4 ms·
How is this different from publishing the pgp signed sha for the digital file, which is an inherent property of the file? For example, how Ububtu authenticates
by mch82 5y ago
How is this different from publishing the pgp signed sha for the digital file, which is an inherent property of the file?
For example, how Ububtu authenticates its Linux installers, https://ubuntu.com/tutorials/how-to-verify-ubuntu#1-overview https://ubuntu.com/tutorials/how-to-verify-ubuntu#1-overview
- zwarag 5y agoI'm not sure, but I think the trust part (münchhausen trilmma) is handled differently between pgp and blockchain. In pgp you have the web of trust and you choose your "entry" point and who you trust. With nfts you're on blockchain, and have the hole network have a consensus. Therefor the "entry" point for trust is different.
- Macha 5y agoWith blockchain, is it not the person who spends money to verify it gets trusted? There's no sense that the other nodes endorse the message, just that it won the auction.
- zohch 5y ago> With nfts you're on blockchain, and have the hole network have a consensus. Consensus of something completely different than the element you are trusting with the web of trust, so ... basically ... blockchain is better because it does not address the problem at all but instead solves a problem that does not even exist in PGP? Or what is it exactly that you are saying?
- zkldi 5y agoWhat?
- api 5y agoMy name is President Joe Biden, and I endorse this SHA-256...