3 ms·

Search for "on stake and consensus" I'd link to it, but I'm on mobile right now. Btw- I know you're new to blockchains because poelstra has been writing seminal

Search for "on stake and consensus" I'd link to it, but I'm on mobile right now. Btw- I know you're new to blockchains because poelstra has been writing seminalpapers for years and you haven't even heard of him. We know things are true in a blockchain because energy went into it. If only a single Bitcoin node were alive, and 99% of the network disagree with him - the energy would point to the truth. This is what makes it objective. With subjective chains there are a huge suite of problems in the form of stake grinding and long/short range attacks. Why does this matter? Non-objective quorum based sysetms have been around for decades. They're not more efficient than centralized systems for many economic reasons, some of which were really well laid out in a Isabella kaminska paper on the subject (let me know if you need a link) Tldr: blockchains aren't paxos or raft. Blockchains mostly suck, but do one thing well: resist censorship

sanswork 11y ago
That entire paper is about PoS and dynamic membership systems where the participants aren't known. Relevance to the current discussion? Zero. All those attacks are based on unknown dynamic participants which isn't an issue. So ya argue against a system they aren't trying to implement.
- brighton36 11y ago
  So if the trust each other - passing messages in the style of SMTP (or soap) is drastically better than a hokey round robin "blockchain emulation" architecture. At least you understand now that immutability requires energy...
  - pyre 11y ago
    This discussion might be a little above my pay-grade, but you're seeming to conflate "unknown participants" with "untrusted participants" from this response. It's possible to know someone without trusting them. I think that it's pretty apparent that there is a difference between an unknown anonymous actor, and an actor that you know, but just can't 100% trust.

sanswork 11y ago

That entire paper is about PoS and dynamic membership systems where the participants aren't known. Relevance to the current discussion? Zero. All those attacks are based on unknown dynamic participants which isn't an issue. So ya argue against a system they aren't trying to implement.

brighton36 11y ago
So if the trust each other - passing messages in the style of SMTP (or soap) is drastically better than a hokey round robin "blockchain emulation" architecture. At least you understand now that immutability requires energy...
- pyre 11y ago
  This discussion might be a little above my pay-grade, but you're seeming to conflate "unknown participants" with "untrusted participants" from this response. It's possible to know someone without trusting them. I think that it's pretty apparent that there is a difference between an unknown anonymous actor, and an actor that you know, but just can't 100% trust.