←back to thread

238 points nafnlj | 4 comments | | HN request time: 0s | source
Show context
echelon ◴[] No.46241293[source]
We need a P2P internet.

No more Google. No more websites. A distributed swarm of ephemeral signed posts. Shared, rebroadcasted.

When you find someone like James and you like them, you follow them. Your local algorithm then prioritizes finding new content from them. You bookmark their author signature.

Like RSS but better. Fully distributed.

Your own local interest graph, but also the power of your peers' interest graphs.

Content is ephemeral but can also live forever if any nodes keep rebroadcasting it. Every post has a unique ID, so you can search for it later in the swarm or some persistent index utility.

The Internet should have become fully p2p. That would have been magical. But platforms stole the limelight just as the majority of the rest of the world got online.

If we nerds had but a few more years...

replies(6): >>46241355 #>>46241429 #>>46241474 #>>46241556 #>>46241646 #>>46242453 #
baq ◴[] No.46241556[source]
Centralization is simply more efficient. Redundancy is a cost and network effects make it even worse. You’d have to go the authoritarian route - effectively and/or outright ban Google and build alternatives, like Yandex or Baidu.
replies(1): >>46242089 #
doganugurlu ◴[] No.46242089[source]
For a lot of things we don’t opt for the cheapest solutions that also lack redundancy for a lot of things. Why not for the “information highway”?

Most efficient = cheaper. A lot of times cheaper sacrifices quality, and sometimes safety.

replies(1): >>46242666 #
1. AnthonyMouse ◴[] No.46242666[source]
It's not even that. It's that "centralization is more efficient" is a big fat lie. If you look at the "centralized systems" they're... not actually technologically centralized, they're really just a monopolist that internally implements a distributed system.

How do you think Google or Cloudflare actually work? One big server in San Francisco that runs the whole world, or lots of servers distributed all over?

replies(1): >>46242747 #
2. baq ◴[] No.46242747[source]
I know exactly how they work, but they have a single entry point, as a customer you don't really care that the system is global, and they also have a single control plane, etc. Decisions are efficient if they need to be taken only once. The underlying architecture is irrelevant for the end user.

Why do you think they're a monopoly in the first place? Obviously because they were more efficient than the competition and network effects took care of the rest. Having to make choices is a cost for the consumer - IOW consumers are lazy - so winners have staying power, too. It's a perfect storm for a winner-takes-all centralization since a good centralized service is the most efficient utility-wise ('I know I'm getting what I need') and decision-cost-wise ('I don't need to search for alternatives') for consumers until it switches to rent seeking, which is where the anti-monopoly laws should kick in.

replies(1): >>46242886 #
3. AnthonyMouse ◴[] No.46242886[source]
> Decisions are efficient if they need to be taken only once.

In other words, open source decentralized systems are the most efficient because you don't have to reduplicate a competitor's effort when you can just use the same code.

> Obviously because they were more efficient than the competition and network effects took care of the rest.

In most cases it's just the network effect, and whether it was a proprietary or open system in any given case is no more than the historical accident of which one happened to gain traction first.

> Having to make choices is a cost for the consumer

If you want an email address you can choose between a few huge providers and a thousand smaller ones, but that doesn't seem to prevent anyone from using it.

> until it switches to rent seeking

If it wasn't an open system from the beginning then that was always the end state and there is no point in waiting for someone to lock the door before trying to remove yourself from the cage.

replies(1): >>46242987 #
4. baq ◴[] No.46242987{3}[source]
> just use the same code

This is the great lie. Approximately zero end consumers care about code, the product they consume is the service, and if the marginal cost of switching the service provider is zero, it's enough to be 1% better to take 99% of the market.