The China-centric video sharing platform is expanding with a new international app and an English version of its website – with no identity verification (for now at least)
The China-centric video sharing platform is expanding with a new international app and an English version of its website – with no identity verification (for now at least)
I mean. This is… Not worse than now, I guess?
Any entity that can take a chunk out of YouTube’s dominance is a good thing.
But I’m not going to celebrate another corp making a shamelessly monetized video site. They’re independent-ish now, yes, but for how long? It’s just a different flavor of attention gaming and enshittification in the long run.
And I’m skeptical they have the muscle to significantly displace YouTube’s iron grip on everything, even with Google tightening the screws.
Yes! We have peertube which is actually cool.
Unfortunately Peertube is going to be… very difficult to scale.
It needs more people to engage and more exposure.
This won’t help PeerTube become mainstream either.
Hosting videos is extremely expensive, thus there is a financial incentive against hosting a PeerTube instance.
Unless PeerTube fixes the incentives, it won’t be sustainable long-term.
They should do something like Nebula. A subscription maybe in cryptocoins so you still get privacy over your credit card.
It’s more that (as I understand it) the storage requirements increase exponentially relative to the number of users.
how so?
Ah, I was misremembering an aspect - though the requirement is still there that as users join the storage* requirements will increase in the worst case exponentially, it’s not inherent to the ideal system, rather it’s a byproduct of cacheing requirements. At a certain point there will be enough users that popular content being accessed will overwhelm the instance hosting it - this is somewhat fixed by caching popular videos on other instances, but as the network of instances grows the need for cached resources grows with it. There may be ways to get around this, if so I’m simply not familiar enough with the current state of the software to know what those solutions are.
nah i think that the storage requirement grows approximately linear with the number of users, and maybe in the worst case quadratically; but the typical case should be linear.
In an ideal scenario for sure, but in implementation you’re going to run into issues like network throughput and resource access distribution optimization (i.e. feynman parallarity). I got out of the house without a whiteboard or a single grad student (you just never remember the important things) so I can’t confirm more than napkin math, but you may indeed be right that it’s at worst quadratic.