\n<\/p>\n
Twitter has released the code that chooses which tweets show up on your timeline to GitHub<\/a> and has put out a blog post explaining the decision<\/a>. It breaks down what the algorithm looks at when determining which tweets to feature in the For You timeline and how it ranks and filters them.<\/p>\n<\/div>\n According to Twitter\u2019s blog post, \u201cthe recommendation pipeline is made up of three main stages.\u201d First, it gathers \u201cthe best Tweets from different recommendation sources,\u201d then it ranks those tweets with \u201ca machine learning model.\u201d It then filters out tweets from people you\u2019ve blocked, tweets you\u2019ve already seen, or tweets that are not safe for work. <\/p>\n<\/div>\n CEO Elon Musk has been promising the move for a while \u2014 on March 24th, 2022, before he owned the site, he polled his followers about whether Twitter\u2019s algorithm should be open-source, and around 83 percent of the responses said \u201cyes.\u201d In February he promised it would happen within a week, before pushing back the deadline<\/a> to March 31st earlier this month.<\/p>\n<\/div>\n Musk has also been preparing his audience to be disappointed<\/a> in the algorithm when they see it (which is, of course, making a big assumption that people will actually be understand the complex code). He\u2019s said<\/a> it\u2019s \u201coverly complex & not fully understood internally,\u201d and that people will \u201cdiscover many silly things,\u201d but has promised to fix issues as they\u2019re discovered. \u201cProviding code transparency will be incredibly embarrassing at first, but it should lead to rapid improvement in recommendation quality,\u201d he tweeted.<\/p>\n<\/div>\n There is a difference between code transparency, where users will be able to see the mechanisms that choose tweets for their timelines, and code being open-source, where the community can actually submit its own code for consideration, and use the algorithm in other projects. While Musk has said<\/em> it\u2019ll be open source, Twitter will have to actually do the work if it wants to earn that label. That involves figuring out systems for governance that decide what pull requests to approve, what user-raised issues deserve attention, and how to stop bad actors from trying to sabotage the code for their own purposes.<\/p>\n<\/div>\n Twitter says people can submit pull requests that may eventually end up in its codebase<\/p>\n<\/div>\n<\/div>\n The company does say it\u2019s working on this \u2014 the readme for the GitHub says \u201cwe invite the community to submit GitHub issues and pull requests for suggestions on improving the recommendation algorithm.\u201d It does, however, go on to say that Twitter\u2019s still in the process of building \u201ctools to manage these suggestions and sync changes to our internal repository.\u201d<\/p>\n<\/div>\n The decision to increase transparency around its recommendations isn\u2019t happening in a bubble. Musk has been openly critical of how Twitter\u2019s previous management handled moderation and recommendation, and orchestrated a barrage of stories<\/a> that he claimed<\/a> would expose the platform\u2019s \u201cfree speech suppression.\u201d (Mostly, it just served to show<\/a> how normal content moderation works.)<\/p>\n<\/div>\n But now that he\u2019s in charge, he\u2019s faced a lot of backlash as well \u2014 from users annoyed about their For You pages shoving his tweets in their faces<\/a>, to his conservative boosters growing increasingly concerned about how little engagement they\u2019re getting. He\u2019s argued<\/a> that negative and hate content are being \u201cmax deboosted\u201d in the site\u2019s new recommendation algorithms, a claim outside analysts<\/a> without access to the code have disputed.<\/p>\n<\/div>\n