Financial Innovation and the Internet 2019 Fall Kenji Saito Professor, Graduate School of Business and Finance, Waseda University [email protected] Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.1/35
technology tries to answer Physical model with beakers and newspaper Functional layers (validity, existence, uniqueness) Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.4/35
answer Physical model with beakers and newspaper Functional layers (validity, existence, uniqueness) Possibilities, Impossibilities and True Worth Applications people talk about, and thoese workable in reality Problems of blockchain Blockchain’s true worth and the “last will test” Science Fiction Prototyping What about it? Assignment Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.5/35
Satoshi Nakamoto) Satoshi him or her or themselves called it “distributed time-stamp server” Not a good word for representing a concept (catchy, but manipulating the impression) Something implemented by Chain of ← actually, backward list of Blocks ← actually, sets of data For example, we don’t call TV “picture tube” today (or do we?) Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.6/35
so that no one can deny the contents and existence of them Denying = rejection, deletion, alteration, fabrication 2) Anyone can verify that ≪ to realize 1) ≫ 3) No one can stop the above ≪ to realize 1) ≫ “Anyone” and “no one” are defined stakeholders Which is more difficult to achieve: stakeholders are “all humankind” vs. “just the parties involved”? ⇒ As if records were locked up in the air ⇒ Double problems Is that really realized? If it is realized, can we apply it to many cases? Issue of reliable linking between physical entities and digital records · In supply chains, is it possible to reliably track things through processing, etc.? Challenges such as real-time performance, scalability, adaptability, and sustainability (can it be dependable?) Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.7/35
send money whenever we want, and never let anyone stop us”? Distrust of (central) bank money Bitcoin’s “Answer” Cannot depend on any particular service provider ⇒ Exchange digital coins over the Internet by P2P (peer-to-peer) What if they deny that they sent a coin? ⇒ Use digital signatures (collateral for verifiability and non-repudiability of contents) Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.8/35
by digital signatures alone Need to prevent double spending (want to ensure non-repudiation of existence) ⇒ Put the evidence of the transaction in newspaper What if refused for publishing or service is discontinued? ⇒ Place evidence of a transaction in “newspaper” (as collective evidences of events) issued by a crowd And thereby records are locked up in the air Theft of coins based on this idea always follow the story made typical by the Mt.GOX or CoinCheck incident “Don’t let anyone stop us from spending our own money whenever we want to” ⇒ Has to prove that the user is oneself by their own → Zero-knowledge proof of possession of the private key → Anyone with the private key is the user oneself ⇒ Transaction is verifiable by all but irrevocable → Stolen coins can be tracked but not recovered Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.9/35
of liquid of no value to mankind Contained in a tank Individuals can hold as many beakers as they like, measuring down to 1 100,000,000cm3 (it has a locked lid) Only “editor” selected every 10 minutes on average can pump now 12.5cm3 into their beaker Chosen by a special lottery The winning lottery is held in each box, and each person draws the lottery with all their strength → non-stoppable procedure Coordinate the proportion of winning lots so that someone is chosen every 10 minutes on average Volume pumped is reduced by half every about 4 years (every 21 thousand pages of “newspaper” described later) Started from 50cm3 in January 2009 Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.10/35
of fluid between beakers Recorded as “a signed article” by the pourer Post the article in the “newspaper” made by everyone Selected “Editor” verifies the articles and publishes them in the last page of the newspaper Page carries the evidence of winning the lottery Editor also gets “overflow” of trades on the page If people publish a page with the same page number. . . Longer sequence of pages wins People sometimes lose the key of their beakers Create this digitally, and pretend that it’s a currency → Bitcoin There is no money or currency that does not need pretension Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.11/35
number : n page number : n+1 page number : n+2 Cryptographic digest of the previous page (must be less than or equal to the target value) some extra number (Nonce : Number used Once) (random value to make the digest less than or equal to the target) Page digest (output by a cryptographic hash function) must be less than or equal to target We don’t know how to manipulate the original data to get the right digest This is the principle of the lottery, which requires the same amount of cost to fake the history Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.13/35
page number : n+1 page number : n+2 page number : n+3 page number : n+1 page number : n+2 page number : n+3 page number : n+4 This history is valid Sometimes page sequences are split when someone else wins the lottery at about the same time History is the hardest to tamper with when the lottery cumulatively costs the most Everyone agrees that it is the official history (strict consensus is not achieved) Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.14/35
n+1 block_no : n+2 block_no : n+3 block_no : n+1 block_no : n+2 block_no : n+3 block_no : n+4 Histroy with the largest cost to record or modify (history the most difficult to alter) is chosen Cryptographic digest of the previous block Transactions are digitally signed To create a block, its cryptographic digest needs to be below some certain number (Proof of Work) or one needs to win by voting weighted by the stakes in cryptocurrency (Proof of Stake) [both costly] Creator of a block can record the reward in cryptocurrency in the block, which is effective only when the block is included in the chosen history Means are provided to confirm existence of transactions validity existence uniqueness In case of Proof of Work, the cost of power is balanced against the market value of the native currency Everyone confirms that records are not tampered with by the mechanism protected by the price of the native currency Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.15/35
access (no need for mediation), agreed real-time settlements, business rule descriptions, and confidentiality Corporate behavior (automating corporate management, especially in financial matters) Real-time execution and confidentiality control of share splits, capital reductions and consolidations, share transfers and exchanges, mergers, third-party allocation of new shares, etc. Supply Chain Traceback of materials, and record and search from production, storage to sales (beware of linkage problem) Master Data Management Only authorized personnel can update and designated reviewers approve it Sharing Economy and IoT Smart cities/towns, transportation, healthcare/fitness, retail, architecture, education, etc. (implicitly real-time and on a large scale) where trust is not necessarily established Red letters denote parts that blockchains are not good at Within the problems we want to solve, there are sub-problems that have not been solved yet Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.17/35
. . Remittances that bypass banking networks That’s a huge impact Proof of Existence Ex. Proof of Existence, Everledger, Factom, . . . Embed arbitrary digests in a blockchain (piggybacking hack) There is also a method of embedding a single digest of a large number of records Proof that a record has existed and has not been tampered with Origin Certification (traceability, tracking and accounting) Originally designed application category (alternative to “newspaper”) Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.18/35
0SJHJO$FSUJpDBUJPO4FSWJDF .FSLMFUSFF .FSLMFSPPU 0CUBJOTVCUSFF 4FBSDIGPS.FSLMFSPPU &NCFEBUBDFSUBJOQPJOUPGUJNF ʜ ʜ 3FD 3FD 3FD 3FD EJHFTU EJHFTU EJHFTU EJHFTU EJHFTU EJHFTU "1* EJHFTU *OGBDU JUJTQPTTJCMFUPNBLFB.BSLMFUSFFXJUI TFWFSBMUFOTPGUIPVTBOETPGSFDPSET &WFOJOUIFDBTF UIFIFJHIUPGUIFUSFFJT BCPVUUP *OPSEFSUPDPOpSNUIFFYJTUFODFPGSFD UIF.BSLMFSPPUDBOCFDBMDVMBUFEJG POMZUIFTVCUSFFTIPXOJOCMVFJTPCUBJOFE Services can be distributed (made large-scale and fast) by existing technologies (distributed KVS and DHT) Can be used for accounting in a broad sense Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.19/35
of trial and error In fact, a lot of new designs are being tested If we do not have governance for (or if we do not know how to accommodate) technological changes, we cannot use it in society Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.20/35
participants gain half of the hash rate, blockchain cannot be guaranteed to work correctly Risky in principle if the hashrate is doubled quickly → And it is happening On the other hand, what if it doesn’t double rapidly? → Dilemma of providing room for malicious participants What if it suddenly halves? → Very risky in principle Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.21/35
a new business on the beach: If a customer pays in bitcoin, a drone flying overhead will drop them a can of juice When should the drone drop the can? Reality that goes in real-time and blockchain’s behavior are very different But as a business decision, a risk taker can drop the canned juice the moment they detect a payment As long as they are in a position to use social infrastructure, they can act disruptively Not because it is a perfect technology But because it is a fairly cheap platform Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.22/35
decentralized investment fund built on Ethereum Split (fund split) was recursively called, and 360,000 ETH (5 to 6 billion yen) was stolen (2016/6/17) Choices Do nothing Soft fork (maintains compatibility → freezes the address of the thief) Funds are not returned Hard fork (No compatibility → rewrites history; who controls the present controls the past) Worst occurrence of “Oneness Trap” (described later) in a sense Community chose “hard fork” ! (executed on 2016/7/20) “Most interesting. Gravity’s silhouette remains, but the star and all its planets have disappeared. How can this be?” “Because someone erased it from the archive memory.” — from Star Wars: Episode II – Attack of the Clones So the incident never happened Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.23/35
the cost of maintaining data structures rises linearly as transactions increase It does not scale out Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.24/35
nodes does not solve or mitigate performance challenges Must be “the world is one” to work System does not work correctly if the network is partitioned by a large-scale disasters or political change Difficulty of governance to advance technology You cannot “try something different partially, and if it works, apply it to the whole” Impossibility of governance : Agreement by the “whole” must be maintained, but the “whole” cannot be defined ⇒ Powered few changes the technology instead ⇒ Those are disadvantages of non-decentralized structure Conversely, there are great expectations and potential for truly decentralized “record fixation device in the air” Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.25/35
ETH as a currency crashes and declines Supported by miners’ motivation to get ETH When the value of ETH drops, miners withdraw Can people who want to run apps (smart contracts) buy ETH to maintain the price? If ETH’s market participants are primarily app users, may be . . . (but they aren’t) The design of the raw Bitcoin is goal-consistent, but . . . In other words, “Bitcoin cannot survive if BTC declines” would be fine But as proof applications such as Everledger and Factom advance, similar problems arise Either way, the future of the system depends on the interests of the miners ⇒ Need to separate application platforms from currency systems That is where the recent ledger technology is going, hopefully (because many still aren’t) Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.26/35
1. Guarantee of Validity (For Bitcoin : UTXO data structure + digital signature) Self-contained and verifiable by everyone ⇒ Beneficial (yet confidentiality is a problem) 2. Proof of Existence (For Bitcoin : Hash-chain w/ proof of work) Evidence base for nonrepudiation of existence ⇒ Not as strong as expected 3. Consent of Uniqueness (For Bitcoin : Nakamoto consensus) Trial to finalize transactions ⇒ Problem in synchronizing with reality In addition, the non-decentralized structure and incentive mismatch by 2 and 3 Necessary to simultaneously solve the problems and explore applications on the assumption that the problems will be solved Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.27/35
digital signature Proof of Existence e.g. hash-chain with proof of work Consent of Uniqueness e.g. Nakamoto consensus Description of Rules e.g. transfer of bitcoins - The content of a transaction cannot be altered, - not contrary to past history of transactions regarding the asset, - and the transaction is cast by a legitimate user - Cannot delete the evidence of an existing transaction in the past, - and cannot fabricate an evidence of a transaction that did not exist - When two mutually contradicting transactions are cast, (eventually) everyone chooses the same one to place in the history - Application logic to decide what valid transactions are Blockchain ≪or alike≫ (tries to) bring the End-to-End philosophy of the Internet into reality in the control of assets, thereby making a Record Fixation Device in the Air The center is automated Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.28/35
digital signature Proof of Existence e.g. hash-chain with proof of work Consent of Uniqueness e.g. Nakamoto consensus Description of Rules e.g. transfer of bitcoins - The content of a transaction cannot be altered, - not contrary to past history of transactions regarding the asset, - and the transaction is cast by a legitimate user - Cannot delete the evidence of an existing transaction in the past, - and cannot fabricate an evidence of a transaction that did not exist - When two mutually contradicting transactions are cast, (eventually) everyone chooses the same one to place in the history - Application logic to decide what valid transactions are Digital signatures suffice! Needed to rely on trust! Design-dependent feature! Application layer! Blockchain ≪or alike≫ (tries to) bring the End-to-End philosophy of the Internet into reality in the control of assets, thereby making a Record Fixation Device in the Air The center is automated Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.28/35
the thing you can never do without blockchain? Ex1 : Digitizing the last will and testament (in a thought-experiment sense) After the death of the person, the private key used for the digital signature may no longer be a secret Can’t believe a notary saying, “it is as signed before the death” (possible collusion with malicious heir) Ex2 : Online banking passbook data as proof (for the liabilities of banks) If you download it as a CSV file, the data anyone can create is not considered as evidence Even with the digital signature of the bank, once the private key is leaked, the data can be created by anyone Prove “data digitally signed at cetain past date has not been tampered with” (instead of believing those who insist so) “The Last Will Test” is to ask them if they can do that with their blockchain A test to see if something that someone has been pitching as “it’s a blockchain” really makes sense An idea (hash-chain with proof of work) that may satisfy this true worth, combined with existing technologies around the idea is the Bitcoin blockchain Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.29/35
that can pass the “last will test”? Public blockchain may pass the test while the market value of the native currency is high Defense such that it would cost a lot to tamper with When the price of the native currency drops or crashes it becomes unreliable Private ledger systems in general only insist, so they do not pass the test Mostly, “blockchain made and operated by XXX Inc.” is meaningless ← please be careful We are building new technology to make it pass the test BBc-1 (Beyond Blockchain One; https://github.com/beyond-blockchain) There is an inherent challenge of linkage between records and entities Includes areas that cannot be solved by technology alone (does the public key really belong to the person?) Including the openness of the source code, it is roughly the problem of . . . “How can we trust automated mechanisms?” Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.30/35
on or disguised as the thoughts and ideas of science and technology The world with existing science and technology is the real world Ex1 : Medical drama Ex2 : Economic novels The world with science and technology unknown to the real world → Fiction Ex1 : Medical drama with nano-machines Ex2 : Drama with digital currency that depreciates (my book “NEO in Wonderland”) Designing new media and putting it into society = living science fiction To work out plans for that = to write science fiction Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.32/35
of the word “smart contract” to fantasize a specific application example, and state the application briefly That is, go ahead and write a very short science fiction Deadline and how to submit November 27, 2019 at 17:59 JST From Course N@vi Lecture 8 : Blockchain (2) — FinTech — Financial Innovation and the Internet 2019 Fall — 2019-11-22 – p.34/35
ks91
hsugita
__allllllllez__
mitsuzono
pelelgrino
coconala_engineer
taishin
1ftseabass
rinchsan
takakiyo
nagix
manarobot
kenichirokimura
holman
moore
addyosmani
tanoku
philhawksworth
paigeccino
skipperchong
maltzj
kevinliebholz
ufuk
destraynor
kneath
