Vitalik Buterin, co-founder of Ethereum, believes bitcoin is more likely than Ethereum to have an intentional split in 2017, according to a tweet he posted in response to a tweet about the United Arab Emirates choosing Ethereum over bitcoin for a real estate pilot program. Ragnar Lifthrasir, chairman of the International Blockchain Real Estate Association, had tweeted the UAE is being irresponsible by using Ethereum because it is too unproven, too mutable and unstable compared to bitcoin to be used as a real estate database.
@Ragnarly Past performance not indicative of future results; I actually think BTC is more likely than ETH to split in 2 in 2017
— Vitalik Buterin (@VitalikButerin) December 30, 2016
Lifthrasir’s comments touched on the proof-of-stake versus proof-of-work debate.
Lifthrasir, in criticizing the UAE’s decision, pointed to the exploit of the DAO, caused by a vulnerability in its smart contract programming language that allowed the removal of $145 million worth of Ether. Ethereum developers created a new network with new rules. Miners adopting the new version executed a hard fork on July 20, 2016. Such events showed a small group of individuals can make significant changes to Ethereum for investment-boosting or political reasons, Lifthrasir said.
Lifthrasir Cites Ethereum Issues
The hard fork created another problem because not all Ethereum miners accepted the changes and continued to mine the older chain called Ethereum Classic, causing the two chains to compete, Lifthrasir further noted. This created two databases. Hence, a real estate title application built on the earlier database could have two competing title claims, which Lifthrasir cited as a fatal flaw.
Ethereum has been hard forked intentionally three times between August and December of 2016 to address technical issues, he added. It experienced an unintentional hard fork on November 24, 2016, when Parity and Geth clients fell out of sync, temporarily creating two blockchain versions.
Ethereum developers considered switching to proof-of-stake from proof-of-work, posing a risk to its functionality, stability and security.
Bitcoin has nearly 400,000 times the hashing power of Ethereum and its market capitalization is 23 times greater, Lifthrasir said, two data points that indicate the level of network hacking difficulty.
Buterin Clarifies
Buterin later clarified that he was referring to an intentional, ETC/ETH style fork, not an accidental consensus split.
@Ragnarly To clarify I’m referring to an ETC/ETH style fork, not accidental consensus splits
— Vitalik Buterin (@VitalikButerin) December 30, 2016
Buterin also tweeted that “the notion that blockchains are ‘all about’ one-X-to-rule-them-all is silly maximalism.”
Lifthrasir responded that if the UAE government has a two-step in when a blockchain fails and splits into a pair of competing chains, they have selected the wrong blockchain.
Buterin replied that the UAE can also simply “choose the chain that makes sense.”
@Ragnarly @macwind Competition = free market. The notion that blockchains are “all about” one-X-to-rule-them-all is silly maximalisn.
— Vitalik Buterin (@VitalikButerin) December 30, 2016
(The Dubai Museum of the Future Foundation has established the Global Blockchain Council to explore transactions through the blockchain platform, CCN reported earlier this year. The council will examine the blockchain’s impact on future business and finance, and its role in facilitating transactions within both the financial and non-financial sectors as ways to improve efficiency and reliability levels.)
Also read: Ethereum successfully hardforks to address vulnerable vectors
Buterin Expands On Proof Of Stake
Buterin continues to expand on proof-of-stake. He recently explained his position in Medium. He observed that a blockchain protected only by social consensus would be too slow and inefficient to address disagreements; economic consensus thereby services an important role in protecting safety properties in the short term.
Where proof-of-work operates on a logic of massive power incentivized by massive rewards, proof-of-stake relies not on rewards for security, but on penalties.
The best protocols are those that work well under various models and assumptions, Buterin noted – “economic rationality with coordinated choice, economic rationality with individual choice, simple fault tolerance, Byzantine fault tolerance…”
Consensus protocols that work as quickly as possible carry risks and must be approached very carefully if they should be approached at all, he noted.
Image from Flickr.
Our Social Networks: Facebook Instagram Pinterest Reddit Telegram Twitter Youtube