• minorninth@lemmy.world
    link
    fedilink
    English
    arrow-up
    11
    ·
    1 year ago

    Spending an hour on Reddit or Twitter means downloading a few megabytes of content.

    Spending an hour on YouTube means downloading a few gigabytes of content. The cost to serve that is massive.

    YouTube lost money when Google bought them. It continued to lose money for years. It was only after YouTube finally got large enough and their ad targeting got good enough that they started to turn a profit on YouTube.

    I’m really skeptical that anything other than a big tech company could provide a similar platform like that for free.

    Sure, it could work if you could get people to pay $10/month, like YouTube Premium, but people wouldn’t do that without there being enough content to make it worthwhile. It’s a chicken-and-egg problem. The only way to get past that point is with a massive amount of initial investment.

    • neonpeon@programming.dev
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      I have pinned video to IPFS, before, and it streamed well from an IPFS gateway. I lack an understanding of what the impact to a gateway would be, though, if many users were doing this.

      One thought would be software that would automatically pin any video you viewed, or the partial content you viewed, for a period of time. Popular videos with many views per day would also be supported by many peers to pull from. Videos viewed infrequently might be pinned only by the content creator or a handful of others. The like/upvote concept could be tied into pinning, even on a scale such that an ordinary upvote would subscribe you to, say, 3 days of pinning, and a super-upvote would subscribe you to 20 days.