This pandemic has brought out a lot of misinformation on social media and platforms have been scrambling to help people figure out which is real and which is misleading. Twitter started labeling tweets that have been flagged as false or contentious but there have also been some questions as to how they have been flagged as such. Now they are launching a new community-based platform called Birdwatch that aims to help in the fight misinformation by asking help from the users themselves.
Birdwatch is currently a separate site from Twitter. Users can sign up to report tweets that they have identified as misleading and add notes to give further context. They can also rate the helpfulness of notes from other contributors, which may be kind of similar to the upvote or downvote button on Reddit and other sites. The first phase of this project is intentionally keeping the site separate from Twitter as they are still in the process of determining if users will find the context “helpful and appropriate”.
— Twitter Support (@TwitterSupport) January 25, 2021
They eventually plan to make the notes visible on tweets for everyone once they have managed to reach a consensus from “a broad and diverse set of contributors”. They’re also working on systems that will make all of this transparent. Eventually, all the data in Birdwatch will be publicly available and downloadable as TSV files. The algorithmic codes they’re using will be published publicly as well. Actually, the initial ranking system can already be viewed and the reputation and consensus systems will follow later on.
How we’ll see the notes eventually on the main Twitter app is something that they should work on. Since there will probably be a lot of back-and-forth in contentious tweets, there should be a way to view it that will not cause further confusion. They will also be facing a lot of challenges in building this so hopefully they can work on it now while they’re still basically in the testing phase. Twitter says they have received “broad general support” for Birdwatch from the more than 100 qualitative interviews they have already conducted.
For now, birdwatch.twitter.com is only available in the U.S and only people who are there can test it out. There is no timetable yet as to when this will roll out more widely so we’ll just have to wait for further announcements.