Social Online Video Experiences

Abstract

Online video sites are becoming extremely popular in today's world. Because the act of watching video is online, viewers are able to simultaneously watch video and interact with each other. However, it is unclear whether the passive act of watching a video can be combined with active social interaction; watching a video and interacting with others both require attentional resources, and these activities may be mutually distracting.

The goal of this thesis is to design for social online video experiences. It will demonstrate that the activity of watching together - chatting with others while watching online video - can, and often does, result in a fun, entertaining, and socially engaging experience. It will also demonstrate that watching together can be used as part of a human computation process to learn about video content. For example, chat transcripts from groups of people who watch a video together can be mined to learn a set of tags that describe a video, or a rating for that video. Learning this information directly from the video content is computationally intractable, and collecting it through contributions to an online video community may result in less accuracy because of lurking and free-riding behaviors.

This work will contribute to the field of HCI by furthering our understanding of how the Internet can have a positive impact on social relationships, and how new technologies such as peer-to-peer streaming video can be used both for entertainment and community-building. This work will also contribute to computer science by demonstrating that social processes and interactions can be used to solve computationally intractable problems.

Thesis proposal: Download [pdf]