<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Rodrigue Afota on Medium]]></title>
        <description><![CDATA[Stories by Rodrigue Afota on Medium]]></description>
        <link>https://medium.com/@Rodrigue.Afota?source=rss-ba3c60f60fd2------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*4WDZevSj3ewPkgIYH_FP0A@2x.jpeg</url>
            <title>Stories by Rodrigue Afota on Medium</title>
            <link>https://medium.com/@Rodrigue.Afota?source=rss-ba3c60f60fd2------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 20:02:37 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@Rodrigue.Afota/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[The fallacies and perils of AI-powered credit-risk engines for Fin Techs]]></title>
            <link>https://medium.com/@Rodrigue.Afota/the-fallacies-and-perils-of-ai-powered-credit-risk-engines-for-fin-techs-b9296144958a?source=rss-ba3c60f60fd2------2</link>
            <guid isPermaLink="false">https://medium.com/p/b9296144958a</guid>
            <category><![CDATA[saudi-arabia]]></category>
            <category><![CDATA[credit-risk]]></category>
            <category><![CDATA[ai-finance]]></category>
            <category><![CDATA[digital-lending]]></category>
            <category><![CDATA[fintech]]></category>
            <dc:creator><![CDATA[Rodrigue Afota]]></dc:creator>
            <pubDate>Thu, 29 May 2025 19:52:36 GMT</pubDate>
            <atom:updated>2025-05-29T19:52:36.270Z</atom:updated>
            <content:encoded><![CDATA[<p>“This KSA Fin Tech is very smart; they use AI to reduce risk and future credit losses”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iEbWsd3GyMe_McgZDDNXpQ.jpeg" /></figure><p>When I heard this comment from an otherwise smart and experienced investor, it just did not sound right and did not make sense to me, and it got me thinking about this topic.</p><p>Let’s, for a moment, assume that we are not dealing with one of those “break things and move fast” Fin Techs; such an executive philosophy would be entirely at odds with the careful and well-planned regulatory environment in KSA, and that philosophy would not be tolerated for much longer by the Central Bank.</p><p>Instead, let us project ourselves into the future and assume that all participants in the private credit and lending market use perfectly trained AI engines on massive amounts of relevant data. Banks, Finance Companies, Fin Tech… Everybody! Would that change in any way the quantum of credit risk in the system? Zero! Not one single iota. It’s worse actually: when executives get convinced by their own spin that they have better systems than anybody else, the opposite does happen, and caution is out the window. Risk within the system increases, and in some cases, it can reach a bubble-type situation.</p><p>Not too dissimilar with the US non-conforming, no-doc, mortgage craziness in the early 2000s. And we all know how that turned out.</p><p>No; what a good, well-trained, AI-engine could do is potentially reduce the PD (Default Probability) in the loan book. It would not, however, make any change to LGD (Loss Given Default), which depends on external factors such as the legal system and recovery practices (for readers unfamiliar with credit risk management: expected credit losses in any loan book are equal to PD x LGD).</p><p>The problem, you see, is that every single executive is going to think they have a better AI-credit decisioning engine than anybody else, and the overall result is that, on average, PD in the system won’t change as a whole! What may change is how losses are allocated among market participants; some participants get better results than before, and others get worse results than before. But the overall amount of risk in the system is the same.</p><p>This brings me to the second critical insight: which player will get improved PD and which player will lose out? And this gets very interesting because I now believe that any player (generally Fin Tech trying to shine and charm Series A and Seed investors and telling them what they want to hear) now claiming to use AI front and centre in their credit model is deluding himself. Can I ask: AI-engines trained on what data, exactly? The commercial banks have data, and for the most part, they do not use AI yet, and rely on traditional credit scoring metrics and their own data on their customers. What about Fin Tech? Well, they do not have any serious data. They do not host bank accounts or wallet accounts so have no clue about cash flows and net balance patterns of the people or companies they lend money to. Nor could they, even if they wanted to. A Finance Company acquiring a digital bank or EMI license is Saudi Arabia; well, let’s just say it has never happened, and I think it will be extraordinarily difficult to overcome Central Banking objections.</p><p>So, what is a Fin Tech doing, exactly, when using AI for credit-decisioning? I think it is worse than having no AI model at all, unless someone can prove otherwise to me. That Fin Tech AI engine will be very poorly trained. <a href="https://hbr.org/2018/04/if-your-data-is-bad-your-machine-learning-tools-are-useless">And the predictive capability of a poorly trained engine, whichever is the training topic, is by then probably worse than going to the casino</a>.</p><p>So, I have a general piece of advice to investors. Any digital lending FinTech pitch in front of you putting AI front and centre, throw that pitch in the bin. You will outperform fellow VC investors. Unless there is a very clear, convincing, mitigating fact, such as agreed-upon access to one of the large commercial banks’ retail data.</p><p>In the long run, AI is just a tool, like credit risk scoring was and still is. And like any tool, AI will move from team to team, from bank to bank to Fin Co to FinTech, etc.; the real and only long-term competitive advantage is the share of relevant data, through customer ownership and share of wallet. Data ownership will be king for a very long time in credit risk!</p><p>That, in a nutshell, is what I think.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b9296144958a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Can Bitcoin Ever be Improved Upon?]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@Rodrigue.Afota/can-bitcoin-ever-be-improved-upon-c0fc826fee96?source=rss-ba3c60f60fd2------2"><img src="https://cdn-images-1.medium.com/max/2240/1*o2VoIRtjnmJuCFnBTQYJRw.jpeg" width="2240"></a></p><p class="medium-feed-snippet">One statistic reported by BitPay, the leading crypto processing payment company, caught my attention.</p><p class="medium-feed-link"><a href="https://medium.com/@Rodrigue.Afota/can-bitcoin-ever-be-improved-upon-c0fc826fee96?source=rss-ba3c60f60fd2------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@Rodrigue.Afota/can-bitcoin-ever-be-improved-upon-c0fc826fee96?source=rss-ba3c60f60fd2------2</link>
            <guid isPermaLink="false">https://medium.com/p/c0fc826fee96</guid>
            <category><![CDATA[digital-wallet]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[payments]]></category>
            <category><![CDATA[bitcoin]]></category>
            <dc:creator><![CDATA[Rodrigue Afota]]></dc:creator>
            <pubDate>Wed, 17 Jun 2020 09:35:24 GMT</pubDate>
            <atom:updated>2020-06-17T09:35:24.862Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Looking Beyond The Crypto Winter]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@Rodrigue.Afota/looking-beyond-the-crypto-winter-3c8bc7186e8e?source=rss-ba3c60f60fd2------2"><img src="https://cdn-images-1.medium.com/max/600/1*HG7FLSPpQrDVnfA_M6FEWQ.png" width="600"></a></p><p class="medium-feed-snippet">One year into the current Crypto Winter and, as the saying goes, it&#x2019;s when the tide turns that we can see who is swimming naked. Trading&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@Rodrigue.Afota/looking-beyond-the-crypto-winter-3c8bc7186e8e?source=rss-ba3c60f60fd2------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@Rodrigue.Afota/looking-beyond-the-crypto-winter-3c8bc7186e8e?source=rss-ba3c60f60fd2------2</link>
            <guid isPermaLink="false">https://medium.com/p/3c8bc7186e8e</guid>
            <category><![CDATA[bitcoin]]></category>
            <category><![CDATA[security-token]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[finance]]></category>
            <dc:creator><![CDATA[Rodrigue Afota]]></dc:creator>
            <pubDate>Sun, 10 Mar 2019 14:18:46 GMT</pubDate>
            <atom:updated>2019-03-13T06:36:11.769Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[From Initial Coin Offerings to Tokenised Securities: Far From the End Game]]></title>
            <link>https://medium.com/@Rodrigue.Afota/from-initial-coin-offerings-to-tokenised-securities-far-from-the-end-game-a38a74ac27ce?source=rss-ba3c60f60fd2------2</link>
            <guid isPermaLink="false">https://medium.com/p/a38a74ac27ce</guid>
            <category><![CDATA[ico]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[bitcoin]]></category>
            <category><![CDATA[tokenization]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Rodrigue Afota]]></dc:creator>
            <pubDate>Wed, 10 Oct 2018 13:28:50 GMT</pubDate>
            <atom:updated>2018-10-12T17:39:57.601Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/325/1*mnJBubeH63u11mu16R1jKA.png" /></figure><p><strong>From Initial Coin Offerings to Tokenised Securities: Far From the End Game</strong></p><p>On September 12 this year, a US federal judge <a href="https://www.law.com/newyorklawjournal/2018/09/18/edny-u-s-securities-laws-can-be-used-to-prosecute-ico-fraud/?slreturn=20180908100959">issued an order</a> finding that a defendant had promoted digital tokens (RE Coin and Diamond) through initial coin offerings in violation of U.S. securities laws, granting a victory to the US Department of Justice and the SEC until a jury can make a final decision. The promoter of these coins was charged on, essentially, two counts: illegally proceeding with an unregistered offering of securities and making misrepresentations and false claims. For the crypto world, the core issue was whether those coin offerings qualified under US Securities Laws and whether the coins should have been treated as securities. In issuing his order, the judge applied the “<a href="https://en.wikipedia.org/wiki/SEC_v._W._J._Howey_Co.">Howey Test</a>”, a test arising from US case law and used to determine when an investment offering instrument is a “security”.</p><p><em>“The test is whether the scheme involves an investment of money in a common enterprise with profits to come solely from the efforts of others. If that test be satisfied, it is immaterial whether the enterprise is speculative or non-speculative or whether there is a sale of property with or without intrinsic value.”</em></p><p>Much has been written about the Howey Test and its applicability to ICOs, and the September 12 decision was the second time a federal judge issued a finding to that effect. By now, there is a broad consensus that almost all ICOs done to date are, in fact, unregistered securities offerings, exposing their promoters, enablers, advisers, brokers, underwriters and even lead investors to prosecution and fines, and this is a important factor explaining the slowdown in the market. Few people are willing to go to jail now that the warnings have been made.</p><p>The ICO market has clearly cooled down since its May 2018 peak (chart courtesy of ICORating.com).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b_2xlMP0aBcjGfoG4YwYVg.png" /></figure><p>The fear of legal prosecution is only one factor behind the decrease in ICO closings. A more important one is simply that the market has turned and many ICO are under water. This has upset investors, the majority of which were retail individuals and it has led to threats of legal action, open revolt in some telegram chat rooms, class actions and the involvement of regulatory agencies. It has also cast a spotlight on dubious practices including insider trading, outright scams and misrepresentations by ICO promoters, and shed light on some structural issues with ICOs, including the lack of contractual recourse and proper governance and the perverse incentives for entrepreneurs. After all, why not forget about the project and keep the money when the going gets tough?</p><p>Yet, for all the negatives, there is no denying the critical innovation of ICOs as a decentralised blockchain-based crowdfunding tool. The number of projects in the pipeline from various industry and countries keeps increasing (ICORating.com), with a total industry wide fundraising in excess of US$8 billion year to date.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YjzKVC7AnGbeQgJhOfXQBw.png" /></figure><p>And this is where “tokenized securities”, otherwise known as security token offerings or “STO” come into play. What if the industry could collectively re-work how ICOs are structured and make them compliant with securities laws? It’s a very logical idea, one that holds the promise of attracting large and more stable institutional investors and bringing liquidity to many illiquid assets and fundraising capacity to many entrepreneurs.</p><p>Many companies have entered the space, from “compliance protocols” to “security tokens exchanges”: “Compliance protocols” purpose is to mimic registered investment banks distributing traditional securities in compliance with regulations, while “security tokens exchange” purpose is to create trading platforms where tokens can be exchanged between qualified buyers and sellers via accredited brokers, in the process addressing some of the concerns expressed by regulators on the lack of proper “know your customer”, “anti money laundering” and “counter terrorism financing” procedures.</p><p>Reading from the “evangelical” predictions and statements made by some of these companies, one would be forgiven for thinking STOs will revolutionise capital markets . We doubt that: the public credit markets and the main stocks markets are effective and efficient and they are tested. There is no need nor demand from institutional investors to instead switch to a technology (public blockchain) that is still evolving, does not know yet how to both scale and remain decentralised and that is still prone to security breaches, hacks and market abuse. Instead, the believe the potential market will come from start-ups and young companies struggling to attract capital through traditional means as well as asset-backed tokenization, markets which retail investors can’t easily access.</p><p>That most of the ICO/STO activity would occur on Ethereum, a very creative token platform that is <a href="https://medium.com/@Rodrigue.Afota/on-public-blockchains-scaling-and-re-centralisation-a309a58cf549">struggling to remain decentralised</a> under the weight of its own success is a concern too: if a STO platform is not truly decentralised, then what is the point of using public blockchains?</p><p>While we applaud the new wave of entrepreneurs building up the STO infrastructure, we believe the majority of these companies either misunderstand the scope of regulatory challenges or misrepresent their role. Adequate disclosures are seldom found; instead, there is a tendency to narrowly equate lawful offerings and regulatory compliance with a “tick the KYC box” attitude, and there is in our estimation an untested and over-optimistic view about the ability of smart contracts code embedded into protocol tokens to ensure that a crypto asset will never, anywhere, fall into the wrong hands. This is implausible.</p><p>Let us explain. A traditional bank such as HSBC spends in excess of US$2 billion per year on compliance alone, with over 6000 employees across the globe whose tasks are: keep up to date with the ever-changing requirements issued by law makers and regulatory bodies in each jurisdictions, maintain active relationships with such regulators, keep track of sanctioned entities and individuals, perform increasingly complex background checks and fraud identifications, and enforce these regulations on their customers and their front offices. And even then, they occasionally get it wrong and get fined.</p><p>Here comes a start up focusing on compliance protocol, staffed mostly by IT and computer science entrepreneurs, with little to no financial or capital market background, but a strong belief that encoding rules and regulations in a smart contract will do what HSBC is actually struggling to do. These start-ups have no relationships with mainstream regulators, no enforcement experience and rely instead on snapshots memorandums compiled by law firms with a summary of restrictions on selling restrictions. The memorandums come with pages of qualifications and exclusions. Yet, these snapshots restrictions are then encoded in smart contracts without any concept of minimum denominations, hoping that that will work out over time once the token is issued.</p><p>Strikingly, there are no STO specific laws, no clarity, no proper regulatory framework in any major jurisdiction. As far as we can tell, few (if any) of these companies maintain any active supervisory relationship with any mainstream regulator (taking the sandbox concept aside). In fact, It could be the case that the current STO eco-system is not a real improvement on the ICO market: instead of enforcing standards consistently, it creates the impression of legality and legitimacy and may end up unwittingly persuading investors to invest and companies to issue STOs that are, in fact, not much better than ICOs from the standpoint of securities laws and investor protections.</p><p>In a number of major jurisdictions, government authorities are taking steps to establish a framework to allow distribution of tokens (in <a href="http://www.france-finance.fr/2018/09/17/assemblee-nationale-cadre-juridique-levees-fonds-crypto-monnaies/">France</a> and <a href="https://www.parliament.uk/business/committees/committees-a-z/commons-select/treasury-committee/inquiries1/parliament-2017/digital-currencies-17-19/">UK</a> most notably). The indications are that countries will set up a filing and registration protocol, providing a “safe harbour” to tokens distributed locally. The current proposals in France apply to tokens that have substantial economic links to France: more generally, we believe these framework will make the most sense for regionally focused and distributed transactions, more on this later. The United States is in a more complex situation due to the overlap between federal and state agencies, law makers, justice departments and the number of regulatory bodies involved.</p><p>We are aware of a new wave of interesting ideas, involving partial re-centralisation through digital Central Securities Depositories (a digital “Euroclear” for crypto assets). These ideas need more work, and they need substantial involvement from mainstream regulatory bodies such as the European Central Bank or the Financial Conduct Authority. At BlockTrust, we are ourselves focusing on being part of the long term STO infrastructure, focusing on the Middle East, India and South East Asia. To explain our approach and philosophy, here is a comparison with traditional asset backed securities. What happens when a security backed by assets such as leases or real estate mortgages is issued and when the issuer, in effect, goes “public”?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZQhL_n-Bt9SQ4zaYWWg0Pg.png" /></figure><p>Our key insights are:</p><ol><li>All third parties involved in a deal are companies under the oversight and scrutiny of major regulators: structuring bank, trustee &amp; custodian, registrar (CSD), exchanges and brokers</li><li>Relationships between parties are contractual and enforceable</li><li>A Trustee / Custodian will interface throughout the duration of the deal between investors (which are often scattered) and the company. Trustee can and often do sue the Company or its management when it misbehave. Trustee &amp; Custodian also ensure, along with the Company, compliance with standard listing requirements such as IFRS reporting, filings and corporate disclosures</li><li>Rating agencies and research firms usually provide an enhanced degree of scrutiny to the deal</li><li>The security must be registered in most jurisdictions where it will be distributed</li></ol><p>So, here are a number of interesting ideas:</p><p><strong>First</strong>, compliance protocol companies and STO exchanges could expand their focus to ensure that issuers / companies fully understand the consequences of going “public”. Whenever there is a security listed on a proper and legitimate US or EU exchange, the company has to report its accounts under IFRS or US GAAP and comply with mandatory disclosures and regular filings. In the process ensuring that these companies do not later get in trouble and aligning the STO market to something closer to Alternext (Euronext for small caps) or AIM (London Stock Exchange equivalent).</p><p>In the United States, <a href="https://www.sec.gov/fast-answers/answers-regdhtm.html">Regulation D</a> offerings provide a limited exemption to filings and registration requirements. The scope of Regulation D is fairly tight: to qualify, an offering must be and remain at an aggregate of less than US$5 million, or its distribution must be limited to sophisticated investors, along with strict transfer and trading restrictions. We do not believe Regulation D is a viable option for the most part.</p><p><strong>Second</strong>, Trustees &amp; Custodians. The lack of Trustees, who usually carry a fiduciary duty to investors, is the one single biggest structural weakness in the ICO/STO space, and none of the current compliance protocols will change that. A “Digital Trustee” can front investors and face companies and issuers, there is a proper financing agreement or equity subscription contract with the issuer. It shall have no part in the initial distribution process focusing instead on being a solid and low-volatility player. Think Bank of New York or State Street versus Goldman Sachs. When things go wrong, a Trustee can contractually sue and enforce an actual contract, it has no conflict of interest. Compare this with an ICO: there is no contract, it’s nearly impossible to sue anybody. Trustees provide an enormous degree of investor protection, and, I submit, STO structured with a Digital Trustee will sell faster and at a premium.</p><p><strong>Third</strong>, regionalisation. Forget about companies issuing STO on a global basis; regionally focused platforms will have a more sustainable business model. An example to illustrate why: take an office building in Dubai; the owner wants to get liquidity against it, which presumably he can’t get at attractive terms from the banks via a lease-back. Such a real estate-backed STO would 99% be distributed to regional Middle East investors familiar with the property or the management company. It would be listed on a regional exchange (such as a “digital” Nasdaq Dubai). Ensuring primary and secondary trading compliance with AML/CTF regulations would be easier as we’d only need to fully understand the rules and regulations in the Middle East. Issuers would only need to understand regional reporting and filing requirements in jurisdictions they already understand. Focused regional platforms stand a much better chance of getting it right consistently in our estimation and not fall foul of the law.</p><p><strong>Fourth</strong>, the industry and major regulators could agree (in the absence of a clear regulatory regime) minimum disclosures and behavioural criteria for promoters, exchanges and compliance protocols. A <a href="https://ag.ny.gov/sites/default/files/vmii_report.pdf">recent survey by the New York Attorney General</a> covering crypto currency exchanges contains a number of valid points and ideas along with this interesting observation (verbatim):</p><p><em>“…The OAG could not review the practices and procedures of non-participating platforms (Binance, Gate.io, Huobi, and Kraken) concerning manipulative or abusive trading. However, the Kraken platform’s public response is alarming. In announcing the company’s decision not to participate in the Initiative, Kraken declared that market manipulation “doesn’t matter to most crypto traders,” even while admitting that “scams are rampant” in the industry….”</em></p><p>We won’t get into the details of that report, but its main conclusions are common sense for experience traders: fair market practices, restrictions on employees and principals, disclosures of conflict of interests, market manipulation policy, disclosure of historical outages, insurance, custody and cybersecurity policy, etc.</p><p>Contact us at rodrigue.afota@blocktrust.com and alan.desira@blocktrust.com.com and stay tuned with the progress of <a href="http://www.blocktrust.com/">www.BlockTrust.com</a>, our latest project now under development.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a38a74ac27ce" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[On Public Blockchains: Scaling and Re-Centralisation]]></title>
            <link>https://medium.com/@Rodrigue.Afota/on-public-blockchains-scaling-and-re-centralisation-a309a58cf549?source=rss-ba3c60f60fd2------2</link>
            <guid isPermaLink="false">https://medium.com/p/a309a58cf549</guid>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[bitcoin]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[dapps]]></category>
            <dc:creator><![CDATA[Rodrigue Afota]]></dc:creator>
            <pubDate>Mon, 25 Jun 2018 13:24:31 GMT</pubDate>
            <atom:updated>2018-06-28T14:39:36.161Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/325/1*E49xSK3k7TB7sE7ddyr_Bw.jpeg" /></figure><p><strong>On Public Blockchains: Scaling and Re-Centralisation</strong></p><p>On July 14 this year, William Hinman, a senior official at the S.E.C., stated in a speech that the Ethereum public blockchain was probably not a security. Hinman mentioned the large decentralisation of the blockchain as a key factor, perhaps a warning for younger and more centralised “platform” blockchains such as E.O.S. entering into the space.</p><p>The announcement was welcome by the crypto industry: the stakes were high, with <a href="https://www.coindesk.com/6-3-billion-2018-ico-funding-already-outpaced-2017/">over US$6 billion of ICO fund-raising year to date</a> (the vast majority raised through the Ethereum platform) and a lot of Capex invested into developing “DAPPS” (decentralised applications) running on the Ethereum Virtual Machine (the global network of computers running the Ethereum protocol software). There are now over <a href="https://www.stateofthedapps.com/stats">1600+ DAPPS built or underway</a> and a reclassification under US securities law might have caused havoc.</p><p>Whilst bitcoin can be seen as a pure and vastly decentralised protocol with currently an estimated 100,000+ validating nodes, its software offers limited tools for DAPPS developers, mostly in the field of security, trust and payments. Ethereum was, from the start, an attempt at expanding the bitcoin experiment into a global platform, capable of running a wide range of applications and using its native cryptocurrency (Ether) to regulate access and usage of its network. Ethereum has been so far a resounding success, attracting investments and interest from a range of industries and governments, with applications developed in gaming, fund raising, exchanges, collectibles and public records. The vast majority of I.C.O. (initial coin offerings) today operate by issuing a smart contract (called ERC20 tokens, which are fungible) on the Ethereum blockchain, while games and collectibles use a non fungible type of smart contract called ERC721 and essentially operate by exchanging data against payments in Ether. The most well known DAPP so far, <a href="https://en.wikipedia.org/wiki/Cryptokitties">CryptoKitties</a>, operate by trading and breeding digital cats, and some users have paid in excess of US$100,000 for a single digital cat! This might seem to be an inefficient use of an expensive network, but it’s undeniably a proof of the utilitarian value of the concept, and collecting digital cats is no more absurd than collecting physical pokemon cards.</p><p>Nevertheless, CryptoKitties embodies the challenge faced by the Ethereum platform. It was launched in December 2017 and nearly brought the blockchain to a halt, using at times about 25% of its computational power and severely stressing the ability of validating nodes to remain synchronised at all times.</p><p>Deep down, the core issue faced by Ethereum (and other ambitious DAPPS blockchain platforms) boils down to the following question: can public blockchains scale and remain decentralised? At first, one has to be skeptical: it will always be much harder to synchronise in near real time thousands of computers and servers than just one or two. It’s 101 logic and this is about a lot more than just processors capacity or individual digital storage capacity. It’s also about bandwidth and network latency, and this is where the comparison between Bitcoin and Ethereum becomes telling: a bitcoin transaction essentially consists of the sender’s digital signature, his public key, a hash of the recipient public key and an amount of bitcoin being transferred. A CryptoKitty transaction is all that plus a digital image in binary code: it’s far more computationally intensive. This trend is reflected in the graph below, charting the average size of newly mined blocks and illustrating the massive increase in data processing faced by validating nodes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FU9L0iFS8DnM3fHpMIj4ng.jpeg" /></figure><p>Why is that so important? Because a very diversified pool of independent fully validating nodes is core to the whole blockchain value proposition: trust-less, censorship resistant and single fault tolerant. In the long run, if users come to see a public blockchain as centralised, and putting aside sovereign crypto currencies, the value of its native cryptocurrency will eventually fall to zero. As for Ethereum, it means the pressure of the growth of the ecosystem must be solved for the network to be healthy. Its ledger is now approaching (and maybe past) the 1 terabyte mark, well in excess of the bitcoin ledger. A situation where independent validating nodes keep dropping out is not a viable one.</p><p>The Ethereum core development knows all this of course, but it has to find solutions to problems that were never anticipated and it has to face enormous expectations from third party investors, businesses and developers who have staked a lot on the ability of the network to successfully grow.</p><p>What are the solutions? There are two competing schools of thoughts.</p><p>The first school believes that blockchains don’t really scale and need to “regulate” their growth in order to ensure strong decentralisation as measured by the number of fully validating nodes. Bitcoin is the best example: its flow is regulated by a cap on individual block size (under the SegWit protocol update, it’s actually a bit more complicated) and a pricing charge per data. Ethereum is not capping individual blocksize (which are experiencing exponential growth) and it relies mainly on “gas limit”, a concept capturing the maximum amount of processing cost for all transactions in a block combined. The idea is similar to the maximum block size in Bitcoin (measured in bytes), with a purpose to keep block propagation and processing time low, thereby allowing for a sufficiently decentralized network. In contrast to Bitcoin however, this limit can and has been increased by block producing miners, it’s not constant.</p><p>We have tried to compile reliable data on the the number of fully validating nodes, the best indicator for the true state of decentralisation for Ethereum, but, curiously, have not found any. What we have found however is repeat circumstantial evidence that less powerful nodes are gradually downgraded to light nodes, only verifying block headers but not all transactions, in effect having to trust other nodes. At current growth rates, whether due to network latency, inconsistent internet connections or the processing demands of the network, <a href="https://gist.github.com/Quiark/a3a2981fd4a21a1d5b95">most nodes will eventually fail to sync</a>.</p><p>If this trend remains unchecked, is the future of Ethereum one that is really dominated by a finite number of super nodes, akin to the newer EOS network? Many respected developers have suggested Ethereum will eventually have to have a block size limit if it is to survive and remain decentralised. And it will have to manage some anger from the DAPPS community in the process, with DAPPS substantially slowed down, not working and / or facing more expensive gas prices. This is the Ethereum conundrum.</p><p>The chart below, from the <a href="https://twitter.com/StopAndDecrypt">StoPandaCrypt</a> developer and blogger, illustrates this point schematically.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MGLGl-jqdPgyAVA5e-hOhA.jpeg" /></figure><p>The team behind EOS, the highest profile new platform blockchain, fully recognises the limitations of trying to both run a decentralised blockchain and offer fast processing time: it only has 21 super nodes (data centres really), connected by high capacity wires. It’s not decentralised and it’s a question mark whether its native currency will have a long term adoption or fail. Without a trusted native currency to regulate the data flow into the blockchain, that blockchain has no future as a mass market DAPP platform, so building trust in its native currency is key to its success.</p><p>For now, the industry is watching. <a href="https://finance.yahoo.com/news/eos-locked-7-accounts-implications-080014369.html?guccounter=1">EOS got off to a rocky start</a> when it recently migrated its currency out of the Ethereum platform and suffered an embarrassing hack in its first week. Its governance infrastructure was then questioned when it decided to freeze hacked accounts. When I get change at the supermarket after buying some goods, and it turns out the actual bills were used by a customer who was using stolen money, I trust 100% that nobody is going to come after me. No way I could ever trust a currency when this hypothesis is called into question.</p><p>But the EOS founding team has a point: maybe truly public blockchains just don’t scale, at least not to the extent needed to live up to “Web 3.0” expectations. The coming years will show us the kind of mass market DAPPS and public blockchains that are actually working and flourishing, while private blockchains will by construction remain limited in their purpose or mostly used as corporate data tools.</p><p>There is a second school of thought for how to scale, with a more pragmatic bias perhaps, headed by Ethereum charismatic leader, Vitalik Buterin.</p><p>Fully recognising the collection of challenges embodied in the word “scaling”, the Ethereum foundation has encouraged and is spearheading a number of new scaling techniques, some of them actually similar to the lightning protocol in Bitcoin. These techniques have yet to reach technical viability and go through extended peer testing but they have attracted a lot of interest in the industry, and can be classified in two categories.</p><p>Layer 1 techniques, such as <a href="https://www.ethnews.com/buterin-lays-out-ethereums-next-3-5-years-explains-sharding">“sharding”</a> are implemented at the base protocol level and, if integrated, would substantially impact how blockchain consensus is achieved among independent nodes. “Sharding” works by subdividing the amount of processing and validating work performed by the nodes, so that not all nodes need to verify everything. In effect, groups of nodes would have to trust each other, and the outcome would be a magnified processing capability for the network. On the other hand, It would also re-introduce the need for trust in the system, a violation of a core Satoshi principle.</p><p>Layer 2 techniques, such as <a href="http://www.plasma.io/">plasma</a>, <a href="http://www.jeffcoleman.ca/state-channels/">state channels</a> or <a href="https://truebit.io/">truebit</a>, are more akin to bitcoin’s lightning (this is a simplification), by pushing some of the activity off the main blockchain before, ultimately, going back into the blockchain for validation. The idea behind is not new: crypto exchanges already perform most of their clients buying and selling offchain by maintaining and modifying their own ledger, thereby reducing the overload of small transactions on public blockchains. There are many tech firms focused on developing these techniques. In our opinion, layer 2 techniques are broadly consistent with the core principles of public blockchains.</p><p>Will Ethereum succeed and build a high capacity blockchain that remain broadly decentralised? This is a tricky question. Using Vitalik Buterin’s words, the long-term goal is for the platform to be able to process transactions at “Visa-scale transaction levels” or beyond. Can Ethereum live up to this goal without introducing hard caps on block size?</p><p>Contact us at rodrigue.afota@blocktrust.com and alan.desira@blocktrust.com.com and stay tuned with the progress of BlockTrust.com, our latest project now under development.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a309a58cf549" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Privacy Challenge to Bitcoin and Crypto Currencies]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@Rodrigue.Afota/the-privacy-challenge-to-bitcoin-and-crypto-currencies-9e03b2b47fe?source=rss-ba3c60f60fd2------2"><img src="https://cdn-images-1.medium.com/max/600/1*K6LHRXdj317sKkslu_jXlQ.png" width="600"></a></p><p class="medium-feed-snippet">On 20 January 2016, international enforcement authorities arrested 10 men in a coordinated raid targeting an illegal online drug operation&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@Rodrigue.Afota/the-privacy-challenge-to-bitcoin-and-crypto-currencies-9e03b2b47fe?source=rss-ba3c60f60fd2------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@Rodrigue.Afota/the-privacy-challenge-to-bitcoin-and-crypto-currencies-9e03b2b47fe?source=rss-ba3c60f60fd2------2</link>
            <guid isPermaLink="false">https://medium.com/p/9e03b2b47fe</guid>
            <category><![CDATA[regulation]]></category>
            <category><![CDATA[privacy]]></category>
            <category><![CDATA[bitcoin]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <dc:creator><![CDATA[Rodrigue Afota]]></dc:creator>
            <pubDate>Thu, 26 Apr 2018 12:47:19 GMT</pubDate>
            <atom:updated>2018-04-26T20:46:50.045Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[RoadMaps for Central Banks Digital Currencies]]></title>
            <link>https://medium.com/@Rodrigue.Afota/roadmaps-for-central-banks-digital-currencies-5da8f961b149?source=rss-ba3c60f60fd2------2</link>
            <guid isPermaLink="false">https://medium.com/p/5da8f961b149</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[bitcoin]]></category>
            <category><![CDATA[central-bank]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[regulation]]></category>
            <dc:creator><![CDATA[Rodrigue Afota]]></dc:creator>
            <pubDate>Mon, 05 Feb 2018 14:47:52 GMT</pubDate>
            <atom:updated>2018-02-06T12:57:40.118Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>RoadMaps for Central Banks Digital Currencies</strong></p><p>February 5th, 2018</p><p>China, Canada, Sweden, Russia, Korea, Estonia, Tunisia, Poland, Japan, and others… What these countries have in common is an interest in issuing or backing a crypto currency, perhaps capitalizing on the hype generated by public blockchains such as bitcoin or zcash.</p><p>Even the Bank of International Settlement got involved, producing last year a report on crypto (<a href="http://bis/">https://www.bis.org/publ/qtrpdf/r_qt1709f.pdf</a>) with this very intriguing diagram.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/400/1*3VoKeApaikzE2FtRmYwrzw.jpeg" /></figure><p>These are early days: a still evolving blockchain technology plus a history of security hacks into wallets and exchanges suggest to us that Central Banks and regulatory bodies are likely to proceed carefully, since sovereign backed crypto currencies will share many of the same challenges as public blockchains.</p><p>Security of digital assets is a key concern. Digital currencies don’t just rely on a network of servers to power a blockchain, they also rely on an ecosystem of point of entries such as software clients, digital exchanges and wallets providers. The oldest public blockchain network, bitcoin, has never been hacked, yet its point of entries have been hacked many times. As a first building block of issuing a Central bank backed digital asset, it stands to reason that the crypto tech ecosystem required to support that asset will have to operate under a regulated framework and undergo a massive shift in security culture and regulatory compliance. Old style conservative banking culture mixed with young techies will be far more effective in the long run.</p><p>Just over the past 6 months, we had major hacks in a number of exchanges (Coincheck), repeat loss of funds in one of the main Ethereum client, Parity, hacks in a number of online wallets, spread across various popular coins such as IOTA, ETH and NEM, and fears over a number of popular digital exchanges’ ability to support a massive increase in customers activity. Of course, traditional banks also get hacked occasionally, but at least, as regulated entities, they have to have cyber security policies in place, operational and financial audits, their licenses can be suspended or restricted and they can receive massive fines.</p><p>Assuming a regulatory licensing regime is in place, then the real strategic questions arise for governments and central banks authorities. Before rushing into a test mode with a blockchain partner, decision makers shall carefully review the economic impact of seemingly innocuous technological choices. Among the most important one are:</p><ul><li>Consensus mechanism: How do we ensure (or, more precisely, maximize the probability) that the network will always manage to produce genuine new blocks accepted by all the nodes? It’s a critical question: what would happen to a Central Bank credibility if, overnight, consensus new blocks are fraudulent or frivolous and e-depositors accounts are now empty? Consensus algorithm can be «trustless», such as bitcoin, Ether or Zcash’s proof of work («POW»), or rely on a faster protocole such as proof of stake («POS») or practical byzantine fault tolerant («PBFT») algorithms such as the one used by Ripple. While the main drawback of POW is well known, energy consumption, POS and PBFT all rely somehow on validators behaving rationally and being trusted to not collude. There is just not enough history and years of testing here and POS and PBFT algorithms appear to share some of same drawbacks as centralized ledgers.</li><li>Convertibility: Will central banks guarantee the convertibility of an e-coin at par with fiat? Alternatively, authorities could take a more restricted approach, such as allowing or requiring only certain payments to be redeemed in e-coins, such as traffic fines or tolls.</li><li>Restricted or unrestricted supply of e-coin? In other words, shall depositors be able at will to shift all their banking deposits into their e-wallets and vice versa? There is a trade-off here. Restricting the supply of e-coin will substantially reduce the impact on banks liquidity in case of banking crisis, as depositors would not be able to shift all their banking balances out of banks. On the other hand, a restricted supply will all but guarantee that the e-coin will trade at a basis with fiat. An unrestricted supply will solve that, but it will inevitably reduce commercial banks’ grip on fiat and their ability to act as allocator of capital over time, being replaced by new Fin Tech business models.</li><li>Private, «permissioned» or public blockchain network? Central banks may decide to initially run and test a blockchain with only a few nodes, the existing regulated banks. That would not amount to a real game changer, there are cheaper technologies than blockchains for that, and such a small network may in fact be less fault tolerant than a centralized one. A few dishonest employees could collude and hijack the blockchain, making a mess of banking records across banks. We believe central banks will converge towards a far more diverse network of nodes, including many new economy point of entries, such as regulated wallets, exchanges or large corporate. Ultimately, the higher the number of network nodes, the greater the security of the blockchain</li><li>Traceability versus anonymity: Central banks will likely opt for traceable networks, such as bitcoin (and unlike Monero or Zcash), with the ability to track down ultimate owners through regulated point of entries. Central banks E-coins will be far harder to use than cash for money laundering</li></ul><p>We reviewed the major public policy implications of issuing Central Banks and identified the most interesting one.</p><p>Reduce reliance on cash in the economy, as in Sweden or Korea. Reducing reliance on cash saves a lot of money to the economy, from fewer ATM, smaller back offices, simpler check-outs at the supermarket, to a smaller black market economy and higher reportable taxable incomes.</p><p>Allow Central Banks to bypass commercial banks in the implementation of its monetary policy. This is a particularly relevant concern whenever a banking system is unable to satisfactorily pass on Central Banks’ decisions onto end users. In the long run, allowing displacement of commercial banks will create new challenges for a central bank: commercial banks run on a regulated system of fractional reserves and share with the central bank the ability to create liquidity and add or reduce leverage in the economy. Initially, central banks-backed e-coins won’t, at least until a proper regulated eco system can emerge.</p><p>Managing counterparty and liquidity risks: E-coins can be held in a manner not exposing the holder to counterparty risk, a useful tool for banks in the era of Basel III. It also allow banks to shift undesirable non-operational customer deposits onto a third party, freeing up balance sheet capacity in the process.</p><p>Under-banked economies. This is particularly true in parts of Africa or South East Asia. In Tunisia, roughly 40% of the population does not have a bank account. However, mobile phone penetration is 90%+.</p><p>General purpose funding or project finance funding alternatives, as in Estonia. Regardless of what one thinks about ICO, it’s now clear that there is a new funding tool that public authority are soon going to explore to fund public projects. In theory, a government can issue a token through an ICO, with a given supply and convertibility and powered by a blockchain, and mandate that from a given date in the future, these tokens have to be used by users of certain public infrastructures, such as housing, tax, traffic fines, pollution rights or tolls. The funding mechanism may at times be more attractive than issuing public debt or borrowing from banks, and it will serve a strategic public policy goal, encouraging the development of a tech ecosystem required to support these tokens. These tokens will also be used as a safe storage of value whenever concerns of a run on banks are on the rise.</p><p>We believe Central banks backed e-coins will co-exist with the most successful public blockchains. They serve a different purpose and, by construction, their governance is centralized. Like public blockchains though, they have the potential to truly disrupt commercial banking as we know it, allowing users to bypass banks altogether over time. We will keep monitoring developments in this area.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5da8f961b149" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Ripple Paradox]]></title>
            <link>https://medium.com/@Rodrigue.Afota/the-ripple-paradox-1a48582e4629?source=rss-ba3c60f60fd2------2</link>
            <guid isPermaLink="false">https://medium.com/p/1a48582e4629</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[ripple]]></category>
            <category><![CDATA[banking]]></category>
            <category><![CDATA[bitcoin]]></category>
            <dc:creator><![CDATA[Rodrigue Afota]]></dc:creator>
            <pubDate>Sun, 14 Jan 2018 17:21:43 GMT</pubDate>
            <atom:updated>2018-01-15T06:43:54.285Z</atom:updated>
            <content:encoded><![CDATA[<p>In the space of a month from December 4 to January 4, XRP (the acronym for the crypto currency powered by the Ripple blockchain) market capitalisation and US$ price went up x14, before retreating back to x7. There was no specific logic to such price increase and I have felt compelled to look into the Ripple story when multiple friends, new to crypto, where enthusiastically jumping into XRP.</p><p>Ripple Labs is the company behind Ripple Net, a decentralised semi-public blockchain network, running the xCurrent software protocol. Ripple Net and xCurrent are the products that Ripple is actually pitching to banks throughout the world with some success (over 100+ banks signed up globally), getting them to test these products. As of now, these products are essentially a near real time messaging service confirming cross border interbank payments and represent a clear emerging competition for the Swift messaging service. This is where the hype is coming from: Swift is a form of co-operative representing 11,000+ banks and processing over 15 million messages every day, in effect powering most of the global FX market. What if Ripple could disrupt all of that?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RtkcKdQ2sbiQwcAks9W4HA.png" /></figure><p>Ripple products are actually not displacing the Nostro / Vostro model of correspondant banking, a legacy and costly framework under which, say, a European bank without a US banking license needs an account in a US bank to hold its US dollars, and vice versa. So when a US and a EU bank interact in the foreign exchange market, 4 banks and 4 back offices are usually involved. Because the Nostro / Vostro framework is unchanged, the Ripple’s disruption is relatively limited and middle men are still involved. Ripple would of course like to do more, it would like to be truly disruptive and, in the future, get banks or anyone to bypass costly Nostro/Vostro arrangements and instead have market participants use XRP. That is not what the product is doing, and, with Central Banks unlikely to let regulated banks hold crypto-currencies for a reasonable capital charge for the foreseeable future, we very much doubt this will happen anytime soon.</p><p>In effect Ripple Labs products and XRP are two very different things. Ripple Labs products could turn out to be a success with no or very little impact or utility role for XRP at all. The company Ripple could be a great success and XRP could tank.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/515/1*c9MHHq_sbnMbewyCrvt_-A.png" /></figure><p>It is a well documented fact that the correspondant banking model is expensive. There are widely spread misconceptions though that need to be debunked.</p><ul><li>Swift is actually not that inefficient: Swift processes millions of same day messages every day</li><li>It is the back end, banks back offices and treasury operations across multiple weakly coordinated entities, that is usually slowing things down and preventing an efficient and near real time management of Nostro and Vostro accounts</li><li>Swift has the clients and it has a strong incumbent advantage. It is currently testing its own decentralized ledger technology and the game is clearly Swift’s to lose because regulated banks are very careful and deliberately slow moving entities when it comes to handling assets and money. Blockchains are new, and Ripple’s practical byzantine fault tolerant consensus protocol, while therotically very hard to attack or game, is also theoretically open to attack</li></ul><p>Will Ripple succeed? It has to contend with Swift and a number of other institutional competitors. It is also facing strong competition from new players such as Stellar (powering the crypto-currency Lumens), a company that was created in 2014 and is now backed by IBM. Stellar code was initially similar to Ripple’s and subsequently forked its own software after it found flaws in the previously similar Ripple and Stellar code.</p><p><a href="https://bravenewcoin.com/news/why-the-stellar-forking-issue-does-not-affect-ripple/">https://bravenewcoin.com/news/why-the-stellar-forking-issue-does-not-affect-ripple/</a></p><p>We believe the only possible rationale for last month very rapid XRP price increase is if one strongly believe that banks will eventually move their FX transactions to more than just a blockchain powered messaging service.</p><p>One has to believe banks will get rid of the Nostro / Vostro correspondant banking model and settle through XRP. It might happen, but we doubt that, and it will take a lot and a very long time for that to happen. Today, Ripple’s core argument to get the banks to use XRP in a limited way is through transfer fees intended to help secure the blockchain against spam attacks: the new technology promoted by Swift and others do not use such native currencies, and are using alternative secure identification methods that Banks are far more likely to be familiar and comfortable with.</p><p>So what happens if Ripple software is a success but gets no traction with XRP (i.e. the software is still just a messaging service and does not use XRP)? Can XRP still prosper on a standalone basis as a credible alternative to pure protocol public blockchains, starting with Bitcoin or the likes of Zcash? That seems unlikely for a number of reason. The Bitcoin core community is incredibly conservative and intentionally slow moving, and has a perfect safety track record against system wide attacks. Its monetary base is fixed (22 millions BTC against 100 billion XRP) and its network is far more decentralised. Bitcoin proof of work consensus is expensive and slower, no doubt, but it can always fork into faster variants such as Bitcoin Cash and, probably this year, adopt the critical SegWit upgrade and Lightning off-chain capability, in which case transfer fees and confirmation time will be substantially reduced . On the other hand, XRP has the obvious disadvantage that Ripple’s insiders are controlling 60% of the XRP supply!</p><p>XRP could still be a success and we know many knowledgeable believers. Our money is on Bitcoin though.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1a48582e4629" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[On Public and Private Blockchains]]></title>
            <link>https://medium.com/@Rodrigue.Afota/on-public-and-private-blockchains-1b1cdf0e9c2b?source=rss-ba3c60f60fd2------2</link>
            <guid isPermaLink="false">https://medium.com/p/1b1cdf0e9c2b</guid>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[bitcoin]]></category>
            <category><![CDATA[regulation]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Rodrigue Afota]]></dc:creator>
            <pubDate>Thu, 04 Jan 2018 13:28:39 GMT</pubDate>
            <atom:updated>2018-01-04T13:28:39.717Z</atom:updated>
            <content:encoded><![CDATA[<p>January 4th, 2018.</p><p>Contrasting with open-access public blockchains (where anybody can download a copy of the open source software and act as a node or a validator and maintain a copy of the ledger), private and consortium blockchains are built on two different pillars.</p><ol><li>Governance is privatized to an entity or group of entities</li><li>Read and/or Write access on the blockchain ledger are restricted</li></ol><p>In our previous newsletter (“The Blockchain Governance Trade-Off”), we made the point that the word “governance” was a misnomer for public blockchains. In Bitcoin, there is no governance. There is instead a consensus among miners, nodes and developers, slowly evolving the software protocol in order to adapt, maintain those crucial network effects and, most important of all, ensure the integrity of the network against any attack, in the process building up trust globally. That the bitcoin developers were willing to let go of their control of the software back in 2009 was a radical decision, and, with the exception of the semi-public Ripple protocol, all public blockchains including Ethereum have let go of centralized governance. Even if we now have foundations gently promoting but not imposing upgrades, these foundation proposals can be tainted or suspected of favoring the insiders. Even the Ethereum foundation faced much resistance when it tried to promote the DAO hard fork, a software upgrade intended to roll back a portion of its blockchain history in order to un-do a hack caused by weaknesses into the DAO code, in the process jeopardizing the core idea of transaction irreversibility. That proposal led to the Ethereum Classic hard fork, and, when the foundation tried to un-do the Parity hack in 2017, it had to give up.</p><p>Governance is tricky, if anything because any software update relying on cryptographic protocols needs an extended and intensive peer-review process, debate, testing and an openness to not go ahead. That is not the way a centralized body is likely to run things.</p><p>Bitcoin was also an exercise in radical transparency. The entire blockchain history is open to everyone and transactions are entirely traceable.</p><p>Financial institutions recognized early the potential for blockchains to provide the technology for upgraded, more secure and decentralized shared database, speeding up and lowering the cost of a number of inter-bank transactions. But they did not like the open-access framework, at odds with the need to enforce data protection, a key concern for regulated entities. Hence they have promoted the use of private or consortium blockchains.</p><p>In a private blockchain, Read/Write access and software protocol governance are controlled by a private entity or a group of entities. The network will be smaller, less decentralized, but it will be faster and cheaper. It won’t benefit from large network effects though and, in the long run, it’s questionable whether their own crypto-currency can be trusted enough by market participants to reduce or eliminate the counterparty risk inherent in any digital contract or confirmation. We believe that most sovereign crypto-currencies will operate through some private blockchain (operated by a Central Bank) or consortium blockchains (operated perhaps by a group of banks and other approved regulated entities), and we will track developments in this area. Other uses of private blockchains include medical records, government records and real estate ownership.</p><p>There is an on-going debate though about whether private blockchains are worth the hype, and indeed private investment in that area appears to have stalled in 2017. The basic reason lies in whether private blockchains are superior to secured shared database when factoring in the hassle. As banks and their IT officer get more knowledgeable, they also become more aware of the issues involved in operating shared ledgers with other entities and managing a common protocol.</p><p>There is also a lack of track record for consensus mechanisms other than Bitcoin’s proof of work (“POW”). Whilst the drawback of POW is well known (energy usage), its network itself has been resilient and never been hacked since 2009. It just does not rely on trusting any participant and is sometimes referred to as a low trust blockchain (i.e. there is no need to trust anybody). Private blockchains are higher trust networks: they are far less decentralized and rely on trusting the behaviors of nodes or validators to choose not to collude. This will be the subject of a separate newsletter. These are early years though and we need more years of data and track record to understand which blockchain technology can withstand the passage of time successfully.</p><p>Ravelin Finance Consulting and Analytics Ltd</p><p>Financial and blockchain technologies for Institutions. info@ravelinfinance.com</p><p>971 2 418 7566</p><p>+971 56 992 1349</p><p>We are licensed and registered by Abu Dhabi Global Markets. Registration 0000633</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1b1cdf0e9c2b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Blockchain Governance Trade-off]]></title>
            <link>https://medium.com/@Rodrigue.Afota/the-blockchain-governance-trade-off-6ab52b01d061?source=rss-ba3c60f60fd2------2</link>
            <guid isPermaLink="false">https://medium.com/p/6ab52b01d061</guid>
            <category><![CDATA[banking]]></category>
            <category><![CDATA[regulatory-compliance]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[bitcoin]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <dc:creator><![CDATA[Rodrigue Afota]]></dc:creator>
            <pubDate>Fri, 29 Dec 2017 11:28:19 GMT</pubDate>
            <atom:updated>2017-12-29T11:28:19.342Z</atom:updated>
            <content:encoded><![CDATA[<p>December 21st, 2017.</p><p>On August 1st this year, Bitcoin, the best known public blockchain, split and forked into into two different coins, Bitcoin Cash and Bitcoin, sharing their Blockchain history up to the split. The split, otherwise known as a “hard fork”, was the result of the ongoing scalability debate.</p><p>The scalability problem is essentially a confrontation between those who primarily want to further Bitcoin appeal as a transactional currency, cutting trade confirmation time and improving efficiency. On the other side of the debate are the purists and libertarians, who’d rather err on the side of conservatism and caution, maintain relatively low barriers to entry to set up a node and in the process strengthening bitcoin status as a digital gold.</p><p>There is no denying that the bitcoin network is facing challenges due to its success. The chart below is a common metrics used to measure the backlog of unconfirmed trades.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9efrV9a0rzlOYDl9XuGweg.jpeg" /></figure><p>The Bitcoin Cash split was not consensual: a subset of the community decided to go ahead anyway and adapt their own modified version of the software protocol and hope to entice enough nodes and miners to join in. Holders of Bitcoins (those in full possession of their private keys) ended up owning two different coins. The scalability debate is also at the heart of the SegWit soft fork (more on this below) as well as the cancelled SegWit2x hard fork. And that debate was also at the heart of past forks, Bitcoin XT, Bitcoin Classic, Bitcoin Unlimited.</p><p>It should seem scary to a wannabe investors, and yet, bitcoin survived and, in a manner of speaking, came out stronger. In a way, it should not really matter, a hard fork would result into two different coins more or less replicating the value of the legacy blockchain. Even then, there really is very little anyone or any participant can do to prevent a fork.</p><p>This is because blockchains are run by decentralized nodes (you and me and many others) choosing to run any version of the protocol they want. There is little the founder of a blockchain protocol can do: Once the protocol is out, anybody can promote an “upgrade” in the protocol.</p><p>And, seen from that light, any successful digital coin can and should expect many lawful “hijacking” attempts.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/604/1*vwt-Wt4ldkcewfL01OSF0Q.jpeg" /></figure><p>The diagram above, courtesy of bitcoin.org, is schematically what happens in a hard fork. Which begs the question, how many nodes and miners are needed to back a protocol upgrade? What are the rules? Who decides?</p><p>We tried hard to answer these questions, and there really is no clear answer. Let us state flat out that there is no formalized governance for public blockchains (private blockchains have their own issues, to be sure). It is a misunderstanding to refer to “oversight”, there is not even a clear concept of majority rule: even assuming 95% of miners (creating blocks) and nodes (verifying consistency with rules and blockchain history) would agree or not on an upgrade, the dissenting 5% can break away. Of course, their small 5% network will probably decline and benefit from far smaller networks effects.</p><p>That is why the community refers to collective consensus, an amorphous concept intended to capture the willful cooperation of many core developers, miners and nodes. The fear of damaging existing powerful network effects working to the benefit of the community at large is a powerful force, placing greater value in unity than in being individually right. And that fear was perhaps the reason SegWit2X was cancelled. There was no break-away community this time. Away from Bitcoin, even Vitalik, the informal leader of Ethereum, could not convince the community to execute a hard fork that was intended to roll back a hack incurred by Parity, an Ethereum client.</p><p>Indeed, we would be remiss in not mentioning that there is usually a darker story going on. Behind proposed forks, we have often found out after the facts problems in the upgraded code, some of them intentional and opening the door to back door hacks or hidden attribution of coins.</p><p>In the case of Ethereum main selling point, the ability to execute third party smart contracts, third party code was found to be the cause for a number of hacks in 2016 and 2017. Vitalik successfully promoted a controversial hard fork roll-back in 2016, essentially erasing a small and recent portion of Ether transaction history, but, in 2017, the community ignored him. Roll-backs are controversial and can appear to conflict with the core principle of immutability, benefiting a few and rewarding the reckless behavior of developers rushing out new code.</p><p>Which leads us back to Bitcoin extreme cautiousness: we cannot not wonder whether maybe that’s the right way to go for digital gold. After all, blockchains are cryptographic-based protocols that do require intensive and extended peer reviews over a long period of time. Maybe moving too fast is actually not the way to go. The link below will refers to a vulnerability found relatively easily in IOTA by a MIT researcher this year.</p><p><a href="https://medium.com/@neha/cryptographic-vulnerabilities-in-iota-9a6a9ddc4367">cryptographic vulnerabilities in IOTA</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6ab52b01d061" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>