Procedural Generation has been used in video games for years, and is where in for example minecraft, the whole gameplay map does not exist at the start of the game, but instead is randomly generated and loaded while the player moves around the world.
I think this is something that could be used for the explore page, or a separate timeline, on Byte. The idea is that a user could start at the top of a timeline, choosing what they like and don’t like, which would influence the content that comes after it.
The way this would work is by using factors such as:
What other users who like the same posts also liked, for example if user 1 likes video a and video b, and then user 2 likes video a, they could also be recommended video b.
The genre/tags of videos that they’ve liked, for example if a user likes 10 videos with #dogs, they can be recommended more #dogs posts.
A degree of randomness? May sound bizarre but platforms like Spotify often recommend songs that are tailored to be slightly out with what you normally listen to, to help you find new genres etc that you enjoy!
Let me know your thoughts on this, especially the last point about having some randomness to help you find new content!