Bram Cohen’s crypto project Chia Network took another important step in the right direction, last month.
Cohen alongside computer scientist Krzysztof Pietrzak released the project’s Green Paper in which they explained how Chia Network is planning to solve the many issues a non Proof of Work (PoW) is facing in a decentralized environment.
What’s the next step? Obviously launching a testnet and put the Proof of Space (PoSpace) consensus protocol to work, probably towards the end of this year. If all goes as planned, the Chia Network mainnet could launch sometime in 2020.
Until then, cryptocurrency specialists & programmers alike can try out Chia’s PoSpace prototype using the code on GitHub.
Before Venturing Forth
Chia Network is not the only project proposing disk space usage instead of computational power to build a blockchain. SpaceMint was another such project that didn’t ultimately materialize into a live and functional blockchain.
In fact, Cohen’s partner at Chia, Krzysztof Pietrzak was involved in SpaceMint for a while, contributing to the project’s whitepaper. Chia took several ideas proposed in the SpaceMint whitepaper and developed them further.
Another cryptocurrency project Chia Network’s Green Paper makes reference to is the pseudo-unknown Ouroboros Praos paper released in November 2017. Ouroboros Praos was a proof-of-stake (PoS) blockchain idea meant to solve some of the theoretical problems the original PoS encountered early on.
Why Green And NOT White Paper
The team chose Green instead of the already established White Paper term to make a valid point. Instead of using computer power that requires a fair amount of electricity that exponentially increases with adoption – like in Bitcoin’s and Ethereum’s case – Chia Network will require disk space in order to participate in the consensus protocol.
Therefore, it’s a green project well-adapted to the modern needs of a society that doesn’t rely on reckless consumption. As a matter of fact, it’s quite possible to become a Chia miner or farmer, you’ll only need a micro-computer like the versatile Raspberry Pi that uses minimum power to function. You won’t need to buy specialized hardware to earn Chia tokens as rewards like the popular ASICs (Application Specific Integrated Circuits) and their many cryptocurrency applications. The life cycle of such hardware is rather compact, thus miners have to change them regularly and basically throw the old ones away as they can’t be used for anything else other than mining.
Proof of Space (PoSpace) & Verifiable Delay Functions (VDF)
What will these so-called farmers need then if not computer power? They will need disk space, whether it’s unused space on their PCs and laptops or internal/external hard drives bought specifically for participating in the Chia consensus.
The consensus algorithm is called Proof of Space and the active participants, space farmers. Since it’s not about solving complex mathematical problems that require a lot of computing power, Chia Network must determine a way to identify who gets to add a block of transactions into the blockchain. To do that, they adapted an idea from SpaceMint and assigned a quality to each PoSpace ‘iteration.’ Whoever has the best quality PoSpace, gets to extend the blockchain.
That’s not all, though. Cohen & Co. took the next step and introduced another consensus algorithm called Verifiable Delay Functions (VDF) to further secure the future network. This is a separate process that needs to be done by the so-called time lords. The blockchain needs at least one time lord to finalize the process of adding blocks into the main chain.
Non Proof of Work (PoW) Vulnerabilities
Satoshi Nakamoto didn’t choose the Proof of Work (PoW) consensus protocol at random. Nakamoto knew it was indeed the most secure possible way to create and distribute a legitimate ledger that could stand the test of time.
Indeed, non PoW based blockchains (PoS & PoSpace for that matter) are much more vulnerable and have their own unique set of problems.
For example, since PoSpace is quite cheap to ‘solve’ in terms of computing power, a malicious attacker could use his powerful CPU or GPU to calculate multiple proofs. The attacker could either try to create many different blocks with a single proof before deciding which one to broadcast on the network, a vulnerability known as grinding (digging). Moreover, the attacker could try to extend multiple chains, not just the ‘longest’ one. In this case, a ‘lesser’ malicious (with double-spending) chain could eventually replace the ‘longest chain.’ This attack is known as double dipping.
Such attacks like grinding and double dipping don’t make sense in a standard PoW based blockchain since the attacker can’t quite distribute his computing power to such tasks.
There is also the long-range attack, another PoS and PoSpace-specific vulnerability, that would allow an attacker to alter the current main chain in private, broadcast it on the network and replace the ‘good’ chain.
All these problems have a solution in Chia Network’s Green Paper.
Chia Network’s Blockchain Structure
To prevent attacks such as digging, the Chia team adapted an idea from the SpaceMint whitepaper and split the blockchain, thus any block, into two sub-chains/sub-blocks.
Much like the Bitcoin’s block header which contains the nonce which is the solution to the PoW, the trunk of a block in Chia Network contains the PoSpace linked by a time parameter to the VDF ‘solution.’
The foliage (similar to Bitcoin’s block body) contains the details of all transactions included in that particular block, plus obviously the HASH of the previous block to link them all together.
Both PoSpace and VDF in any given trunk of the blockchain are unique to prevent digging attacks.
A 38.5% Attack?
Satoshi Nakamoto thought very carefully a good enough consensus protocol that could prevent most malicious attempts to take over the network. The PoW makes any hostile takeover really hard to accomplish, especially on the major blockchains like Bitcoin or Ethereum.
To be able to include double-spending in the blockchain, the attacker must control at least 51% of the network’s hashing power in any traditional PoW system.
Nevertheless, non-PoW systems are trickier and more often than not, attackers need less than 51% to take over the network. Specifically, according to Chia Network’s Green Paper, the network is secure and double-dipping-free as long as ‘73.1% of the space is controlled by honest farmers.’ That means an attacker would actually need just 26.9% of the space for a hostile takeover.
To lower the honesty threshold, the Chia team will actually let honest farmers to double dip ‘to some extent,’ without using much more computing power. This would actually set the network takeover limit from 73.1% down to 61.5%. As a result, an attacker will need 38.5% of the space to actually bend the Chia Network to his/her/their own will. Definitely, this isn’t as much as the 51% majority in traditional PoW systems, yet much more than the initial 26.9%.
What is your take on the latest Chia Network’s Green Paper? Are you excited about the project’s future? Let us know in the comment section below!
Images courtesy of Chia.net, Wikimedia & Needpix.com.