73
The decentralized web is growing
(jlai.lu)
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 18 Jul 2023
73 points (97.4% liked)
Memes
45189 readers
1566 users here now
Rules:
- Be civil and nice.
- Try not to excessively repost, as a rule of thumb, wait at least 2 months to do it if you have to.
founded 5 years ago
MODERATORS
YouTube might be the biggest challenge yet given the extraordinary amount of storage needed to recreate it.
Its also getting the content creators onto the new platform. Thats a bigger challenge I think, without creators it's a dead site really, and making videos is significantly more difficult than image or text posting.
For storage, if we assume the format would be WebM at 1080p, 60fps and 20 minutes in length, it turns out to about 1GB. Even a cheap VPS instance usually offer 50GB of storage (with not too expensive storage upgrades).
So if its distributed evenly, we can host a good bit of videos (nothing compared to YouTube though).
Let's not forget that there's money to be earned by being a youtube person. Creating a model that would make this possible in a federated approach would be bonkers as hell and probably just invite predatory dipshits who then lure creators with seemingly good offers and then start to hold them hostage in ways YouTube hasn't dared so far.
good. i don't want capitalist advertising bs on the internet anyway.
Most professional YouTubers survive primarily off of Patreon support and sponsored videos. YouTube ads provide only a small fraction of what they earn. If they could increase their Patreon or sponsorship income by cross-posting to PeerTube, then they could be enticed to do so. The current issue there is that sponsors are going to want accurate analytics, and PeerTube isn't going to be able to offer the kind of depth of audience analysis that YouTube can.
The problem is, the cost of hosting videos -- both in terms of storage and in terms of bandwidth -- is kind of prohibitive. That part needs to be solved.
Not to mention the computing to re-encode stuff for different framerates and resolution
Someone needs to invent middle-out compression and install it on a network of smart fridges
Couldn't get past the third season of that show. Got too repetitive. Is it worth finishing?