Blockchain 103: Working Hand in Glove with Nakamoto
Agreements are only as good as those witnessing it. Without witnesses, its enforcement will often boil down to the caprice of time, space, and whichever side of the bed they wake upon.
Like a broom, the more the witnesses, the more the difficulty in breaking the agreement.
Still, the best witnesses remain those that stand to lose something equally valuable if they should act contrary to the terms of the agreement. In the traditional world, what is often at stake rarely strays beyond the witness’s localized reputation — a value that may be worth much less than the ink they signed their signature with.
By giving witnesses personal stakes in events they wish to validate, blockchain technology is making the validation of any agreement worth its weight in gold.
However, just how effective is this system?
Nodes are witnesses to any and all events on the blockchain, but, not all nodes can validate the same event at the same time. Doing so would be sheer chaos; an enormous waste of resources since the ledger will mostly be documenting the same response over and over again from individual nodes resulting in excessive bloat, insecurity, and instability.
At its core, blockchain is a linear chain of blocks chained together with each block containing data from the preceding blocks. Each block possesses a unique signature that must first be verified by the network before being added to the chain by a validator. This is known as block selection.
Choosing a validator is based on a lottery system and the probability of them depends on the magnitude of required resources committed — computing power for PoW and stake for PoS systems.
Nakamoto’s Secret
Nakamoto Consensus is governed by two rules:
- The rule of the first
- The rule of the longest
The rule of the first sets the condition for determining who wins the lottery. Nodes compete among themselves to submit answers to a mathematical puzzle. The first to submit the nearest possible answer wins the lottery and gets to validate the block. However, this rule is not without its own complications.
What if two nodes submit the same result at the same time? What if a block is added without containing information from the previous block due to network delay? What if a node maliciously chooses to ignore certain blocks? In most traditional database systems, the problem is either one Ctrl+Z away or database restore. However, when dealing with immutable data with no central authority pulling the strings, anticipating the slightest possible deviation becomes mission-critical. Therein, lies the beauty of the second rule.
The second rule will only consider the longest chain on the network as the only valid chain based on the assumption that this chain has the most validated computational resources or stakes committed to it, representing the majority of the nodes on the network. Blocks that do not conform to this rule are automatically rejected by the rest of the nodes on the network. Nakamoto-styled Consensus (NSC) blockchains have a corruption threshold of up to ½. What this means is as long as 50% of the nodes remain honest, the network can withstand any attack to change the data.
Turning a New Leaf
Block finality is the certainty that transactions executed on the network cannot be altered after a number of blocks have been added to the network. For NSC blockchain, once a certain number of blocks have been added to the initial block that executed the transaction, the cost of undoing the transactions of this block becomes prohibitively expensive due to the rule of the longest chain. The more blocks added, the more expensive it becomes. This is known as probabilistic finality.
The block probabilistic finality for Bitcoin is six blocks. The average wait time on Bitcoin’s blockchain for the creation and confirmation of one block and the next is 10 minutes. A six-block wait for finality means Starbucks having to wait an hour to be absolutely sure you paid for that latte.
Wait, what? Pause galactic domination for an hour because someone had to grab a cuppa joe? That just wouldn’t cut it for any self-respecting overlord.
Absolute finality is a vast improvement over this, finalizing transactions as soon as the block is validated. The efficiency of this finality boils down to having a democratic relationship in which block validation is proposed by a leader to an elected committee of validators. As soon as there’s sufficient support for the proposal from the committee, the block is validated and the transaction finalized. Leadership is chosen from a pool of validators, changing every epoch. Being validators themselves, any malicious action or acts that jeopardize the liveliness of the network may result in having their entire stake slashed. However, this finality can only tolerate a corruption threshold of ⅓.
But what if you can marry both layers together to eliminate uncertainties and nip any malicious action in the bud?
Welcome to Concordium.
Hand in Glove with Nakamoto
Concordium has developed a finality layer that can be bolted over any NSC blockchain, allowing the network to dynamically ‘checkpoint’ the blockchain through a Byzantine agreement that identifies and marks the blocks validated by honest users as final.
This two-layer approach provides the best of both worlds, ensuring that as long as corruption remains below ⅓, blocks are finalized as soon as they are validated.
However, when corruption is between ⅓ and ½, the network accedes to the finality of a pure NSC blockchain, to resolve the conflict.
Join #TeamConcordium in our different channels:
Twitter: https://twitter.com/ConcordiumNet
Telegram: https://t.me/concordium_official
Reddit: https://www.reddit.com/r/Concordium_Official
Developers Hub: https://developers.concordium.com/tech
Gitter: https://gitter.im/concordium_official/community
Discord: https://discord.gg/xWmQ5tp
Learn more: https://concordium.com
