Decentralized Transactional Trust

G. Grace

March 13, 2020

Abstract

Secure communications, over distance, enable our digital infrastructure and facilitate e-commerce activities on the internet.  Currently, online messages are securely transmitted by encryption: asymmetric cryptography.  Distributed ledger technology (DLT) employs asymmetric cryptography to securely facilitate transactions between arbitrary users in an adversarial environment: the internet.  Digital implementations of append-only ledgers, DLT can facilitate trustworthy transactions via network consensus.  A consensus protocol is a mathematical innovation which propagates user trust via shared-agreement across all DLT network nodes.  Furthermore, DLT systems utilize a record-keeping apparatus, or ledger type, to enable management of ledger data throughout the consensus process.  The combination of consensus, encryption, and ledger technologies can result in DLT platforms that enable disintermediation of costly trusted-third-party services.  The Bitcoin Network was instantiated in 2009 to provide a hedge against privacy-inept, centralized, third-party authorities’ systems through the creation of a secure and decentralized payment network.  The Bitcoin Network transacts value between users in the form of a cryptocurrency: bitcoin.  DLT networks, such as the Bitcoin Network, can provide ecommerce participants unprecedented strategic potential: boost internal value-driven capabilities emergent from DLT applications and mitigate the risks of trusted-third-party services through secure user-centric communication platforms powered by trustless technology.

Decentralized Transactional Trust

Secure Communications and DLT

        Economic participation in global commerce requires technological advancements which fuel societal activities. Networks of billions of devices and servers populate an invisible web of transaction pathways.  One user sends information to a recipient; the recipient typically sends a message back.  Data security is accomplished, in part, through cryptography: the science of applying mathematics to encrypt and decrypt data (Kessler 2016).  Encryption technology allows device users to securely share, store, and transact data.  Encryption mechanisms are used by DLT systems to ensure secure communications while operating in an assumed-adversarial environment: the world wide web.  

Despite environmental adversity, DLTs can enable transformational capabilities for system participants (users): trustfully exchanging, storing, and creating secure data with advanced networks (Rauchs et al. 2018).  Properly aligned DLT capabilities distill beneficial competitive advantages for the astute strategist, but such strategic considerations are beyond the scope of this paper.  Rather, the organizational technologies used to unlock said capabilities are discussed below, from a user-centric perspective.  

DLT systems (such as the Bitcoin Network) enable so-called “trustless” technology due to consensus: a shared-agreement propagation protocol.  A consensus protocol fosters the establishment of user trust through a single, shared transaction history (Bonifacio, Bouquet, and Traverso 2002).  At any point in time, the agreed-upon data recorded in the historical ledger is referred to as the present state.  The current state of the network is updated by nodes that reach consensus by following protocol rules.  (Node server operators are “distributed” across arbitrary geographic locations, hence the term distributed ledger technology).  A DLT’s consensus protocol is a layer of mathematical mechanisms nodes must follow to properly update to the current state of the network (Ferdous et al. 2020).  Consensus systems enable DLTs to propagate shared agreement of a single historical record; the recordkeeping archetype (or, “ledger” in distributed ledger technology) establishes the sequential data process all nodes must follow in order to participate on the record.

DLT systems require a record-keeping system to enable data management and to support transactional considerations.  Some DLT systems present compelling strategic possibilities for users to improve value-driven capabilities.  Others, like the Bitcoin Network, can mitigate centralized financial system risks; unrelated, unknown, untrustworthy users can securely agree upon an “official” transaction history in a decentralized system (Rauchs et al. 2018).  The Bitcoin Network is the first decentralized blockchain ledger to directly offer a natural economic counter to centralized money system risks (Nakamoto 2008).  

Emergent possibilities, such as improved decentralization capabilities, are derivative of the deliberate application of DLT’s capabilities toward solving real-world problems.  The disintermediation of a trusted-third-parties catalyzes the improvement of commercial and financial participation in a global network.  Encryption, consensus, and ledger technologies can form secure and capable DLT systems, built for both organizational and individual users, which enable economic and resource hedges against centralized system risks.

        The following article explores the three primary infrastructural aspects of DLT: encryption, consensus, and ledger technologies.  These interconnected technologies power DLT’s core capabilities; therefore, the goal of the article is to provide a strategic roadmap to bridge the gap between DLT’s transformative considerations and contemporary organizational core capabilities.  Additionally, the purpose of the article is to provide a clearer path to improved awareness of DLT’s potential role in the future of commerce for organizational users who may consider developing crypto capabilities.  Furthermore, a user-centric discussion of the Bitcoin Network is presented and future considerations, emergent opportunities, and DLT system concerns are explored.

Record-keeping Ledgers

        A DLT system’s primary function is to provide users the ability to transact and store data on an append-only distributed database in an adversarial environment (Kannengießer 2019).  The structure of the database is formed by two characteristics: 1. the network connections between node operators, and, 2. the transaction format or data structure (Ballandies, Dapp, and Pournaras 2022).  Database actions occur within chronologically-stored records called, entries: immutable user-initiated transaction groupings.  Not all DLT systems utilize the same entry type, however.  

The resultant dynamism emergent from entry-data structure, shared through node connectedness, forms the basis for the record-keeping system, or simply: the ledger.  The manner in which transactions within entries are processed results in the overall data structure.  Continuing the example of the Bitcoin Network: bitcoin uses a block entry-type and has a blockchain data structure.  However, a DLT’s data structure presents a trade-off between important network characteristics of security and transaction throughput (Kolb 2020).

        The data structure inherent to any particular DLT network contains various data types.  Meta-data (such as, timestamps), digital value representations (like, a token), and executable program code (e.g., smart contracts), as examples, may be contained within the entries.  Trade-offs occur due to the data structure and the replication of the entire journal/database across the network.  As an example, more nodes usually mean more security, but longer transaction times.  Regardless of the capability trade-off at the infrastructural level, DLT enables a replicated, distributed database and ensures user-transaction security via encryption.

Encryption

        DLT systems utilize cryptographic methods to disguise the message-sender’s data via encryption, then subsequently allow the recipient to decrypt (un-disguise) the data back into practical form (Kessler 2016).  Stored-data encryption versus transmitted-data encryption necessitates alternate cryptographic approaches.  Symmetric cryptography can be effective for securely storing data due to requiring both sender and recipient to possess the same private key; hence the term “symmetric”, both users need the same key.  

Alternately, asymmetric encryption is effective for securely disguising transmitted-data through the use of two different, but paired, cryptographic keys: a public key (known/shareable) and a private key (secret/not for sharing).  Each key is a value, a long number; one key encrypts the sender’s data while the other key allows decryption by only the intended recipient.  Cryptographic algorithms called, ciphers, are combined with key values to produce the message.  Called, ciphertext, encrypted messages are disguised and transactable data decodable only by the intended recipient. 

Asymmetric cryptography is also known as, public-key cryptography, which enables the security of transmitted messages between two or arbitrary parties.  The encryption type, asymmetric versus symmetric, requires different levels of disclosure for the users, as well as each’s ability to physically or otherwise securely share the private key.  Thus, communicating securely online necessitates asymmetric encryption via public-key cryptography.  

Furthermore, encryption strength can be measured by the size of the ciphertext as measured in bits (e.g., 256-bit encryption).  A 256-bit encrypted ciphertext is a number that is no more than 77 digits long.  Regardless of plaintext message input size a 256-bit hash function provides a fixed-length output message in the form of a 77-digit number.  Ciphertext size is held constant by a hash function in DLT systems.  Generated from the “hashing” of arbitrary message data previously combined with key values, hash functions output secure cyphertext, called a hash code or a message digest (Tasca, Paolo, and Tessone 2017).  

Attackers must spend irrational amounts of computer resources to attempt decoding a message digest due to the use of asymmetric cryptography.  The rules of mathematics allow only one specific data set to result in the exact message digest of any transaction or entry, so all inaccurate input data skews the output digest unpredictably.  The original message, encrypted with the sender’s private key, then hashed, is the only mathematical means to feasibly reach any specific message digest post-hoc.  The constant ciphertext length acts as an encoded digital proxy for any message, which enables trustworthy transmission: attempted discovery of the private key is economically favorable to hackers versus decoding the message digest.  

Asymmetric cryptography allows for the transmission of mathematically-disguised data using a private and public key pair.  Additionally, it is economically infeasible to deduce the private key from the public key (Kessler 2016).  The public key and a hash function encrypts data and the private key decrypts data.  In such manner, anyone with your public key can encrypt data with a common hash function plus your public key to send you a transmission, but only you can decrypt it with your private key because it is paired with your public key.  In practical application, large quantities of transactions occur consistently which necessitates the need for an “agreed-upon” state of transactions and entries.  A DLT’s consensus mechanism enables the growth of user trust in the network through the application of a shared-agreement system.  

Consensus

        A DLT system’s consensus protocol fosters the establishment of user trust through a single, shared transaction history; a concept called state machine replication (SMR) (Kolb 2020).  The premise of SMR is: a computing device can provide deterministic and pre-defined outputs from arbitrary inputs, and these inputs can compel a desired change-of-state.  DLT system nodes (i.e., network computing devices) share replicas of historical state-changes: transactions within entries.  

A DLTs consensus protocol establishes the rules of agreement on the network.  Rules of agreement within consensus protocols foster user trust in two manners: 1.  Establish one authoritative order/sequence of ledger transactions (the changes-of-state), and, 2.  Prevent the double-spending of money (Nakamoto 2008).  Effective consensus protocols propagate shared-agreement of a single transaction-history; system nodes process state changes, independently and sequentially, desirous of reaching the same state through the atomic broadcast (Ferdous 2020).

Any consensus protocol must ensure the atomic broadcast maintains full order, agreement, validity, and integrity of all valid messages for all applicable nodes.  Additionally, considerations are given to atomic broadcast timing: synchronous, asynchronous, and partially synchronous consensus algorithms maintain different message delivery and state-change time limits.  Furthermore, trustworthy consensus protocols must satisfy two primary conditions: 1. Liveness – the network must eventually deliver a sent message and cannot be “dead”; 2. Safety – a combination of proposed transmission validity and consistency of agreement amongst nodes (Ballandies, Dapp, and Pournaras 2022).  Satisfaction of the aforementioned conditions can result in distributed consensus; however, assumptions must be made regarding security.  

A consensus protocol must also provide fault tolerance against two types of failure: crash failure and Byzantine failure.  Crash failure results from greater than 50% of all nodes crashing; Byzantine failure results from greater than 33% of nodes behaving maliciously against the network’s “real” state.  The onus of the tolerance is derivative of network synchronicity (Tasca, Paolo, and Tessone 2017).

Lastly, a substantial distinction of consensus protocols is, incentivization.  Incentivized consensus algorithms reward or otherwise remunerate the node operator for processing entries while favorably meeting the two primary network conditions for users.  The Bitcoin Network employs an incentivized consensus protocol to reward node operators with satoshi’s, the name for the 100 million indivisible quantities comprising one bitcoin.  Non-incentivized consensus protocols do not issue rewards to node operators; such operators and systems are typically assumed to be trustworthy as all participants must be pre-authorized to use the system.  

The agreement rules of a consensus protocol are enabled by secure atomic broadcast and SMR in a distributed network.  Effective distributed consensus results in a live, safe, and fault tolerant DLT system.  Incentivized and non-incentivized consensus protocols are two broad categories of consensus; however, only incentivized DLT systems can enable a permissionless, decentralized, peer-to-peer, payment network, like the Bitcoin Network.          

Bitcoin

        The Bitcoin Network is the world’s first decentralized, peer-to-peer, permissionless, immutable blockchain payment network.  Launched in 2009 but presented to the world through academic white paper in 2008, the Bitcoin Network was invented by a pseudonymous author or group of authors: Satoshi Nakamoto (Popper 2015).  Currently, the true identification of Satoshi remains a mystery, but evidence suggests that an accomplished group of computer programmers called the Cypherpunks may be linked to the first-ever successful cryptocurrency.

        A founding member of the Cypherpunks published a manifesto in early 1993 outlining the importance of privacy for world wide web users.  “Privacy” means that users get to choose when to reveal themselves and how much about themselves to reveal while exchanging data online.  A private transaction allows a user to disclose only the information that the other party needs to complete the transaction, but nothing more.  User transactions on a DLT system, like the Bitcoin Network, results in a historical “memory” of the transaction for each user; furthermore, the entry details cannot be altered or deleted post transaction (immutability).  Thus, the deliberate socialization of historical ledger entries results in potential privacy risks for transactors.  

The Cypherpunks acknowledged that private transactions and privacy enablement must be sacrosanct and defended; the development of open-source computer code along with ancillary tools and systems became guiding principles of the group (Hughes 1993).  They championed a network communication model called a “web of trust”, used to enable a decentralized certificate authority for cryptographic keys.  Cryptography and the encryption of transaction data drove the Cypherpunks’ goals toward enabling private online transactions.  

Over 15 years after the Cypherpunks published A Cypherpunk’s Manifesto, the world’s first cryptocurrency was created.  The Bitcoin Network went live in January of 2009 when the genesis block was mined.  Satoshi Nakamoto posted the following quote in the block (Di Salvo 2021): “The Times 03/Jan/2009 Chancellor on brink of second bailout for banks.”

Satoshi posited that cryptography could replace trusted-third-parties on a decentralized network, thus resulting in improvements to privacy, transaction dynamism, and security for users (Nakamoto 2008).  Encryption enables users to transact pseudonymously: a user’s public address is not private but also does not directly reveal the user.  A chronological, timestamped, proofing mechanism is used to prevent double-spending of funds, as well.  Termed, proof-of-work, the Bitcoin Network employs a proof-of-work (PoW) validating mechanism as part of the consensus protocol.

The proof-of-work process used by the Bitcoin Network’s node operators to enable so-called, trustless technology, functions similarly to cryptographical web-of-trust models for online digital certificate issuers (Bertelloni 1998).  A socio-technological concept, a web-of-trust network model enables trust by social agreement (Caronni 2000): new node operators choose other node operators to trust, based on their social network, and connect to the network through trusted transactions.  The Bitcoin Network operates similarly because the PoW mechanism requires all nodes to agree about the validity of each new block added to the blockchain.

Agreement in a web-of-trust type network is built through known social affiliation; agreement in a PoW-type network is built on solving a cryptographic puzzle (Herlihy 2019).  Contrary to some opinions, the Bitcoin Network does not arbitrarily burn electricity, rather, the proofing mechanism chosen by Satoshi Nakamoto requires the Bitcoin Network node operators to use computer clock cycles to solve the puzzle.  And because computers can perform faster clock cycles only by performing faster hashing calculations, the Bitcoin Network requires node operators to “spend” energy attempting to solve the puzzle by finding a nonce.  

Defined as “an unknown value”, a nonce is a number that can only be discovered by guessing.  The PoW mechanism requires the miners (node operators) to solve for an unknown number, and because of the one-way functionality of the hashing algorithm, all miners have to start at 0, run the hashing algorithm, check if the resultant hash equals the correct hash code, and if not, try a different input.  The hashing and comparing process occurs by all live miners, until a miner discovers the correct nonce.  

Because the miner discovered the correct nonce, they can gather a group of the most profitable transactions waiting in their mem-pool to make a block entry, with one caveat: the miner can include a deposit of bitcoin (satoshi’s) to their own wallet for a block reward.  Miners can therefore gain a block reward for nonce discovery and the associated fees paid by users’ transactions included in the block entry (Nakamoto 2008).  These remunerations are incentives which reward honest node operators by requiring block-reward recipients to maintain consensus via blockchain.

Each block comprising the blockchain on the Bitcoin Network is represented as a hash code.  Each subsequent block entry added to the “chain” of blocks uses the previous block’s hash code as input for nonce discovery.  All blocks are chained in this way, thus, any modification to any transaction data from the past would result in incorrect hash codes.  The rules of the hash function prevent the malicious miner from adding an incompliant block to the chain via replicated blockchain entries disseminated via atomic broadcast.  The Bitcoin Network prevents amending past block data via consensus of the blockchain: all other nodes must change ledgers to match the offending ledger, as well as all subsequent hash codes.  Hence, the Bitcoin Network is tolerant to 51% of nodes behaving maliciously identical.

Conclusion

        Deliberately-applied combinations of distributed ledger, encryption, and consensus systems enable secure digital transactions and trustworthy communications between arbitrary users on DLT systems.  DLT systems can be assembled to perform specific user-centric functions in the context of real-world commerce and personal finance.  Some DLT systems, like the Bitcoin Network, leverage various technologies and ancillary protocols to provide additional platform layers of secure decentralized transactional trust networks on the internet.    

        The establishment of decentralized and incentivized networks affords the potential for new transaction models, governance arrangements, and organizational structure.  Trustless technologies and newfound capabilities to establish trust on global networks enable the revisioning of legacy transaction networks and information management systems.  It is the opinion of this author that contemporary organizations seeking long-term going-concern relevance, should strategically assess the macroeconomic environment facing traditional financial systems and their organization’s role within.  The ability to externalize transaction cost and the potential to solve myriad problems requires thoughtful consideration of DLT adoption for the modern firm.  

New risk mitigation tactics emerge by applying crypto capabilities toward solving real-world problems.  DLT systems also present decision-makers the potential to mitigate the inherent risks of our centralized financial system.  Democratic economies that operate exclusively in a single-currency environment expose themselves to dangerous consequences of benign policy decisions meant to “generate” wealth through money-printing.  However, new strategic business practices aimed at hedging risks associated with retaining economic value exclusively within a centralized system can be achieved through trustless technology implementations.

Against the backdrop of sustainability, the hot-button topics of DLT system operations, like bitcoin mining, are not in and of themselves unsustainable nor bad for the environment, per se: it is just the requirement of the proofing mechanism for the miner to compete with computer clock cycles against all other miners.  It is up to the miner to establish a renewable energy source for the node.  

Any extant payment network requires power, but not all quantify the amount of power securing the network like the Bitcoin Network.  Consider: the energy consumption of the Bitcoin Network is the measure of cryptographic strength; realize: satoshi’s transact on the most powerfully secure payment network in the world.  The Bitcoin Network is a functional example of technologically-enabled decentralized transactional trust.

In the final analysis, the combination of consensus, encryption, and ledger technologies can result in the disintermediation of traditional third-party services and platforms.  The inter-relatedness of distributed consensus networks and programmable trust models through a ledger’s data structure and consensus mechanism introduces organizational management opportunities.  DLT systems, like the Bitcoin Network, provide a hedge against privacy-inept, centralized transaction systems.  The ability for modern organizations to establish a centralization hedge is not well-known, but perhaps it should be.

References

Ballandies, Mark C., Marcus M. Dapp, and Evangelos Pournaras. "Decrypting distributed ledger design—taxonomy, classification and blockchain community evaluation." Cluster Computing 25, no. 3 (2022): 1817-1838.

Bertelloni, Maud Barret. "The Cypherpunk vision of techno-politics." St. Anne’s Academic Review (1998): 1.

Bonifacio, Matteo, Paolo Bouquet, and Paolo Traverso. "Enabling distributed knowledge management: Managerial and technological implications." (2002).

Caronni, Germano. "Walking the web of trust." In Proceedings IEEE 9th International Workshops on Enabling Technologies: Infrastructure for Collaborative Enterprises (WET ICE 2000), pp. 153-158. IEEE, 2000.

Di Salvo, Mat. “Coinbase Encodes Secret Message on Bitcoin Blockchain on Day of Public Listing.” Decrypt.co, April 14, 2021. https://decrypt.co/66502/coinbase-secret-message-bitcoin-public-listing

Ferdous, Md Sadek, Mohammad Jabed Morshed Chowdhury, Mohammad A. Hoque, and Alan Colman. "Blockchain consensus algorithms: A survey." arXiv preprint arXiv:2001.07091 (2020).

Herlihy, Maurice. "Blockchains from a distributed computing perspective." Communications of the ACM 62, no. 2 (2019): 78-85.

Hughes, Eric. "A cypherpunk’s manifesto." URL (accessed 3 August 2004): http://www. activism. net/cypherpunk/manifesto. html (1993).

Kannengießer, Niclas, Sebastian Lins, Tobias Dehling, and Ali Sunyaev. "Mind the gap: trade-offs between Distributed Ledger Technology characteristics." arXiv preprint arXiv:1906.00861 (2019).

Kessler, Gary C. "An Overview of Cryptography (Updated Version, 3 March 2016)." (2016).

Kolb, John, Moustafa AbdelBaky, Randy H. Katz, and David E. Culler. "Core concepts, challenges, and future directions in blockchain: A centralized tutorial." ACM Computing Surveys (CSUR) 53, no. 1 (2020): 1-39.

Nakamoto, Satoshi. "Bitcoin whitepaper." URL: https://bitcoin. org/bitcoin. pdf-(: 17.07. 2019) (2008).

Popper, Nathaniel. "Decoding the enigma of Satoshi Nakamoto and the birth of Bitcoin." New York Times 15 (2015).

Rauchs, Michel, Andrew Glidden, Brian Gordon, Gina C. Pieters, Martino Recanatini, François Rostand, Kathryn Vagneur, and Bryan Zheng Zhang. "Distributed ledger technology systems: A conceptual framework." Available at SSRN 3230013 (2018).

Tasca, Paolo, and Claudio J. Tessone. "Taxonomy of blockchain technologies. Principles of identification and classification." arXiv preprint arXiv:1708.04872 (2017).