Proof-of-Work and Proof-of-Stake are **sybil control mechanisms**. PoS doesn't achieve consensus by itself. It has to be coupled with a protocol, such as PBFT, or Ben-Or, or Tendermint/Cosmos, or Avalanche, to make decisions.
Show this thread
Ok, there is a terribly wrong framework emerging around consensus protocols. People think that PoW and PoS are consensus protocols, and that they are the only two consensus protocols out there. This is false. Let me explain.
-
-
-
PoW, by itself, isn't a consensus mechanism. In BTC/BCH, it works with the heaviest/"longest" chain selection rule to achieve consensus. PoW by itself can be used to thwart spam, as originally proposed by Dwork&Naor. PoW doesn't get you agreement, it gets you rate limiting.
Show this thread -
There are many consensus protocols (mechanisms), and there are many Sybil control mechanisms. These two kinds of mechanisms are distinct and separate. A system or coin will end up combining one with the other. Of course, not all combos make sense. But many combos are possible.
Show this thread -
One could use Avalanche either with PoW or PoS, if one wanted. One could use a PoS coin, which had in it a PoW mechanism to make sure that a rogue but staked actor does not flood the network layer.
Show this thread -
The big design decisions in coins aren't between PoW vs. PoS. They are between the consensus protocols themselves, as they determine performance, scale, energy expenditure, and security.
Show this thread -
This concludes today's
#microlecture. I hope it clarifies the discourse around coins and avoids confounding mechanisms that are independent.Show this thread
End of conversation
New conversation -
-
-
Would it be right to say POS/POW are how eligible participants are decided (the entry ticket) to vote.... and PBFT/Ben-Or/Tendermint/Avalance -- the actual code of conduct/rules for voting?
-
Looking forward to more of your updates. I will prepare an animation / simplified version of these concepts based on your texts and publish with your permission to dispel the common myths that exist about Distributed Consensus aka The brain of the blockchain
-
Fantastic. Do let me know and I'll circulate them as much as I can.
End of conversation
New conversation -
-
-
What about
#DAG? -
That refers to a third component in system design: the data structure you build via consensus and with the help of your sybil deterrence mechanism.
-
One could use PoW+DAG+modified heaviest chain. That'd be GHOST. One could use PoS+DAG+Avalanche, as in Ava coin. Or do PoS+multisets+Avalanche just as well. Or do PoS+forests or chain+Snowball. All are very similar, only one has a DAG. Certain combos are more efficient.
-
Thanks, this clears up my thinking about Avalanche. I was confused how it solved Sybil attacks and node randomness, but if it's build with a PoS mechanism, it doesn't need to do those things alone.
End of conversation
New conversation -
-
-
Agreed. What about PoA? That kind of rubs me the wrong way? Isn't that just a euphemism for centralized control of nodes? Just sounds cooler to be "proof-of" something? Or, am I missing something?
-
Indeed, that terminology was created to try to fit into the "everything must be proof of something" framework. That framework is broken.
End of conversation
New conversation -
-
-
How do you view
@ChiaNetworkInc? -
I understand its sybil deterrence mechanism. I don't understand its consensus protocol.
-
Got it. We haven't learned about it yet, but are interested to learn more
End of conversation
New conversation -
-
- 1 more reply
New conversation
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.