Topic: one ip one vote (Read 341 times)

At least for shortage of IPv4 address, it's solved by IPv6 address.


I wouldn't say the idea (Proof of Space) failed when Chia coin is failed. Although these days everyone use high-end/enterprise SSD rather than HDD due to various reasons.

Getting more expensive with the IP v4 exhaustion but yes. But yes.
Also keep in mind at least the last time I checked the largest holder of IP space is still the US government (DoD).
You also have mega corps like Apple and Ford with /8 subnets getting them 16,777,216 IPs in the IP4 space more IP6 then you can even think about.

And in the end, basing it on something like this sill makes it PoS just what you are staking is different.

-Dave

Truth be told, running a node just for the sake of the network provides minimum help. Bitcoin isn't a newborn peer-to-peer network anymore. There are dozens of thousands of full nodes that archive the chain, and do help bandwidth-wise. In both clearnet and anonymous network(s), there are just enough to keep things decentralized in case of some shortage.

It's also trivial to just buy lots.

It is trivial for someone to access an arbitrary number of IP addresses. It is also normal for many legitimate users of various services to access a changing IP address.

ip based voting on blockchain would give control to the internet providers, plus allow every "miner" to be tracked through their IPs. Plus creating IP addresses is not a hard task the way PoW computation effort is.



For validating transactions you either need something self-contained within the system so it is therefore controlled by the system itself - this is how Proof of Stake works, or you need something that requires an external scarce resource in its raw form without any need to trust that resource.


Computational work is an external scarce resource. There is also some crypto that uses hard disk space which is another external scarce resource though I guess that sort of mining quickly destroys hard drives so that idea failed. IPs are not a scarce resource so that doesn't work. Identity is a scarce resource but would require trust to verify the identity and the whole point of Bitcoin is that it is trustless. People have at times suggested all that mining computation work should go TOWARD something to make it "useful" but that requires a trusted body to decide what the work goes toward so that doesn't work. The reason nobody has come up with something better than PoW is because it is hard to figure out something that would be better and still accomplish the same thing. You need a raw scarce external resource that itself requires no verification/trust and can't be spoofed, and of course it needs to be able to be directly input in a digital fashion.

It's probably possible but what's the point? Why does the world need such a crypto? ISPs are running short of IPv4 addresses for quite some time already. Creating crypto like that would lead to further shortage and even disruptions on the networks. Also, how to treat dynamic/pool IPs? Many caveats and little reason imho.

The issue with IPs, outside of the ones mentioned is the fact that there are so many (millions and millions and millions) issued to legacy holders.
https://en.wikipedia.org/wiki/List_of_assigned_/8_IPv4_address_blocks

The issue with domains is that was not mentioned it that there are a lot of businesses with massive name holdings that they run for other people.

That is the advantage of PoW, you actually have to DO something. Yes, I know the miners are doing it but you have to invest, and then keep investing or get left behind.

Everything else is just a different version of PoS.

-Dave

Reachable IP addresses don't help the network (like Tor), running a "node" does. And to verify someone is running actually running a node is more complicated than just reaching their IP. In other words it is very difficult, whether it is a Tor node or a cryptocurrency full node.

I think it's certainly possible, and would probably even be very beneficial for certain services. If you have a service where you want many nodes, like creating a decentralized internet or something like Tor, you might want to reward your users based on the IP addresses they're contributing to the system. People would "game" it by renting IPs and all that, but that would only benefit your system. This assumes however that the system would mandate IPs stay online in order to get their "stakes."

Why do you think it's possible? How about IPv6?

---

Do you imply cryptocurrency isn't really decentralized? ICANN have some influence towards internet centralization

---

It's not proposal, saotoshi simple mention weakness of one-IP-address-one-vote. Here's whole paragraph which contain sentence mentioned by OP.

It has never been true, either. It has always been 'one hash, one vote', in a way. The number of votes solely relies on how many hashes you can churn out in a certain amount of time.

Got nothing to do with the software, but yes, Bitcoin Core removed mining functionality.
One H/s = one vote, makes most sense, I guess. Exa-, Peta-, Tera-, those are just metric prefixes. Don't change what unit you're working with.

Not going to work not only because of CGNAT but also because servers can have an IPV4 and IPV6 address, and what's more, multiple IPV6 addresses as they are so cheap to allocate. Also you can run multiple nodes on the same host with different ports and protocols.

It is perhaps a flawed proposal by Satoshi, but fortunately Gavin and the core team of developers ironed those issues out shortly after his disappearance.

Don't forget about people who have dynamic IPs which change easily.

It was 1 CPU 1 Vote and you could say that a CPU is representation of computing power or hash power which in turn means someone with a better CPU got "more vote" than someone with a weaker CPU.
This means nothing has changed, you still get a vote equivalent of the amount of computing power you posses. If you have a better ASIC or more number of ASICs you get more votes.

It's not just about decentralization. It's also about verifiability and consensus. tromp already mentioned that you can't verify the blockchain history after the fact if you use this method. How do you envision to prove the existence of certain IP addresses years later?

It's also interesting to me that you go from a clear explanation as of why 'one-IP-one-vote' is a bad idea:
to contemplating creating a cryptocurrency based on 'one-IP-one-vote':

You quoted Satoshi explaining why the idea is bad in the same sentence..

Which IP exactly?

IPv4 has 2^32 IPs.

IPv6 has 2^128 IPs.

Right. I don't think it's even possible...

I'm not sure it is. Blockchain histories must be fully verifiable.
How does a future verifier check that some IP address was actively involved and entitled to a reward?

It is technically possible, but there is no point, because it's trivial for someone to allocate multiple IP addresses at the same cost. Furthermore, it provides minimum privacy and gives too much control to the ISP hosts. They can de-anonymize every node, and perhaps even forbid someone from participating.

That doesn't mean Bitcoin was ever one-IP-one-vote. And you can still mine from Bitcoin Core, it just pointless without the proper infrastructure.

Please correct me if I am wrong but one node / One vote is not  true today and since a long time.

People are not solo-mining BTC anymore through Bitcoin Core

I don't know what we should say now, one ASIC one vote ? One Eh/s one vote ?

the bitcoin white paper says: "if the majority were based on one-IP-address-one-vote, it could be subverted by anyone
able to allocate many IPs".
Is it possible to create a cryptocurrency in which IP addresses are "staked"?.
If a node acts maliciously its ip address is blacklisted.
So anyone with internet access can act as a validator.