7/ Being tolerant to byzantine failures, by the way, means not only being able to handle node failures of the simplistic kind, e.g. network outages, but rather, any type of malicious behavior the node can muster.
-
Show this thread
-
8/ The easiest way to start building a consensus algorithm is to use some sort of round-robin signature scheme, where each of the validators take turns signing a block, and all signatures are then tallied up by the group before a blocks is deemed valid. That's ultra-simplistic.
3 replies 32 retweets 192 likesShow this thread -
9/ What you'll immediately come to realize when building that type of solution is that if you're using a predetermined order for the consensus round, the whole system will freeze up if one of the validating nodes gets unplugged. So you'll have to come up with a solution for that.
5 replies 28 retweets 214 likesShow this thread -
10/ An ultra-simple solution is to just abandon the round-robin part of the design and instead go for a threshold scheme where the validators' signatures are broadcast and just needs to be tallied up to a certain threshold, instead of waiting for specific "turns" in the rounds.
5 replies 27 retweets 196 likesShow this thread -
11/ I want to say this in the least pompous way possible, but this was something that I was able to figure out as a CS student with no previous experience designing consensus algorithms. Yet yesterday, the whole NEO blockchain failed because they had not thought about this issue!pic.twitter.com/cHFJnUDLeB
22 replies 161 retweets 728 likesShow this thread -
12/ The Discord message above was written by Malcolm Lerider, who's the Senior R&D Manager for NEO. How can a node failure of the most simplistic type possible (network loss) be referred to as an "edge case" scenario, in a supposed Byzantine fault tolerant consensus system?!
8 replies 69 retweets 597 likesShow this thread -
13/ I'll reiterate my point: A *Byzantine* *fault* *tolerant* consenus system's whole purpose is to be designed to handle ANY error or malicious behaviour from a subset of the nodes and still reach consensus. In NEO, their system couldn't handle the most simple error of one node.
5 replies 78 retweets 469 likesShow this thread -
14/ This means that NEO is not a *Byzantine fault tolerant* consensus system. NEO is not even a *fault tolerant* consensus system.
7 replies 87 retweets 464 likesShow this thread -
15/ This happened on the live NEO mainnet, a cryptocurrency valued $7.8 billion. Without exaggeration, this is the most blatant display of sheer incompetence I've ever seen from a project in this space. No project with *any* form of code review would have allowed this to happen.
38 replies 126 retweets 777 likesShow this thread -
16/ Thankfully, there only appears to be 23 deployed smart contracts in total on the NEO blockchain, so no one is really using this crap atm. Let's hope it stays that way (and reading this thread, with $35260 in fees to deploy a contract I suspect it will https://www.reddit.com/r/NEO/comments/7qkaz4/how_mature_are_neos_smart_contracts_for_a_real/ …)
132 replies 98 retweets 701 likesShow this thread
interesting tweet storm on $NEO. would like to see your take on $EOS 

#cryptoassets
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.