CommunityData:Twitch

From CommunityData

We have an opportunity to study Twitch. For now, lets collect ideas and references that might be useful on this page.

Ideas[edit]

Community governance[edit]

How do streamers balance competing interests?

A list of possible interests:

  • community expressiveness (free speech, interacting with the community)
  • entertainment
  • control over the community
  • attracting / pleasing advertisers
  • attracting / pleasing donors
  • compliance with platform requirements
  • attracting and maintaining a community

Streamer success[edit]

Can we predict which channels will take off?

Theories of streamer success:

  • Density dependence
    • Prediction: those among the first to stream a game or activity in the early stages of the game's popularity will succeed
  • Resource partitioning
    • Prediction: those who differentiate from others streaming the same game or activity can also succeed
    • Prediction: those who keep consistent time slots for streaming can succeed
    • Density dependence and resource partitioning looking at viewer overlaps.
  • Cross-site communities
    • Prediction: those who are active in another community will succeed (search gaming sites, forums, Reddit, YouTube, for evidence
    • Prediction: top players in games attract more viewers than average / low rated players.
  • Preferential attachment
    • Prediction: those who get popular quicker get more popular
    • Those who get popular first within a genre continue getting popular.
  • Stage / development model (What is important may vary along developmental trajectories)
    • Normal -> Affiliate -> Partner
    • Affiliate and Partner Twitch programs offer streamers ways to make money not just for themselves but for Twitch as well
  • Mutualism
    • Streamers who enter mutualistic relationships with one another (have their communities spam their faces, advertise for one another, host the other's stream on their channel) will be more successful.
      • Does hosting other streams increase streamer followers / subscribers?

Interaction on Twitch[edit]

Why do people love Twitch? What makes the experience so fun? Why are people so willing to donate? The Chatroom experience seems so important. But how do people experience it? What challenges does the fast paced chat create for moderation? How do overlays encourage interaction in Twitch?

This set of questions are more qualitative or maybe suitable for a laboratory study. This seems to have attracted a fair amount of prior attention.

Social computing papers about Twitch[edit]