cross-posted from: https://lemmy.ml/post/49015971
YouTube is still one of the major points of centralization on the internet, so I’ve been brainstorming ways around the problem.
From the readme:
Torrent-Tube is a set of tools to help decentralize YouTube videos, by moving them to torrents, which can be shared by many people. It includes:
- A Torrent-Tube search site which searches the Torrents-csv search engine to see if the given YouTube video already exists, and is being seeded.
- It does this by extracting the YouTube [VIDEO_ID] from a link, which you can also do manually if you like (IE, the text after
watch?v=...).- A script to download, and create torrent files from YouTube videos, with a uniform naming style and format, taken from TheFrenchGhosty’s YouTube-DL-Scripts.
- You will need to upload these torrent files yourself to a service (details below), and seed them.
In the future, it may be possible to create a browser plugin that checks a video link that you’re currently watching for existing torrents.
Create torrent script
Requirements
Instructions
Copy a YouTube video URL.
# Clone this repo git clone https://github.com/dessalines/torrent-tube # Run the script ./create_torrent.sh [YOUTUBE_URL]The video will download, and is saved in the
videosfolder. The torrent file is saved in thetorrentsfolder.Add the torrent to your torrent app, such as qbittorrent.
I really like this idea, and I suppose it could also be built to include videos from other video hosting cites as well in the future.
Do you think this could become popular enough it to be useful? Public BitTorrent trackers for other media are barely seeded as it is and they have been active for decades. Why would people want to seed media they can watch for free on youtube? (to be clear I understand why but it’s difficult to get enough others to understand something like this.)
Another issue is if it does get popular enough to become useful I could imagine this being a major target from YouTube, or do I misunderstand the copyright laws? Actually considering the db is just one massive torrent.csv file I guess you can’t really sensor that. Although I’m not sure if a copyright troll issue could arise.
Also that 38 MiB csv file is insane.