Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> The issue is that even if video hosting can be done 'cheaply' (and it can't if you want speed!), the ad money won't come to a self-hosted video site.

That's definitely the big barrier to all of this. There's been a few groups trying to build another platform to compete with youtube. So far most of them seem to either have found small niches that are working for them (vimeo et al) or they've got belly up (vid.me, etc.). I'm aware of a ffew that are trying but haven't fully launched yet, such as Linus Media Group's (linus tech tips primarily on youtube) FloatPlane venture, which seems like it's going for more of a youtube-red like model where you pay a subscription fee, and what you watch determines who gets a cut of the fee.

I really do hope that someone manages to get even a modest competitor going for content creators, even if it doesn't smash youtube down it could do a lot to give them a bargaining position on what kinds of content can survive.



>I really do hope that someone manages to get even a modest competitor going

Hey, I did it for cheap.

https://tinydatacenter.com/2018-04-10-ispooge-live.html

It uses a static site generator, CDN for delivery, RasPi's H264 hardware to HLS encode at >1x, and of course Docker. It include scripts to mirror YT playlists and videos.

It's more of a boilerplate site than SaaS/framework/etc. But that's how we do it in Clojure land for the most part anyway. I'm enhancing the RSS and added a Passwordless login and have a history with XMPP.

Features are being driven by taking my own channels off of YT/Twitch as their primary home, and move to the Pi with the POSSE pattern (publish on your own site, syndicate everywhere). There's a video systems compared table + video in the link.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: