The Blockchain PDF

SMU Classification: Restricted Emerging Technologies and Law The Blockchain  SMU Classification: Restricted In this video… The next big thing (banks HATE this) Understanding the blockchain and its implications  SMU Classification: Restri...

SMU Classification: Restricted Emerging Technologies and Law The Blockchain  SMU Classification: Restricted In this video… The next big thing (banks HATE this) Understanding the blockchain and its implications  SMU Classification: Restricted What is Blockchain?  SMU Classification: Restricted The OG Definition “a solution to the double-spending problem using a peer-to-peer distributed timestamp server to generate computational proof of the chronological order of transactions”  SMU Classification: Restricted Double Spend Problem I want a house! Here’s $2m in digital fiat currency! Cool, your digital balance says you have $2m. Ok here’s your house. Seller A SMU Classification: Restricted  SMU Classification: Restricted Double Spend Problem I want a house! Here’s $2m in digital fiat currency! Cool, your digital balance says you have $2m. Ok, here’s your house! Seller B  SMU Classification: Restricted The Solution How do sellers know you have what you purport to give? Physical cash: proof is in possession Digital currency?  SMU Classification: Restricted The Solution Buyer has $2m with me on escrow DigiBank I want a house! Cool, DigiBank says you Here’s $2m in digital have $2m. fiat currency! Ok here’s your house. Seller B  SMU Classification: Restricted The Solution Problem unique to online transactions (and other transactions based on records, even if not electronic) Traditional solution is trusted intermediaries (aka platforms) Why do we trust banks? Laws, enforcement  trust in government Business reputation, profitability  trust in markets Social capital  trust in norms Now enter blockchain  trust in code  SMU Classification: Restricted “Peer-to-peer distributed timestamp server” B has $0m B has $0m B has $0m $2m B has $0m B has $0m B has $0m  SMU Classification: Restricted “Computational Proof of Transactions” Current balances: Buyer B: 2 Seller A: 0 B has 0 2 A has 2 0 B transfers 2 to A B has 0 2 Signed, B A has 2 0 2 B has 0 0 A has 2  SMU Classification: Restricted Tampering Attempt 1: Mis-repeating B has 0 A has 2 B transfers 2 to A Signed, B B has 02 A has 20  SMU Classification: Restricted Tampering Attempt 2: Re-writing Records B has 02 A has 20 B transferred 0 to S B has 0 2 Signed, B A has 2 0 2 B has 0 0 A has 2  SMU Classification: Restricted How to Overcome? Attack vectors: assume the most dangerous situation (where the attacker is the original transactee) Markets won’t work: B has every incentive to ‘undo’ transaction after spending money Law not efficient, too slow No clear social norms So back to code: consensus mechanisms that enforce majority agreement at the point of accepting transactions  SMU Classification: Restricted Cryptographic Hash (#) Functions SHA (or Secure Hash Function) 256 is one (of many) algorithms available for encrypting information Encryption is achieved by converting or (“hashing”) human readable information, like plain English, to some gibberish. For instance: SHA256(“blockchain”)  ef7797e13d3a75526946a3bcf00daec9fc9c9c4d51ddc7cc5df888f74dd434d1 For the encryption to be secure, it is essential that given the RHS output, you cannot easily figure out the what the original LHS output was However, you can guess what the LHS was, and see if SHA([your guess]) produces the same output. If so, you have “decrypted” the info. As we will see later, however, the blockchain does not rely on this bit to encrypt data, because all the inputs are transactions that should be on public record For SHA256 specifically, the output is always 256 characters regardless of input length  SMU Classification: Restricted POW Consensus and Tamper-Proofing B transfer 2 to S Signed, B Block ID 1233 Block ID 1234 Block ID 1235 Date: 28-09-2021 Date: 28-09-2021 Time: 1600:30:30 Time: 1605:00:00 Transactions: Transactions: - B-2, S+2 - … … - Y-3, Z+3 - … Previous block #: Previous block #: abc4032 cdef1234 — X? X: 123 X? Proof-of-work: Find X so that SHA256(“2809202116003030—B-2,S+2…abc4032—X”) starts with “00000” E.G. SHA256(“blockchain”) = ef7797e13d3a75526946a3bcf00daec9fc9c9c4d51ddc7cc5df888f74dd434d1 All blocks are publicly-distributed, so anyone can try to solve (or “mine”). Suppose X is found to be 123. Block 1234’s # = SHA256(“…123”). Say this is cdef1234 for simplicity.  SMU Classification: Restricted POW Consensus and Tamper-Proofing B transferred 0 to S Signed, B Block ID 1233 Block ID 1234 Block ID 1235 Date: 28-09-2021 Date: 28-09-2021 Time: 1600:30:30 Time: 1605:00:00 Transactions: Transactions: - B-0, S+0 - … … - Y-3, Z+3 - … Previous block #: Previous block #: abc4032 Not cdef1234 X? X? Suppose attacker wants to re-write block 1234 as such. They must: 1. Find new X for block 1234 so that SHA(“…B-0,S+0…”) starts with “00000”. 2. If they succeed (say new X is 321), SHA(“…321”) is different too. 3. So must find new X for block 1235 also  SMU Classification: Restricted POW Consensus and Tamper-Proofing B transferred 0 to S Signed, B Block ID 1234 Block ID 1235 Block ID 1236 Block ID 1237 Date: 28-09-2021 Date: 28-09-2021 Date: 28-09-2021 Date: 28-09-2021 Time: 1600:30:30 Time: 1605:00:00 Time: 1610:00:00 Time: 1620:00:00 Transactions: Transactions: Transactions: Transactions: - B-0, S+0 - … - … - … - Y-3, Z+3 - … Previous block #: Previous block #: Previous block #: Previous block #: abc4032 Not cdef1234 X? X? X? X? While solving 1235, network adds 1236, 1237, etc, Attacker must present longest blockchain to win consensus Sets up a race between honest and dishonest nodes By naïve (Poisson) probability model, can show that chance of attacker winning diminishes exponentially as (a) more nodes added, (b) honest/attacker nodes compute faster/slower  SMU Classification: Restricted The Actual Bitcoin Chain Try and spot some of the terms and concepts just covered. Of course, we haven’t covered everything. https://www.blockchain.com/explorer/blocks/btc/804244  SMU Classification: Restricted Other Consensus Mechanisms (optional) Proof-of-stake To mine, must put up deposit The more deposit, the more ‘compute’ in a POW-like sense Slashing conditions (deposit forfeiture) activate if you try to be funny Code-enforced economic disincentives for security BFT, Hashgraphs Complicated structure of broadcasts/re-broadcasts to ensure ‘right’ record is found despite attackers Relies on attacker not controlling too many nodes See Computer Law & Security Review article  SMU Classification: Restricted Other Blockchain Features Efficient ways of identification/authentication/retrieval, etc Through private/public key cryptography that’s a separate system from Proof-of-Work and which exists for lots of non-blockchain tech Private vs Public blockchains No theoretical limit  SMU Classification: Restricted Takeaways Blocks are nothing special. Just packets of data Hash chaining essential to tamper-resistance Only probabilistic resistance. Attacker can still win if: Very few blocks added (slow bit rate) Puzzle too easy to solve Honest nodes have less compute (50+% attack) Dumb luck POW, POS, etc are necessary for so-called computational proof of transactions, but are not entirely bulletproof Horrendous waste but good alternative means elusive (like capitalism) Bitcoin’s blockchain has not yet been hacked Other blockchains have (e.g. Ethereum classic’s 50% attack) Ethereum classic was a POW system, though it was thin)  SMU Classification: Restricted Larger Uses of Blockchain Databases Bitcoin blockchain used to record only transactions of bitcoin But idea of blockchain as a way to store data securely without trusted 3P is more general What else might be stored on a blockchain? Text describing legal or economic rights  ICOs, “tokens” Links to real world assets  Off-chain assets/”tokenisation” Property records  Land register Links to (digital) art  NFT Computer code itself  smart contracts Thousands of sub-cryptocurrencies are implemented as code on Ethereum blockchain  SMU Classification: Restricted Legal Characterisations of Cryptoassets Is Bitcoin (and other cryptocurrencies and derivative assets property)? Note: Bitcoin is just one ‘app’ built using a (not the) blockchain Bybit v Ho Kai Xin and others SGHC 199 Cryptoassets are “not classed as physical assets” but “do manifest themselves in the physical world, albeit in a way that humans are unable to perceive”. Private/public key locking/unlock system appears to be said “physical manifestation at the level of digital bits and bytes”. While not permanent, we can “give a name to a river even though the water contained within its banks is constantly changing” “This description of crypto assets shows that they can be defined and identified by modern humans, such that they can be traded and valued as holdings.” They meet the Ainsworth formula.  SMU Classification: Restricted Bybit v Ho Kai Xin and others SGHC 199 [34-35] Are cryptoassets things in possession or in action (given that all personal property are either one or the other)? My conclusion is therefore that the holder of a crypto asset has in principle an incorporeal right of property recognisable by the common law as a thing in action and so enforceable in court. While it might be said that this conclusion has an element of circularity in that it could also be said that the right to enforce in court is what makes it a thing in action, this type of reasoning is not strikingly different from how the law approaches other social constructs, such as money. It is only because people generally accept the exchange value of shells or beads or differently printed paper notes that they become currency. Money is accepted by virtue of a collective act of mutual faith. What is the court’s role here? Is it recognizing the broader societal act of mutual faith in crypto? Or, is it creating one? Is this an issue with law, tech, or both?  SMU Classification: Restricted In this video… The next big thing (banks HATE this) Understanding the blockchain and its implications