what is the distinction between a blockchain and a ...blockchain has a shared and replicated ledger...

Blockchain Byte

The Blockchain Byte features a question from the distributed ledger space

R3 ResearchEmily Rutland

For more Blockchain Byte posts, R3 members visit:https://r3-cev.atlassian.net/wiki/x/HgEwAw

What is the distinction between a blockchain and a distributed ledger?

While often used interchangeably, a blockchain and adistributed ledger are distinct – though subtly different –technologies.

A distributed ledger is a record of consensus with acryptographic audit trail which is maintained andvalidated by several separate nodes*. Thiscryptographically assured and synchronized data can bespread across multiple institutions. Distributed ledgers canbe either decentralized, granting equal rights within theprotocol to all participants or centralized, designatingcertain users particular rights. Actors typically employdistributed ledgers when they need a tool which permitsthe concurrent editing of a shared state while maintainingits unicity. The ledger’s state is determined through aconsensus algorithm, which can vary in its mechanics butultimately serves to validate information from inputs to thenetwork.

Blockchain has a shared and replicated ledger comprised of information stored in “blocks” and sits below a distributed ledger and acts as a way to verify transactions submitted by producing a new “block” to the chain.

Distributed ledger is a record of consensus with cryptographic audit trail maintained and validated by nodes. It can be decentralized or centralized.

2


Often described as the technology thatunderpins the Bitcoin network, ablockchain data structure has a shared,replicated ledger comprised of digitallyrecorded and unchangeable data inpackages called blocks. A blockchain sitsbelow a distributed ledger and acts as away to order and validate the transactionsin the ledger. While a blockchainautomatically produces a new “block”after certain predetermined criteria is met,a distributed ledger only verifies atransaction once it is submitted, asopposed to pushing out empty blocks. Ablockchain is a way to implement adistributed ledger, but not all distributedledgers necessarily employ blockchains.

*In a distributed ledger, it is not necessarilythe case that all nodes either receive allthe information or, if they do, canunderstand it. In the Ethereum model, forexample, all nodes receive andunderstand all the information. In Corda,only the nodes involved are aware that atransaction exists.

What are Blocks and Why Aren’t They Necessary?

3



A block structure allows for global broadcast of data andlarge amounts of unrelated transactions to be confirmedat once. While blocks make sense for public blockchainsystems like Bitcoin, this logic does not apply to trusteddistributed ledger networks used within financialinstitutions, like Corda.

A “block” is a bundle of transaction data*. Most closelyassociated with Bitcoin, a blockchain structure connectsblocks such that each block includes the hash value of theprevious block, thereby forming a chain of validtransactions.

“Miners” create new blocks by validating transactionsthrough a Proof of Work (PoW) exercise. This PoW exerciseallows anyone (without permission) to create valid blocks ifthey can expend the required computing power(calculating hashes). This resource-intensive process alsocreates a financial barrier to deter malicious actors fromadding fraudulent blocks.

Proof-of-Work groups transactions into blocks and broadcasts it to unrelated parties.

Therefore, blocks are not suitable for use in a trusted distributed ledger network between financial institutions.

2


This PoW system deliberately takes timeboth to prevent arbitrary coin generation(which would lead to hyperinflation of thecurrency) and to prevent too manysolutions at a given time (once a minerfinds a solution, it must propagate aroundthe network, which takes time and maylead to collisions and forks). In Bitcoin, thisinterval is an average of 10 minutes. Toaccommodate for this delay, transactionsmust be batched together and confirmedin blocks.

In trusted, private distributed ledgernetworks, PoW and global broadcast oftransactions are not necessary - it does notmake sense to group unrelatedtransactions into blocks that are thenshared with many, unrelated parties. Thoseprivate blockchain systems that do useblocks most likely use the Bitcoin structurewithout considering why Bitcoin works theway it does.

Corda is different. Recognizing that theblock structure is a flawed design decisionfor financial services, R3 deliberately chosenot to use blocks for Corda and itsfinancial services clients


3


What is the difference between a centralized and decentralized blockchain?

A blockchain can be either centralized or decentralized. Itis important, however, that decentralized not be confusedwith distributed. While a blockchain is inherently distributed(meaning that many parties hold copies of the ledger), it isnot inherently decentralized.

Whether a blockchain is centralized or decentralizedsimply refers to the rights of participants on the ledger, andis therefore a question of design.

In a decentralized network, anyone can participate andtransact on the ledger. As a result, mechanisms must existin order to combat the vulnerabilities that arise from thisdesign and to ensure that transactions are correct. Bitcoin,for example, is a decentralized blockchain that usesmining and proof-of-work* to maintain the integrity of theledger and to prevent people from corrupting the system.

A centralized network, on the other hand, is made up ofparties whose identities are known. Thus, the system is validbecause only credible and reputable participants canpost to the ledger. Because participants’ identities areknown, their transactions can therefore be audited.

In a decentralized network, anyone can transact on the ledger. Bitcoin’s network uses mining and proof-of-work to maintain the integrity of the ledger.

In a centralized network, only known and identified parties can transact on the ledger. Therefore, their transactions can be audited.

2


Ultimately, a centralized distributed ledgershould be used in any highly regulatedindustry - such as financial services - tominimize vulnerability. While bothdecentralized and centralized blockchainshave respective risks, a centralized networkis much preferred for our purposes in thatthe identity of participants is known and, asa result, an audit trail exists should an actorattempt to corrupt the system. It ispreferable, for example, to fine a knownentity that entered an invalid transactionrather than attempt to account for everypossible manipulation that couldpotentially occur in an anonymous,decentralized network.

*Proof-of-work: a function to deter denialof service attacks and other potentialspam on a network by requiring some levelof work that is usually costly and time-consuming. In the Bitcoin network, minersmust complete a proof-of-work in order toregulate the rate at which new blocks aregenerated.

What is the difference between a centralized and decentralized blockchain?

3


What is Central Bank Digital Currency?

The concept of digital currency precedes distributedledger technology. It began as a way to distinguishbetween Internet versions of digital money (i.e. DigiCash,eCash) and electronic cash that existed on physicalplatforms (i.e. Chipknip and Mondex) and grew morecomplex with the advent of web platforms that providedaccess to accounting (i.e. Paypal). These examples are allprivate market currency solutions that do not involve acentral bank.

Though the concept of a Central Bank Digital Currency(CBDC) remains largely theoretical, the evolution of newtechnologies such as DLT is increasing the feasibility ofputting a CBDC into practice. At a high level, a CBDC is adigital store of value (money) and method of exchangeissued by a central bank. Theoretically, a CBDC introducesa new digital mechanism for real-time settlement betweenindividuals.

Central Bank Digital Currency is a digitally stored currency that theoretically can achieve real time settlement for individuals and banks.

CBDCs are intended to be issued by the central bank and able to exchange for fiat currency. It can also be an alternative to current market currency.

2


CBDCs are intended to be 1:1exchangeable with other forms of money(such as notes, coins and deposits atbanks). They may be issued in an alternateform exchangeable for fiat currency that isheld on deposit by a central bank andpayable on demand to the owner. CBDCscan also be issued as a new form ofmoney supply in addition to traditionalcentral bank money issuance.

One of the main purposes of a CBDC is tobroaden access to central bank liabilities(such as notes and coins) in digital form,thus increasing the ease and conveniencerequired to hold and pay these liabilities. Inaddition to broadening this access, aCBDC system must also be designed to bepractically functional (for example, itcannot only be accessible throughproprietary networks such as SWIFT orFedwire).

What is Central Bank Digital Currency?

3


What is a Digital Signature?

A digital signature is a method to verify the authenticityand integrity of a digital message.

Digital signatures employ asymmetric (public key)cryptography, a cryptographic system that uses a pair ofkeys – a public key, which may be shared widely and aprivate key, which is known only to the owner – to encryptand decrypt messages.

A digital signature ensures the authentication and integrityof data. Authentication of data refers to the verification ofthe source of the data, meaning that the message wasindeed sent by the person or entity who claims to be thesender. This is ensured because the digital signature uses aprivate key that is known only to one specific user.

The integrity of data refers to the content of the messageitself, meaning that the message was not altered in transit.A digital signature ensures this because, if a message isdigitally signed, any change in the message after it hasbeen signed invalidates the signature.

A digital signature also provides non-repudiation, meaningthat the signer cannot later claim that he/she did not signthe message.

Digital signature authenticates the integrity of data by using asymmetric cryptography.

Digital signature is nearly impossible to counterfeit without access to the sender’s private key.

2


A digital signature is created by encryptingthe hash of a document using the privatekey. A recipient can use the public key todecrypt the signature and ensure that theresult indeed matches the hash of thedocument. If a document changes aftersigning, the digital signature is invalidatedbecause the document's hash will notmatch the decrypted signature.

Unlike a handwritten signature, a digitalsignature is nearly impossible to counterfeit.A fraudulent party cannot fake a validsignature without access to the secret,private key, held only by the sender.Furthermore, unlike a handwritten sign-ature, a digital signature not only verifiesthe identity of the signer but also validatesthe contents of the message itself.

What is a Digital Signature?

3


What is a Golden Copy?

A golden copy – or golden record – typically serves as theofficial, master version of a record of data. It is anauthoritative single source, sometimes referred to as the“single version of the truth”, where “truth” is understood asthat which users can ensure is the single correct piece ofinformation which can then be actioned upon.

What constitutes a golden copy varies based on theparticular use of that data. For example, a passport orsocial security number may constitute a golden copy inthe case of customer KYC information, while a driver’slicense can be thought of as golden copy of a form ofidentity at the DMV.

For financial institutions, there is a reliance on a singlesource of data that is untainted, “true” and representssomething that is actionable. In trading, for example, thismeans that the initial booking of a trade and agreementof the attributes of that trade between counter partiesforms a representation that can be used for some action.

A golden copy is a master version of a record of data and referred to as the the “truth” that users can be sure that it is true.

In order to make changes to the “truth”, changes must be validated by participants that has the power to make and validate.

2


In other words, the initial construct of agolden copy consists of the data attributes(of a trade contract information that isboth common like reference data andcounter party specific, like a trader ID) thatdefine the data and confirmation/vali-dation that makes this golden copyreliable and, ultimately, actionable.

While existing processes - like reconciliation- currently carry out this validation processin a bank, distributed ledger technology(DLT) may ultimately be able to carry outthese actions far more efficiently. DLTenhances the notion of a golden copy inthat it enables a group of participants toconfirm that data is true, lowering the costof reconciliation and relieving the burdenof having a single institution validate or actas a golden copy. DLT allows for multiple,distributed confirmations, strengtheningthe claim of factuality.

A golden copy is considered the original,incepting transaction – any amendmentsor changes to the data must be confirmedand validated by participants in thedistributed ledger network that have theauthority to make and validate thosechanges. In this sense, a golden copy is aliving piece of data that is continuouslyvalidated.

What is a Golden Copy?

3


What is Immutability?

Immutability is a frequently used buzzword when discussingblockchain and distributed ledger technology, and theperception of a blockchain as a permanent record ofinformation is one of the greatest benefits of thetechnology.

The term “immutability” refers to a state that cannot bechanged or modified after it has been created. (On theother hand, a ‘mutable’ object can be changed after it iscreated.) As it pertains to blockchain technology,immutability implies that data written to a blockchaincannot be edited, even by a system administrator. This isbeneficial, especially in financial transactions, in that onecan be sure that sent and received data and transactionscannot be altered.

For Bitcoin, immutability of transactional information isachieved through the process of creating a hash itself,which is one-way. Additionally, the process of hashingbundles transaction outputs into blocks which contain thehash of the previous block, thus ensuring consistency ofdata and that previous blocks have not been alteredwithout compromising the entire blockchain and revealingthat this change has occurred.

Immutability implies that data cannot be edited or deleted by participants or system administrator after it is written to the ledger.

2


While immutability is often discussed as anabsolute, innate feature of a blockchain –and one of its greatest strengths – manyclaims of immutability are over-exaggerated and not entirely accurate.Immutability is not an absolute quality, buta relative quality. "Immutability” does notnecessarily mean that data cannot bechanged; rather, it is a measure of thedifficulty required to change data on ablockchain. Additionally, any alterationcan often be detected fairly easily,ensuring the blockchain’s integrity.

The difficulty to change a data entry on aprivate blockchain depends on the level ofbuy-in needed from participants to sign anew replacement block (for example, thenumber of digital signatures necessary tosign this replacement). The more buy-innecessary to make a change, the greaterthe “immutability” of the blockchain.

Even if changing original blockchain datais possible with necessary effort, it ispreferable to instead produce a new linewith this edited information that refersback to the original block. As it pertains tothe financial services industry, regulatorsprefer this design principle as it ensures anaudit trail for reference.

What is Immutability?

3


What is an Oracle?

An oracle is an external source of data to a distributedledger which is considered to be authoritative, trusted anddefinitive. What makes an oracle unique (compared to asmart contract, for example) is this ability to deriveinformation from sources external to the ledger that itsupports.

An oracle is necessary once a smart contract is written tointeract with external data. One example is a bet on theSuper Bowl, where one party picks one team, one partypicks another, and an oracle determines the winner fromESPN and pushes that data to a smart contract to carryout the transaction. Similarly, within financial institutions,external data is often necessary in transactions – forexample, to verify that a loan has reached maturity, onemust gain knowledge about the current time.

Oracles bring in trustworthy external data to a distributed ledger, which is necessary if a smart contract needs to interact with external data.

Information are digitally signed in the transaction and are non-repudiable.

2


An oracle attests to the validity of data bydigitally signing facts within transactions aswell as the state resulting from thetransaction. A signed message from anoracle indicates proof of an event andcan then be used as an input into atransaction and distributed within thetransaction data itself. Once signed, anoracle cannot later “change its mind” andinvalidate transactions that were previouslyfound to be valid (allowing an oracle tochange its mind would result in a loss ofconsensus, undermining the integrity of thesystem).

There are similarities between oracles andsmart contracts and, in fact, they worktogether to carry out transactions thatrequire data from the outside world;however, there are key differences as well.While the integrity of a smart contractdepends on it being contained within theledger itself, an oracle is independent ofthe entities on the ledger it interacts withand can access data outside of theledger. Additionally, an oracle does notcontain legal agreements nor can itchange the state of agreements on theledger as a smart contract does.

Once the integrity of a transaction isexposed to a third party, it createssignificant vulnerabilities. Within “per-missioned” or known-participant systems,trust can be relied upon and fraud andmalicious attacks are much easier tomonitor and police. In a permissionlesssystem, on the other hand, the use of

oracles is more problematic as trust maybe more difficult to establish.

What is an Oracle?

3


What is Provenance?

The term provenance is very broad and used in manydifferent domains including food, manufacturing, businessdata, art and more. At a high level, provenance refers tothe sources and processes that produce a resource, pieceof information, or piece of data.

For example, food provenance refers to the origin of theingredients we purchase and eat to ensure food qualityand safety. Similarly, manufacturers must understand pastprocesses in order to ensure traceability, essential tominimize and discover inefficiencies, waste and excesscost.

The provenance of information is essential to determinewhether a piece of data is trusted and authentic*.However, while provenance is often conflated with trust,these terms are not the same (trust is derived fromprovenance, but is a subjective judgement).

In cryptography, provenance provides the linkage andoptics to determine where something came from. As such,distributed ledger technology can be used in order todetermine the provenance of an asset, which candetermine history of ownership.

Provenance refers to sources/processes that produce information/data.

Provenance in Distributed Ledger Technology can be used to track the history of the asset’s ownership.

2


A distributed ledger could potentially helpact like an added financial control, since(in theory) you would need to compromisesome subsection of the network to changethe historical record (though even then theoriginal cryptographic signature stays thesame for each individual transaction).Instead of consulting multiple systems tosee if something has been rehypo-thecated, DLT would allow one to consulta single logical layer to determine theprovenance of an asset and if it haschanged ownership in the past.

Some startups in the space are currentlyattempting to create “provenanceassurances” using distributed ledgertechnology in order to track theprovenance of certain goods. This couldpotentially be applied more broadly tohigh value or luxury goods, for example,whose provenance would otherwise bebased on paper receipts that could betampered with or altered.

What is Provenance?

3


What is a Smart Contract?

The term “smart contract” is used often when discussingdistributed ledger technology; however, its definition tendsto be broad, unclear and often controversial.

The term “smart contract” itself can be misleading, as thenotion of “smartness” and the extent to which something istruly a “contract” is questionable in many instances. Smartcontracts have become so closely associated withdistributed ledger technology and referred to so often,that in many cases the meaning has strayed from itsoriginal intent and has been generalized to the point ofinaccuracy.

The phrase and concept of a smart contract was originallydeveloped in 1994 by Nick Szabo, a legal scholar,computer scientist and cryptographer. Szabo’sbackground reveals the original intent and requirementsof a true smart contract: a confluence of both legal andcomputer code. In fact, Szabo’s entire methodology iscentered around the improvement of contract law andpractice through electronic protocols.

Smart contract is a agreement whereby execution is both automated and enforceable and runs on a distributed ledger.

Smart contracts can be used to handle common contractual conditions, removing the need for intermediaries.

2


It is critical to distinguish between thetechnically accurate concept of a “smartcontract” and today’s buzzword. There areoften situations where a simple piece ofcode is claimed to be a smart contractwhen, in reality, it is neither “smart” nortechnically a “contract”. A smart contractis not simply any piece of self-executingcomputer code, but is rather anagreement whose execution is bothautomated and enforceable.

A smart contract is automated in the sensethat the core business terms (the actual“transaction” among parties) is expressedthrough, and independently executed by,computer code and which no party canblock or otherwise tamper with. It isenforceable in the sense that it constituteslegally binding rights and obligations of theinvolved parties.

Currently, most smart contracts consist of acombination of both computer code – aprogrammable transaction protocol thatdefines the business terms of the contract –and legal prose that reflects that thecomputer code constitutes part of thebinding legal agreement between theparties, and is therefore also legallybinding.

Smart contracts are critical aspects ofdistributed ledger technology for financialinstitutions. They allow parties involved inan agreement to conclude with certaintythat they are in consensus at all times as tothe existence, nature and evolution of thefacts shared among them, which are

governed by the program. Additionally,they can be used to satisfy commoncontractual conditions, lower transactioncosts and risk, and eliminate the need fortrusted intermediaries.

What is a Smart Contract?

3


What is a State Object?

A state object is a digital document that represents andcompletely captures all relevant information about anagreement shared between parties, including itsexistence, content and current state.

More broadly, states can be thought of as referring to afact at a point in time (for example, a cash state or anidentity state). In Corda, states usually represent anobligation between parties. For example, a state objectcould represent a $100 obligation issued by a bank, aninterest rate swap, or a zero coupon bond. The CordaVault maintains the current position of state objects uponwhich two firms have agreed upon. In Corda, a stateobject is intended to be shared only with those who havea legitimate reason to see it.

A state object references (in the form of hashes) both thecontract code and legal prose contained by anagreement). States are established and can only bechanged by accepted transactions, which are governedby the rules of the contract code.

A state object captures relevant information about an agreement shared between parties.

State object references both contract code and legal prose contained by an agreement.

2


Digitally signed transactions consume zeroor more states and create zero or morenew states (see UTXO model)*. Sometransactions dictate that a new state becreated while other transactions causecurrent states to evolve into new states.Potential new states may be proposed byparties to an agreement, but a “consensusstate” is only reached once all parties tothe agreement achieve consensus.

*A state object is either current(considered a live obligation) or historic(considered no longer valid). Note thatthese terms are also sometimes referred toas unspent/unconsumed or spent /consu-med.

What is a State Object?

3


What is Tokenization?

Tokenization is the process of referencing an asset with a“token” so that it may be more easily and quicklyexchanged, or so that an existing asset may besubdivided into smaller units of value*.

Moving from Possession to Registry

Tokenization within the scope of distributed ledgertechnology is essentially the process of moving the registryof ownership of an asset onto a distributed ledger or ablockchain. Once an asset is on a ledger, the process ofownership exchange is simplified as an edit to the ledgerthat occurs on a separate cycle, ideally increasing theefficiency and velocity of settlement and potentiallyreducing operational risk at the same time. Themechanical settlement of an asset off-ledger may takeseveral days, or involve the physical transfer of a large orless-liquid asset, while the transfer of the tokenized assetmay occur instantaneously (or within whatever cycle isdetermined by the controller of the ledger, which may becentralized or decentralized).

Tokenization is the process of referencing an asset with a “token” so that it may be more easily and quickly exchanged, or so that an existing asset may be subdivided into smaller units of value.

Tokenization within the scope of distributed ledger technology is the process of moving the registry of ownership of an asset onto a distributed ledger.

2


Tokenization may use a third party thatholds the asset, though it does not have todepend on a trusted third party. A swap,for example, can be tokenized betweentwo parties that trust one another. The thirdparty may add value to the system byremoving counterparty risk, though it mayalso introduce both default and fraud risk.

The “token” itself has no extrinsic value – itis not a real representation of an asset on aledger but is rather a derivative of an asset– an agreement to transfer title or rights(and in some cases, obligations) to theunderlying asset. The owner of a tokendoes not store a token on a computer;rather, the owner stores the keys that lethim create a new entry on the ledger andreassign ownership to someone else.

Tokens may be intrinsic (built-in) to ablockchain or asset-backed (issued by aparty onto a blockchain). Intrinsic tokensare fundamental to the blockchain ofwhich they are a part, and are usually anincentive/reward to validate transactionsor create blocks (BTC on the Bitcoinblockchain, XRP on the Ripple network, ETHon Ethereum).

Asset-Backed Tokens: Digital Represen-tations of Title Over Physical Objects

Asset-backed tokens represent ownershipof an underlying asset. A non-digitalexample of this is bullion depositorycertificates, which represent the ownershipof gold bullion or coins. These certificatesact as proof of ownership for the physical

gold and are then able to be passed, likecash, which is much easier and moreefficient than moving physical gold. Asset-backed tokens are a digital equivalent ofthis bullion depository certificates example.These tokens are passed between partiesand these transactions are recorded on ablockchain. An issuer is required to receivea token in exchange for the underlyingasset.

The movement of asset-backed tokensmust accurately represent the state of theunderlying asset. In order to accuratelytrack a tokenized asset that originates off-ledger (for example, a car), the physicalasset must be tagged with a uniqueidentification method that is associatedwith the corresponding location or hashon-ledger.

A car is already associated with a unique,tamper-proof ID – a VIN number – that inturn can be associated with the on-ledgerhash. If the asset isn’t already easilyassociated with a unique ID, thisidentification method may be a physicalsecure device that ties back to this uniquehash on ledger**.


3


*Another benefit and use of tokenization is to subdivide an existing asset into smaller units of value. In this case, tokenization allows for the association of a single large value asset (such as a cargo ship or aircraft) with a larger number of tokens, each representing fractional ownership of the original asset. An existing example of this would be a Real Estate Investment Trust (REIT), which subdivides large scale real estate investment into shares and allows a wider range of investors access this sector.

**Such methods include a secure tag, digital locking and artificial DNA. This is an emerging field which offers the promise of multiple, indelible or tamper-proof methods of associating a physical object with its representation on the ledger.

Note: Colored coins are problematic as a tokenization method because the lack of standards means you risk the loss of the meaning of the “color” between wallets, and thus risk losing the associated data about the tokenized assets.


4


What is a UTXO?

A UTXO is an unspent transaction output. In an acceptedtransaction in a valid blockchain payment system (such asBitcoin), only unspent outputs can be used as inputs to atransaction. When a transaction takes place, inputs aredeleted and outputs are created as new UTXOs that maythen be consumed in future transactions.

In the Bitcoin network, which uses this model, a UTXO is theamount that is transferred to a Bitcoin address (along withinformation required to unlock the output amount*) duringa transaction. Received amounts (UTXOs) are usedindividually during a transaction and new outputs arecreated – one for the receiver and, if applicable, one forthe amount that is left over (change output). The amountsent to the recipient becomes a new UTXO in therecipient’s address while the change output becomes anew UTXO in the sender’s address that may be used in afuture transaction.

Unspent outputs can be used as inputs to a transaction. Inputs will then be converted to outputs as new UTXO that can be spent in future transactions.

2


Corda uses a UTXO model – an entry iseither historic (sometimes referred to as"spent" or "consumed") or current(sometimes referred to as "unspent" or"unconsumed"), but it cannot be changed.Corda assumes that data being processedrepresents financial agreements betweenidentifiable parties and that theseinstitutions require a significant number ofsuch agreements to be managed by theplatform. Therefore, Corda must be able tosupport parallel execution, while alsoensuring correct transaction ordering whentwo transactions seek to act on the samepiece of shared state. A UTXO modelprovides this, allowing transactions to beprocessed independently and in parallel,even for high traffic legal entities.

What is a UTXO?

3

what is the distinction between a blockchain and a ...blockchain has a shared and replicated ledger...

Documents