Horizen: A Proposal to Fight Against The 51% Attack
Update Aug 22nd, 2019 – Updated “ZenCash” to “Horizen” as ZenCash rebranded to Horizen as of Aug 22nd, 2019.
The past two weeks have been an exciting and trying time for the Horizen team. However, even in the face of adversity, the Horizen team continues to thrive and deliver on our promises to you, the community.
Approximately two weeks ago, our exchange partner was the victim of a double-spend attack, where a malicious actor was able to spend their ZEN twice. Double spends are possible in all proof-of-work chains due to consensus rules inherited from Bitcoin.
This attack targeted the exchange, not the Horizen blockchain. The Horizen blockchain is still secure, your funds are safe, and your private keys are still secret.
We just want to kill this, we don’t want this ever be a problem again for the industry. – Rob Viglione, Co-founder of Horizen
This attack ignited the Horizen team to create a comprehensive protocol level defensive strategy for our project and for the entire industry. The Horizen development team has proposed an enhancement to the Satoshi Consensus in order to prevent 51% attacks.
The longest chain rule, or Satoshi Consensus, worked well in the relatively decentralized environment in which it was introduced in 2009. Mining resources have since concentrated and dropped in cost for lease, such that the original dominant strategy of playing by the rules no longer holds for all proof-of-work (PoW) blockchains that rely on the longest chain rule. As recent events have proven, in some circumstances it can be economically feasible to launch 51% attacks on operational public blockchain networks. This paper proposes a novel adjustment to Satoshi Consensus that makes it exponentially more costly to launch such attacks for any proof-of-work mineable cryptocurrency system.
We must push for engineering solutions, and we are tackling this issue head-on.
Watch Rob Viglione discuss the efforts to combat this industry-wide issue during the June 13, 2018, Biweekly Community update:
We encourage you to read and provide feedback on our latest white paper
June 14, 2018 @ 9:11 pm
Komodo is already doing this. Here is a great read, and a good example of how.
https://medium.com/@samadsajanlal/how-i-helped-save-a-billion-dollar-cryptocurrency-e60b6275767d
June 19, 2018 @ 4:54 pm
Yeah, it’s called dPoW (Delayed Proof of Work) and $Hush is already implementing it through Komodo notarization…
June 15, 2018 @ 2:26 pm
I’m concerned that this proposal may increase the risk of temporary network partitions becoming permanent.
If you have a temporary bilateral partition, each side of the partition sees its own side’s blocks as being submitted without undue delay. But when the partition resolves, each side sees the other side’s blocks as having been delayed, so those blocks will be penalized. This situation will persist because the nodes that were on each side will retain the memory that their branch of the chain wasn’t delayed. So the chain will split permanently, as far as I can see.
June 18, 2018 @ 2:23 pm
Hi Daira,
in case of a non-malicious split of the network, after the two sides start communicating again, each side will assign a penalty on the blocks received from the other side. Each side will then keep mining it’s own branch but the penalty on the other chain will be reduced block by block because, also if we receive new blocks from the other branch, the other side block height will be equal to this side height and then a -1 penalty will be applied. In this way, after the penalty goes to zero, the side that find the first block will have it’s chain adopted also by the other one and they will join.
July 10, 2018 @ 7:54 am
Could this open up the door to an attack, where the objective would be to keep splitting the chains? Similar to what a DDOS does. Obviously today the cost would be high, and the effort not justifiable, but theoretically, could this result in enough fragmentation to disrupt the ecosystem with endless penalties? What if there are 3 chains? How do the penalties work out? What if we scale to 4 and so on?
June 19, 2018 @ 4:11 am
This is a re-hash of dPoW which is the only consensus algorithm that has stood the test of 51%……..unlike zencash and many other equihash coins. Stop reinventing the wheel and start giving credit where it’s due.
June 19, 2018 @ 2:52 pm
Like Zen…. just use DigiSheild & call it a day… a nice solution and great community … or take a look at VeriCoin…. dPOW is “OK”…..
June 19, 2018 @ 5:10 am
How can you say : WOW we have the solution, our Chief Solution Architect designed it. Design what part exactly ?
Notary ideas / dPOW are just a fact, they are working, and are live on Komodo platform, nothing is new here. Renaming the nodes names from Notary Nodes to Super Nodes is your solution? At least giving credits is a minimum.
June 19, 2018 @ 5:48 am
Mentionned at 18:39, my bad.
June 26, 2018 @ 6:08 pm
The biggest mistake Zen can make is to fork agaisnt ASIC they will proove to be your best asset as the weaker fork away example …BTG