Processes of disagreement are important to public deliberation, but research has not examined the dynamics of disagreement in deliberation of political topics with respect to effects of the channel of interaction. This study analyzes the discussions generated via an experiment in which discussants were randomly assigned either to deliberate online via synchronous chat or face-to-face. The study compares the initiation of disagreement, its qualities, and how long it is sustained in the two environments. Discourse analysis suggests that in the online environment initial expressions of disagreement were less frequent, less bold, and were not sustained as compared with the face-to-face discussions. Reasons include the lack of coherence in synchronous chat, which may challenge interlocutors and prevent them from pursuing a disagreement over multiple turns. Implications of these findings for scholars and practitioners are discussed.

AbstractThe YouTube algorithm is a combination of programmed directives from engineers along with learned behaviors that have evolved through the opaque process of machine learning which makes the algorithm’s directives and programming hard to understand. Independent tests to replicate the algorithm have shown that the algorithm has a strong bias towards right-leaning politics videos, including those racist views expressed by the alt-right community. While the algorithm seems to be pushing users towards the alt-right video content merely in an attempt to keep users in a cycle of video watching, the end result makes YouTube a powerful recruiting tool for Neo-Nazis and the alt-right. The filter bubble effect that this creates pushes users into a loop that reinforces radicalism instead of level-headed factual resources.

No abstract

No abstract