4 ms·

You add in a centralized service that enforces these rules and feature-wise it is equivalent, except the service is in control. The leading service is the compa

by Osmose 5y ago

You add in a centralized service that enforces these rules and feature-wise it is equivalent, except the service is in control. The leading service is the company that is most well-funded, well-connected, innovative, etc. Whereas with a proof-of-work blockchain, whoever has 51% of the network's mining hash rate is in control, which is effectively who is the most well-funded, well-connected, innovative, etc. Or a proof-of-stake blockchain, where whoever has 51% of the network's stake, which is effectively who is the most well-funded, well-connected, innovative, etc.

k__ 5y ago
Implying the most well funded is enough to get >50%, right? ;)
- Osmose 5y ago
  If we've learned one thing from capitalism it's that it is _not that unlikely_ for a small group of people to get more than half of all the money. The structure still concentrates power.
eightysixfour 5y ago
This is pretty reductive IMO. Instead of trying to sit and think about how it is an inefficient, slow version of the same databases we are used to using, try and steel man the argument a bit and think about what is unique and why some people value it. It will cost millions of dollars to maintain a one hour attack against a PoW chain. It is crazy inefficient, it is slow, but it provides a level of trust that is unique. A software license NFT that lives on a blockchain is a perfect example - would you rather have that or one that depends on a centralized company keeping their server's running (like Adobe's with CS3 and before)?
- Osmose 5y ago
  > try and steel man the argument a bit and think about what is unique and why some people value it. You were saying that the verifiable provenance is part the value, and I'm saying that that provenance is just as trustworthy as if a centralized service provided it because my trust that what they're asserting (that person X owns token Y) is true depends on how vulnerable they are to tampering, and all these systems are ultimately controlled by traditional factors like how much money you have access to, who you know, etc.
syntheweave 5y ago
They're organizationally different, though, and therefore have different outcomes because the frictions lie in different places. With the centralized solution all responsibility is defined with the operator. In practice this means that virtual goods markets are underexplored because it's a huge platform service investment - only the likes of Valve and Roblox will consider it. And in those cases it's also a monopolistic play since the exits from the market are barricaded in various ways, with restricted APIs and hidden data and so on. The blockchain solutions create a public database with multiple frontends. A third party can add on whichever service they want to, e.g. verify the data's provenance, present trading opportunities or incentives. Therefore the platform builds from an aggregation of independent services, more like the early Internet. This is the "consensus" part of the tech in action, and also differentiates it from signed receipts. The agreement of consent precedes the participation. If you don't like the agreement, it's public data, so you can launch alternatives. Threat of forking keeps stakeholders willing to stay at the bargaining table. It doesn't matter if you are a well-capitalized stakeholder if you torpedo the value of the chain. However, the early entrants to every space in crypto look scammy and try to jump ahead with hype because it's faster to run a scam than to actually work through the civic problems of maintaining the chain. To date, they always wash out with the bear markets.
- Osmose 5y ago
  > Therefore the platform builds from an aggregation of independent services, more like the early Internet. Same as creating an open protocol like the article suggests and getting participants to opt-in to supporting that protocol. The difference being that open protocols would require you to either accept _any_ provider or vet and choose which services you trust, vs the blockchain where the trust is in the chain itself and it's inability to be tampered with, i.e. the consensus you're talking about. And the core of my criticism here is that the factors that give one person or a group of people the ability to tamper are the same; the companies that don't really experience consequences for bad behavior are large, well-funded, well-connected, and these things will also let you buy tons of mining rigs, low-power-cost facilities, and connect you with other wealthy miners who are interested in whenever the next DAO reversal needs to happen.

by Osmose 5y ago

k__ 5y ago

Implying the most well funded is enough to get >50%, right? ;)

Osmose 5y ago
If we've learned one thing from capitalism it's that it is _not that unlikely_ for a small group of people to get more than half of all the money. The structure still concentrates power.

eightysixfour 5y ago

This is pretty reductive IMO. Instead of trying to sit and think about how it is an inefficient, slow version of the same databases we are used to using, try and steel man the argument a bit and think about what is unique and why some people value it. It will cost millions of dollars to maintain a one hour attack against a PoW chain. It is crazy inefficient, it is slow, but it provides a level of trust that is unique. A software license NFT that lives on a blockchain is a perfect example - would you rather have that or one that depends on a centralized company keeping their server's running (like Adobe's with CS3 and before)?

Osmose 5y ago
> try and steel man the argument a bit and think about what is unique and why some people value it. You were saying that the verifiable provenance is part the value, and I'm saying that that provenance is just as trustworthy as if a centralized service provided it because my trust that what they're asserting (that person X owns token Y) is true depends on how vulnerable they are to tampering, and all these systems are ultimately controlled by traditional factors like how much money you have access to, who you know, etc.

syntheweave 5y ago

They're organizationally different, though, and therefore have different outcomes because the frictions lie in different places. With the centralized solution all responsibility is defined with the operator. In practice this means that virtual goods markets are underexplored because it's a huge platform service investment - only the likes of Valve and Roblox will consider it. And in those cases it's also a monopolistic play since the exits from the market are barricaded in various ways, with restricted APIs and hidden data and so on. The blockchain solutions create a public database with multiple frontends. A third party can add on whichever service they want to, e.g. verify the data's provenance, present trading opportunities or incentives. Therefore the platform builds from an aggregation of independent services, more like the early Internet. This is the "consensus" part of the tech in action, and also differentiates it from signed receipts. The agreement of consent precedes the participation. If you don't like the agreement, it's public data, so you can launch alternatives. Threat of forking keeps stakeholders willing to stay at the bargaining table. It doesn't matter if you are a well-capitalized stakeholder if you torpedo the value of the chain. However, the early entrants to every space in crypto look scammy and try to jump ahead with hype because it's faster to run a scam than to actually work through the civic problems of maintaining the chain. To date, they always wash out with the bear markets.

Osmose 5y ago
> Therefore the platform builds from an aggregation of independent services, more like the early Internet. Same as creating an open protocol like the article suggests and getting participants to opt-in to supporting that protocol. The difference being that open protocols would require you to either accept _any_ provider or vet and choose which services you trust, vs the blockchain where the trust is in the chain itself and it's inability to be tampered with, i.e. the consensus you're talking about. And the core of my criticism here is that the factors that give one person or a group of people the ability to tamper are the same; the companies that don't really experience consequences for bad behavior are large, well-funded, well-connected, and these things will also let you buy tons of mining rigs, low-power-cost facilities, and connect you with other wealthy miners who are interested in whenever the next DAO reversal needs to happen.