I guess we should start thinking about building a replacement for youtube now... every time this happens it just feels like we cant build fast enough
And I don't even want to rebuild everything from scratch, but its not like there are other options. all the legacy platforms are caught up in rent seeking and its making them unusable
Build the back bones, let entrepreneurs compete on the front end experience so we have multiple good options.
If you build it, they will come!
But the file storage is still the big problem, right? People having to pay for their own storage and bandwidth? Until we solve that, we won't see a big change, right? Just some of my normie non-dev thoughts
It's easy to solve but the podcasters and video people don't want to pay for storage. YouTube has deep pockets and a normal company, developer can't compete against YouTube or Rumble.
Unfortunately YouTube is where the views are.
It's all about the CDN. Storage costs are nothing compared to CDN costs. If you have a 5 minute HD video that gets viewed 100k times end to end, at low-end public cloud CDN costs ($0.05/GB for 30 TB total) that's $1,500. And that's not including the CDN cache-fill.
Storage costs are teeny tiny in comparison.
Who's paying that $1,500? It'll be cheaper if massively scaled on private infra, but still, who's paying?
Are viewers paying the CDN costs? Are creators? If the latter and the video goes viral then what? 1 million views and then a surprise bill over $10k?
I think you can have both
a) massively scaled private infra
b) commerce or advertising business model
The trick being to break the centralisation, not the business model.
Work to do, obvs, but doable.
I hope so. But having worked in this area there's this X factor which is how accustomed people have become to Rolls-Royce CDNs. YouTube is absolutely a Rolls-Royce, as is Twitch and so on. Even Google Drive.
When you present users with a Toyota Corolla CDN they absolutely notice and it quickly goes downhill. And anything built on private infra is going to be somewhat Toyota, no matter how scaled. Even Bluesky uses Bunny, though the bulk of the rest of their infra is bare metal. Tricky one, I think coming at the business model can be done but still will need some reliance on the big CDNs.
That’s why I’m thinking of a different model: instead of free streaming like YouTube, what if viewers paid a small amount of sats (Bitcoin) to watch a video?
For example: it could cost 30 sats to watch a 5-minute HD video. When you press play, it auto-deducts the sats — 15 go to the creator, 15 to the platform. That covers the CDN cost (which is about 23 sats per full view), and still leaves room for a little profit or scaling.
The idea is to keep the costs small for viewers (just a fraction of a cent), but make the whole system sustainable without ads, surveillance, or big VC burn. Real value-for-value.
Zap stream is pretty much already doing this. The biggest problem is most of YouTube is in the fiat mindset so they won't want to pay to watch a video. Unfortunately that's the only way to make money if you don't have big pockets. YouTube isn't even profitable by the way
I think this is for sure the way, with Cahsu. And a separate wallet/mint, just pulls form the background, minute by minute, zero cognitive overhead (just have to fill up the nuts from time to time).
A lot of people do pay on YouTube, YouTube Premium has has 125 million subscribers and will hit 200 million soon. So I think enough people will pay.
If one HD view is $0.001 per minute in CDN costs then $10 gets you 10k minutes, which is 7 full days 24 hours a day. If you double that for some profit to the creator then still pretty reasonable.
The trick is to engineer all that #cashu stuff in the background, payments being made every minute under the radar. Lightning even with 3% fail rate not going to cut it.
If you have an internal lightning wallet then lighting will work. Sites like wavlake, fountain, zap stream already have internal lightning wallets and it works fine.
People will pay but the main reason podcasters are on YouTube is for the views. YouTube still has the biggest reach.
I don't know anything about cashu but I doubt many of the Bitcoiners are going to want to deal with crypto. Most of us are Bitcoin only.
If I can't send lighting to the sites wallet then I find a different site. Bitcoin only.
Network bandwidth has been growing for years, but the cost of reliable storage has been decreasing at a very slow rate. I can't say that storage costs are static, but they don't seem to be beating inflation.
I think key is figuring out monetization options for creators with the right incentives. YouTube is really good at this, no one uses Vimeo or other services why should creators choose nostr over youtube?
Peertube frontend, Nostr backend?
maybe, does peertube use HLS encoding?
Would it make sense to build a video sharing platform instead of a video hosting platform? Like having your own RSS feed or email newsletter, if people are following you via a Nostr video platform and you usually share YouTube videos - then suddenly share a Peertube video - it won't matter since people can still follow you.
There's a similar idea behind https://docs.polycentric.io/protocol/ though it uses Ed25519 so there's probably not an easy way to build a Nostr based video sharing platform combined with it.
