3 ms·
> Beyond cryptocurrencies? Go on, I'm all ears. I'm always amazed by how often I hear that claim (on HN and elsewhere) and yet nobody seems in a rush to be spec
by dsacco 9y ago
> Beyond cryptocurrencies? Go on, I'm all ears. I'm always amazed by how often I hear that claim (on HN and elsewhere) and yet nobody seems in a rush to be specific about it.
A Byzantine fault-tolerant, distributed NTP protocol that does not rely on a trusted third party. That is a concrete, legitimate use case for a blockchain that meaningfully improves through a trustless, permissionless and decentralized system. For example, this could augment Google's RoughTime NTP protocol.[1]
More abstractly, blockchains resolve many hard problems in cryptographic protocols that would otherwise require trusted third parties. There has been nontrivial academic attention for this particular possibility, including support from DARPA and the NSF.[2]
Now you have heard someone making this claim and being specific about it on HN.
________________
1. https://www.imperialviolet.org/2016/09/19/roughtime.html https://www.imperialviolet.org/2016/09/19/roughtime.html
2. https://eprint.iacr.org/2017/935.pdf https://eprint.iacr.org/2017/935.pdf
- Karrot_Kream 9y agoAre you specifically talking about recording times onto a new blockchain, or do you mean something like a timestamp based on block IDs?
- dsacco 9y agoRecording signed time attestations onto a blockchain, establishing BFT consensus of the current time, using the blockchain consesus as an authoritative source for the current time.
- Karrot_Kream 9y agoThat would be a very interesting alternative to the current NTP infrastructure.