaling (websocket, xhr, email...).

For now the project uses Firebase as signaling server.

Developer side

The project is written in JavaScript and uses npm to manage all dependencies.

Different scripts are available with npm run

• build Concatenate and minify the script
• debug Auto-build on save with source-map
• test Run the tests on the source files
• watch Run tests on file change

Usage

Every element of the project is referenced under the paradone namespace. Once the start function called the script will find the video tags and share the media between users through the mesh. You can pass options as arguments to the start function, see the API for more detailed informations.