Blockchain Tech Big PictureCampbell R. HarveyDuke University and NBER
FUQINTRD 697: Innovation and Cryptoventures
Campbell R. Harvey 2019
Tech Big Picture
Key ingredients• Hashing function: provide the chain (the entire last block as the header
of the next). Bitcoin SHA-256; Ethereum Keccak-256 (SHA-3). One way cryptographic function (infeasible to go the other way)
• We will do demonstrations on the main hashing algorithms• Optional Wednesday lunch session on mechanics of SHA-1
Campbell R. Harvey 2019 3
Tech Big Picture
Key ingredients• Private keys/Public keys: Private key is just a
random number. The public key is mathematically linked to the private key. It is easy to go from the private key to the public key – but very difficult to go from public to private. Current technology uses Elliptic Curve Cryptography (ECC)
• We will spend a session on how these work with simple examples
Campbell R. Harvey 2019 4
Tech Big Picture
Key ingredients• Public Address: This is either identical to the public key (e.g., Ethereum)
or a function of the public key (e.g., Bitcoin). • We will detail how bitcoin public addresses are derived
Campbell R. Harvey 2019 5
Tech Big Picture
Key ingredients• Digital Signatures: When doing a
transaction, you “sign over” your cryptocurrency to someone else using a Digital Signature Algorithm (DSA). The signature proves that you are the owner of the private key. Anyone observing the signature and the public key can verify that you have the private key (without revealing the private key).
• We will spend a session on DSA pointing out future vulnerabilities
Campbell R. Harvey 2019 6
Tech Big Picture
Key ingredients• Transaction mechanics: For many cryptocurrencies like bitcoin, we deal
with unspent transaction outputs (UTXOs). If I have an UTXO of say 10 units and I want to send 7 to Jenna, Jenna generates a private key (and a public key). I generate a (potentially) new private key (and public key). In a single transaction, I sign over 7 to Jenna and 3 to myself (think of this as “change”). I have a new UTXO of 3 units. The old one resides in a blockchain but has no value. Ethereum uses a different system of account balances.
• Cryptocurrency doesn’t move anywhere. Everything remains on the associated blockchain.
Campbell R. Harvey 2019 7
Tech Big Picture
Key ingredients• Cryptography: This is widely used in all aspects of blockchain. It is
particularly important in deriving the public key and the DSA. We will also see that other types of cryptography are used in certain blockchainapplications (for example, you might find it useful to have a contract codified in a blockchain – but you only want it visible to the contracting parties.
• Introduction to Cryptology session will introduce some simple ciphers
Campbell R. Harvey 2019 8
Tech Big Picture
Key ingredients• Consensus Mechanisms: Consensus is
the mechanism by which nodes agree on both the historical blockchain as well as the new additions to the historical blockchain.
• We will initially talk about Proof of Work (PoW) which is currently used in Bitcoin and Ethereum. We will also talk about Proof of Stake (PoS) and some other mechanisms.
Campbell R. Harvey 2019 9
Tech Big Picture
Key ingredients• Incentives: In Proof-of-Work systems, miners are incentivized to verify
transactions and to do the work that allows new blocks. However, there is no incentive to run a node.
• It is possible to introduce incentives for Level 2 systems (Lightning Network for bitcoin)
• Some newer technologies incentivize nodes
Campbell R. Harvey 2019 10
Tech Big Picture
Key questions• Latency: Blockchains have complete redundancy. How can blockchains
compete with Visa with 24,000 transactions per second – and go beyond as blockchain applications grow?
• Transactions are but one level – what happens when smart contracts become more mainstream
Campbell R. Harvey 2019 11
Tech Big Picture
Key questions• Data: Blockchains are stored on multiple nodes. If large amounts of data
are associated with transactions, it may be infeasible to store petabytes of data on each node.
• Possible for blockchains to store summaries of data rather than original data
• Cost of storage declining exponentially
Campbell R. Harvey 2019 12
1981 - $300,000 1987 - $50,000 1990 – $10,000 1994 - $1,000 1997 - $100 2000 - $10 2004 - $1 2010 - $0.1 2018 - $0.01
Cost per GB
Tech Big Picture
Key questions• Connection: Are we headed to a world of millions if not billions of
blockchains – or will there be one Masterchain. • Sidechains will be studied in the course
Campbell R. Harvey 2019 13
Tech Big Picture
Key questions• Privacy: Delicate balance needs to be achieved. No one has worked it out
yet.• This is not just a blockchain issue!
14
Tech Big Picture
Key questions• Real world verification: How do we verify events
in the real world trustlessly on a digital and siloedblockchain?
• Much work needs to be done on oracles (software, hardware, inbound, outbound, and consensus based). How do we trust the oracle?
• RFID tags?
Campbell R. Harvey 2019 15https://blockchainhub.net/blockchain-oracles/
Tech Big Picture
Key questions• Immutability: This is a crucial characteristic of a blockchain. What are the
guarantees on my transaction?• Block confirmations• The issue of forks
• Bitcoin Cash 1 Aug 2017• Bitcoin Gold 24 Oct 2017• Bitcoin SV 15 Nov 2018
Campbell R. Harvey 2019 16https://hackernoon.com/blockchain-forks-b0dca84db0b0 https://www.investopedia.com/tech/history-bitcoin-hard-forks/
Tech Big Picture
Key questions• Governance: We are only now thinking about decentralized governance.
This is very much related to consensus mechanisms – but broader in that the computer programs that generate current blockchains will surely need improvement and enhancements in the future. Who will do this? (BIP/EIP for example.)
• Much research on consensus mechanisms.• What will the world look like with potentially millions
of DAOs?
Campbell R. Harvey 2019 17
Tech Big Picture
Key questions• Use Case: Blockchain is not the panacea for all business applications. In
addition, even if a business application is suited for blockchaintechnology – there are a myriad of potential blockchain choices.
• Technology needs to match the needs of the application
Campbell R. Harvey 2019 18
Tech Big Picture
Key questions• Regulation: Regulators are at a disadvantage because blockchain is not
currently part of the law. In addition, there is a steep learning curve for regulators
• How do we classify tokens?• Regulatory risk may cause
innovation to move off shore to jurisdictions that are less strict
Campbell R. Harvey 2019 19
Tech Big Picture
Key questions• AI: Is there a nexus between AI and blockchain?
• Think of Federated Learning being moderated by Smart Contracts.
• Singularity NET; DeepBrain Chain; Namahe;
Campbell R. Harvey 2019 20https://blockonomi.com/ai-blockchain/