Twitter is expanding a test for its planned upvote/downvote feature, which allows users to vote on whether they think a tweet response is relevant to the thread.

This new round of testing expansions, which initially rolled out in July 2021 and only on iOS, will now also cover web and Android platforms (along with more users overall). According to Twitter’s original statement of intent, the feature is intended for replies on tweets as a way to determine which types of replies to show more or less of. So essentially, it’s all about users providing the data to influence the algorithm that will amplify or hide replies.

The idea was met with a fair amount of uncertainty and resistance when it was first announced, with many users believing that downvotes could be abused. For example, depending on the conversation and the overall audience, votes could be cast to amplify or stifle voices that matched users’ own biases. And while Twitter has stated that it has “…learned a lot about the types of comments you don’t find relevant,” many are still concerned that the feature will be abused. Or that it simply won’t work as intended.

In the recent announcement tweet, Twitter user Wordle Gummidge pointed out that “every time negative voices, veiled or otherwise, have been attempted on platforms, they have only led to a greater silencing of marginalized voices.” He asked, “What makes you think your implementation will be any different?”