This is a pretty big development!
- Twitter releases its recommendation algorithm code on GitHub and explains its decision in a blog post.
- The algorithm has three main stages: gathering tweets, ranking them using machine learning, and filtering out unwanted content.
- Twitter aims for a balance of 50% in-network and 50% out-of-network tweets on the For You timeline.
- Elon Musk polled his followers in March 2022 about open sourcing Twitter's algorithm, with 83% responding "yes."
- Musk hopes the algorithm will be transparent and robust, allowing third parties to understand what is shown to users.
- Code transparency and open source are different, and Twitter will need to establish governance systems for managing community contributions.
- Twitter invites the community to submit issues and pull requests for algorithm improvements on GitHub.
- The decision to increase transparency comes amid criticism about Twitter's moderation and recommendation practices.
- Twitter faces potential competition from open-source projects like Mastodon and Bluesky.