4 ms·

I suspect this is how most "blockchain" projects will end up in 5 years; no utility from the fact that it's a blockchain and practically just a database.

davidgerard 9y ago
I found Bitfury's Exonum blockchain quite reasonable in its claims in this regard: https://github.com/exonum/exonum-doc/blob/master/src/get-started/what-is-exonum.md https://github.com/exonum/exonum-doc/blob/master/src/get-sta... > Like distributed databases, such as Cassandra, MongoDB or MySQL Cluster, blockchains are replicated on multiple nodes. However, compared to ordinary distributed DBs, blockchains are resistant against much wider range of attacks: > 1. Blockchains are resistant against failures of any single node (or even multiple nodes at the same time). The threat model includes nodes being switched off, isolated from the rest of the network, or even completely compromised (say, by a hacker) > 2. Blockchains can be deployed in a decentralized network, where there is no single administrator managing all nodes. This significantly reduces risks of data corruption and the bias in the system (e.g., preferential treatment of some participants) so, they posit Exonum as a particularly resilient distributed database. Which is fine, y'know. I've said for a while that I think the one good outcome of Blockchain(tm) hype is going to be businesses discovering the usefulness of transaction ledgers tamper-proofed with hashes of everything. Developers of course know this as git.
- gaius 9y ago
  Developers of course know this as git. Well, except for the history rewriting "feature".
  - davidgerard 9y ago
    Which turns out to be a highly marketable feature for "blockchains" to organisations who actually have money to pay for this. e.g., https://www.reuters.com/article/us-tech-blockchain-accenture/accenture-breaks-blockchain-taboo-with-editing-system-idUSKCN11Q1S2 https://www.reuters.com/article/us-tech-blockchain-accenture... Remember, in this context we're talking about business use, not cryptocurrency use.

davidgerard 9y ago

I found Bitfury's Exonum blockchain quite reasonable in its claims in this regard: https://github.com/exonum/exonum-doc/blob/master/src/get-started/what-is-exonum.md https://github.com/exonum/exonum-doc/blob/master/src/get-sta... > Like distributed databases, such as Cassandra, MongoDB or MySQL Cluster, blockchains are replicated on multiple nodes. However, compared to ordinary distributed DBs, blockchains are resistant against much wider range of attacks: > 1. Blockchains are resistant against failures of any single node (or even multiple nodes at the same time). The threat model includes nodes being switched off, isolated from the rest of the network, or even completely compromised (say, by a hacker) > 2. Blockchains can be deployed in a decentralized network, where there is no single administrator managing all nodes. This significantly reduces risks of data corruption and the bias in the system (e.g., preferential treatment of some participants) so, they posit Exonum as a particularly resilient distributed database. Which is fine, y'know. I've said for a while that I think the one good outcome of Blockchain(tm) hype is going to be businesses discovering the usefulness of transaction ledgers tamper-proofed with hashes of everything. Developers of course know this as git.

gaius 9y ago
Developers of course know this as git. Well, except for the history rewriting "feature".
- davidgerard 9y ago
  Which turns out to be a highly marketable feature for "blockchains" to organisations who actually have money to pay for this. e.g., https://www.reuters.com/article/us-tech-blockchain-accenture/accenture-breaks-blockchain-taboo-with-editing-system-idUSKCN11Q1S2 https://www.reuters.com/article/us-tech-blockchain-accenture... Remember, in this context we're talking about business use, not cryptocurrency use.