Twitter has been a bit of a mystery to the outsiders since Elon Musk took over. The business plans have changed, the algorithms have changed and the mechanisms have changed. However, Elon Musk has promised, since the source code leak on GitHub that he would share more outlook on Twitter’s algorithm code. And now that has happened.
The recommendation algorithm code of Twitter was leaded onto GitHub on March 31st for everyone to see, offering new insights into what you’re seeing and why you see it, which would be particularly helpful to marketers.
As explained by Twitter: “On GitHub, you’ll find two new repositories (main repo, ml repo) containing the source code for many parts of Twitter, including our recommendations algorithm, which controls the Tweets you see on the For You timeline. For this release, we aimed for the highest possible degree of transparency, while excluding any code that would compromise user safety and privacy or the ability to protect our platform from bad actors, including undermining our efforts at combating child sexual exploitation and manipulation.”
In a tweet from Peter Yang, he lays out the hierarchy to getting boosted, where likes gets you 30x, retweets gets you 20x, and negatives include URL only posts, blocks and unfollows.