Liberate the airwaves
Problem: Most of the world's video content is hosted in YouTube. You don't have a google account, but still wish to follow your favourite channels' videos.
Solution: Use self-hosted software to register your channels, then use YouTube's data API to fetch them. No Google account required!
This project aims to supply that niche with a solution that can work for them. This project enables you to take back ownership of your data while still being able to watch the great content on YouTube.
Create an API key for consuming API data, then register it in the app. If worried about privacy, you can create a throwaway google account for the purpose of consuming from the YouTube API.
In a nutshell, this project includes:
- A Server which reads from the YouTube API
- A Client where user sets an API token
- The server fetches the registered user's subscriptions
- A PWA client which can be installed on a user's phone
See also the companion browser extension.
I need to create an account to get a YouTube API key? But how is that different from having a throwaway YouTube account?
It isn't. Creating an account for the purpose of consuming from the YouTube API is the same idea as having a throwaway google account that you have your subscriptions in. If that seems like a more appealing option for you, don't let me stop you. However, consider the following points:
There is a difference in terms of tracking. This project proxies all assets through the included web server, which makes it impossible for anyone to include any tracking cookies with the assets. This also means you're not running any proprietary javascript when watching YouTube, if that's something you care about.
Note: Thumbnails are currently not proxied
When using this service, it means you're not logged in to google in order to view your subscriptions. This means it's much harder for google to track your movements around the web using share buttons and google analytics and build a profile on you.
Another cool aspect of the proxy server is that it is impossible for YouTube to know how many users are on the other side, since we are using an API token and not an idividual login. It could just as easily be many users that are using this API token as far as google is concernec, which again makes it very hard to build a profile on you.
LibreTube works great if you're in a situation where you are encountering censorship. In the case YouTube is blocked, it is easy to spin up a VPS and put LibreTube on it and watch videos from there instead. That way you are able to use the application as a proxy, going around any blocks that have been placed on the YouTube domain or IP adresses.
Security is something I take very seriously.
- The app uses a strong and modern algorithm to hash passwords.
- The app sets security headers via the helmet security library.
- The app includes CSP by default. The implementation is somewhat hamstrung by Next.js' lacking support though.
If you find any security sensitive flaws, feel free to contact me by email directly. You can find it at henriksen.is.
The app is very modern, using the latest in buzzword technology. The frontend is a serverside rendered progressive webapp. The backend is a nodejs fullstack javascript expressjs powered next.js rendered react app with an Apollo GraphQL API wich is rehydrated on the client.
Enough buzzwords? Okay. Check this out:
INSERT GIF OF INSTALLING THE APP HERE.
That's PWA being installed on a phone. This way it should be able to replace your YouTube app if your OS supports PWAs.
This project will never support any browsers other than the newest version of the major browsers (excluding IE). It probably will work in earlier versions up to a point, but your milage may vary.
The easy way is to use docker-compose to automatically set up and configure docker containers for you. If you set up an "ubuntu docker" one-click app on digitalocean, for example, you can follow the following steps:
First, ensure all the required software is installed on your server:
Then we check the code out from version control.
git clone https://github.com/LibreTubeApp/LibreTube.git
cd libretube
Almost there - now we install the project's dependant libaries.
yarn
Then we ensure all project files are compiled into their runtime forms.
yarn run build
Finally we run the server, which will run it on facing on port 80. Edit
docker-compose.prod.yml if you need to change any of this.
docker-compose -f docker-compose.yml -f docker-compose.prod.yml up -d
When updating, simply pull and update the project.
git pull
yarn
yarn run build
docker-compose -f docker-compose.yml -f docker-compose.prod.yml restart web
TODO: Database migrations. Until this is implemented you may have to wipe your db when upgrading.
To check for updates without updating, do a git fetch and git status. If it
says that you are behind remote origin, then you have updates available to pull.
The GraphQL server assumes a PGSQL database that is setup with a database and
user that the server can access. Configure it using the environment variables
DBHOST, DBUSER, DBDATABASE and DBPASSWORD.
If you want to help out, see the contribution guide. Any and all help is appreciated!
I ONLY CONDONE THE USE OF THIS SOFTWARE FOR PERSONAL USE.
This is mostly because the big G probably would take offense to anyone using a proxy like this to make a competing product. This is entirely intended as an alternative viewing platform for an individual with specific needs, not as a competing product in any way.