4 ms·
As a person who is allergic to trends like this, can someone bring me up to speed on the details of the blockchain? Particularly: 1) Is it resilient to catastr
by ken47 4y ago
As a person who is allergic to trends like this, can someone bring me up to speed on the details of the blockchain? Particularly:
1) Is it resilient to catastrophic failure? Does there exist an N < 100 such that if there is an EMP, a computer virus, or some other anomaly that wipes the data from more than N% of the nodes, the block chain enters an irrecoverable failure state?
2) How scalable is it? Sure, it might work for a very small % of the world's transactions, but could it ever handle the sort of load that is taken on by say, Visa credit cards worldwide, while still allowing transactions to complete in a "reasonable" amount of time (that the average person would tolerate)? If not, what is the upper limit of requests per second while maintaining some sufficient upper bound for transaction time?
Feel free to be completely hand wavy. This is just a thought exercise for now.
- horsecabletruck 4y ago1) Depends on the protocol, chain and platform. Ethereum itself is unlikely to be impacted by an EMP as the number of validators is large and widely distributed across the globe. 2) A chain like Ethereum is positioning itself as a secure and decentralized settlement base layer upon which more scalable higher-level and application-specific layers might be built, typically with rollups that batch many operations together. Other chains are making security and centralization trade-offs to provide scalability at this bottom base layer. The latter may not always work out, as seen with Luna crashing or Solana network going offline.
- twox2 4y agoI'm not learned enough to answer question 1... but I don't think the fiat system is fully resistant against catastrophic black swan events either. As for question 2 - it depends on which blockchain and which use case. Blockchain does not need to replace Visa in it's entirety in order to add value.
- ken47 4y agoI'm not taking a side on the fiat vs. crypto debate. I'm inquiring purely about the technical aspects of the blockchain. That being said, with non-blockchain data storage, one could create backups, so that if all the live running data stores were somehow destroyed, one could recover to some previous point in history with the redundant copies. I'm trying to understand whether this, or any other sort of failure recovery, is possible with blockchain technology.
- twox2 4y agoI see. I think it's possible. If there's a catastrophic event that nukes everything, it's certainly possible to restore from a prior state... or even multiple prior states. You would effectively get a fork or multiple forks and then the ones that folks decide to use en massed would become the dominant chain. It's happened already when there have been disagreements in how to scale protocols, or even in the case of a pseudo-catastrophic event like the DAO hack on ethereum.
- vmception 4y ago1) No, but with sharding it could happen. Sharding is a scalability solution being pushed for. Wouldn’t be failure it would miss some information. An application not designed for this missing information could break irrecoverable. Its a data availability issue. 2) Most of today’s largest blockchains can handle VISA levels of transactions and the bandwidth has been used for something like NASDAQ levels of transactions (but doesn't reach that) since its mostly trading of assets on those blockchains. (The “merchant adoption” use case was always a small use case being masqueraded as a “killer app”by irrelevant people grasping for any use case, but its worked to get more relevant interest into the system, so maybe you heard that, the VISA benchmark is still useful though) When a form of sharding is implemented they get an additional 1-2 orders of magnitude more bandwidth. We’ll see how it pans out.