<?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 Chenxi Cai on Medium]]></title>
        <description><![CDATA[Stories by Chenxi Cai on Medium]]></description>
        <link>https://medium.com/@chenxicai1?source=rss-9bdad0704145------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*YmikDBrNDNj6nimFva9UJg.jpeg</url>
            <title>Stories by Chenxi Cai on Medium</title>
            <link>https://medium.com/@chenxicai1?source=rss-9bdad0704145------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 04:33:05 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@chenxicai1/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[Will cryptocurrencies ever reach mainstream adoption? Does it even need to?]]></title>
            <link>https://medium.com/@chenxicai1/will-cryptocurrencies-ever-reach-mainstream-adoption-does-it-even-need-to-861b34fc5ff1?source=rss-9bdad0704145------2</link>
            <guid isPermaLink="false">https://medium.com/p/861b34fc5ff1</guid>
            <category><![CDATA[investing]]></category>
            <category><![CDATA[bitcoin]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <dc:creator><![CDATA[Chenxi Cai]]></dc:creator>
            <pubDate>Wed, 07 Aug 2019 09:05:24 GMT</pubDate>
            <atom:updated>2019-08-07T09:05:24.883Z</atom:updated>
            <content:encoded><![CDATA[<p>We always hear about people in the crypto world complain about the lack of “mainstream adoption”. Less than 1% of the world owns cryptos. Most stores still don’t accept them. It seems like we have a long way to go in terms of adoption.</p><p>However, how many of the world population own other forms of assets? It would be unfair to compare cryptos to smartphones or computers, since it is not a piece of consumer hardware. We should be better advised to compare it to other forms of financial assets.</p><p>Stocks: well-adopted. It is estimated that over half of all Americans own stocks or mutual funds according to Gallup surveys.</p><p>Bonds: not so well. Only 1.3% of American households directly own bonds, according to a study by the St. Louis Fed in 2016.</p><p>Precious metals: not so well. No good figures available, but typical estimates by experts are that only 1%-3% of the population own precious metals.</p><p>Home equity: well-adopted. The Census Bureau puts the rate of US home ownership at 64% in 2018.</p><p>It seems that of all major asset classes, stocks and home equity are well-adopted, while bonds and precious metals barely reach “mainstream adoption”. It is therefore no accident that the former assets are much more responsive to economic cycles than the latter ones.</p><p>What does this mean for bitcoin? Note that bitcoin, as it currently stands, is presented as a digital alternative to gold. Therefore, its peak adoption might not ever break through the ceiling of current ownership numbers for gold.</p><p>While it may sound pessimistic, it may not be a bad thing after all. It means that we might not need after all find a way to get grandmas to learn about secure private key management practices, which seems pretty improbable to me.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=861b34fc5ff1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A Complete Survey of Consensus Protocols]]></title>
            <link>https://medium.com/@chenxicai1/a-complete-survey-of-consensus-protocols-ec9c4988619e?source=rss-9bdad0704145------2</link>
            <guid isPermaLink="false">https://medium.com/p/ec9c4988619e</guid>
            <category><![CDATA[consensus]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Chenxi Cai]]></dc:creator>
            <pubDate>Thu, 18 Oct 2018 00:17:27 GMT</pubDate>
            <atom:updated>2018-10-18T00:39:32.941Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/660/1*q10gLJihRT5ymQpfEOLpBQ.jpeg" /></figure><p>The most amount of confusion in the blockchain world surrounds consensus protocols. Partly, this is because academics traditionally only focused on the BFT families. As a result, quality research on other modes of consensus has been in short supply.</p><p>The departing point of classifying consensus protocols traditionally starts with the so-called CAP theorem. The CAP theorem states that there is a fundamental trade off between consistency, availability and partition tolerance. Under this framework, BFT families favor consistency at the expense of availability, and blockchain families favor availability at the expense of consistency. <strong>However, contrary to popular perception, </strong><a href="http://blog.franslundberg.com/2013/12/the-cap-theorem-is-not-theorem.html"><strong>the CAP theorem is not a theorem, but rather a conjecture</strong></a>. Therefore, it is actually still possible to build systems that satisfy all three properties.</p><p>The first blockchain consensus, Nakamoto consensus, combines the longest chain rule with Proof of Work. <a href="https://eprint.iacr.org/2017/775.pdf">It has been proved that PoW provides the security guarantee without relying on a random oracle. </a>However, economies of scale dictates that PoW mining power must be concentrated, which negates the decentralization promise.</p><p>Alternatives blockchain consensus such as Proof-of-Stake, Proof-of-Space-Time has been explored to reduce mining centralization. However, it has been proved that <a href="https://arxiv.org/pdf/1809.06528.pdf">previous Proof-of-Stake design using deterministic mining under the longest chain were all vulnerable to an adaptive adversary</a>. Thus far, three possible alternatives has emerged to address this problem. (1) Using an unbiasable randomness beacon at the start of each round, e.g. <a href="https://dfinity.org/pdf-viewer/pdfs/viewer?file=../library/dfinity-consensus.pdf">Dfinity</a>, <a href="https://ethresear.ch/t/minimal-vdf-randomness-beacon/3566">Ethereum (proposing)</a> (2) Using a modified longest-chain rule, e.g. <a href="https://eprint.iacr.org/2018/378.pdf">Ouroboros Genesis </a>(3) Using adaptive honest miners to combat adaptive adversaries, e.g. <a href="https://eprint.iacr.org/2017/656.pdf">iChing</a>. Proof-of-Space-Time appears to face a similar issue, and current proposals also require an external randomness beacon.</p><p>The BFT protocols, traditionally hindered by its message complexity and known participation list, have experienced a revival thanks to the discovery that <a href="https://eprint.iacr.org/2017/913.pdf">it can be used as an accelerating overlay on top of blockchain protocols</a> under optimistic conditions.</p><p>There is also a class of “gossip-of-gossip” BFT protocols which attempt to reduce message complexity through recursive communication. However, this class of protocols still mandate a known participation list, and it is not clear how to implement anti-Sybil measures.</p><p>For more discussion on Consensus Protocol Design, join our telegram group</p><p><a href="https://t.me/joinchat/GHNsv0WnGt7ayOe_4HO0BQ">https://t.me/joinchat/GHNsv0WnGt7ayOe_4HO0BQ</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ec9c4988619e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to implement zk-SNARK on Ethereum gracefully]]></title>
            <link>https://medium.com/@chenxicai1/how-to-implement-zk-snark-on-ethereum-gracefully-bdac7f32b48f?source=rss-9bdad0704145------2</link>
            <guid isPermaLink="false">https://medium.com/p/bdac7f32b48f</guid>
            <category><![CDATA[zksnark]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[ethereum]]></category>
            <dc:creator><![CDATA[Chenxi Cai]]></dc:creator>
            <pubDate>Tue, 02 Oct 2018 04:45:15 GMT</pubDate>
            <atom:updated>2018-10-02T04:45:15.478Z</atom:updated>
            <content:encoded><![CDATA[<p>zk-SNARK is an awesome magician’s hat where many rabbits can be pulled out. <a href="https://ethresear.ch/t/on-chain-scaling-to-potentially-500-tx-sec-through-mass-tx-validation/3477/21">Scaling the blockchain</a>, <a href="https://codaprotocol.com/testnet.html#item-4">compressing the blockchain storage</a>, <a href="https://blog.ethereum.org/2017/01/19/update-integrating-zcash-ethereum/">issuing privacy tokens</a>, and etc are some of the top use cases. However, the actual implementation barrier remains high even for experienced developers due to the complex rituals involved. In this article, I review the process of generating and verifying zk-SNARK proofs along with some tools to simplify the process significantly.</p><p>First, we need to preprocess the program into “R1CS” (<em>Rank-1 Constraint Systems</em>), a language that is similar to arithmetic circuit satisfiability. This is important since R1CS is as unreadable as machine code and therefore extremely difficult to program in. Under the hood, this is often done by flattening the program into an AST tree and then encode the instruction sets one by one with intermediate variables inferred. <a href="https://github.com/JacobEberhardt/ZoKrates">Zokrates</a> contains source code that does exactly that for Python-like scripts. More recently, O(1) Labs released <a href="https://github.com/o1-labs/snarky">snarky</a>, a OCaml-based DSL that does the same job.</p><p>Once we have the R1C1 code, we run key generation on it and generate a proving key and a verification key. Zokrates handles it for us through libsnark. Under the hood, we do a number of things: (1) Represent the R1CS constraints by a set of polynomials over a finite group G using QAP (<em>Quadratic Arithmetic Programs</em>) (2) Compute the elliptical curve parings of the generated polynomials and a target polynomial evaluated at a secret point, as well as up to parings of the powers of the secret point up to certain millions. The pairings of the generated polynomials are used as proving key, and the pairings of the target polynomial and scale parameters is used as verification key.</p><p>A prover can then use the proving key and a private input (witness) to generate a proof of knowledge. Zokrates also handles it for us through libsnark. Under the hood, the prover plugs in the witness into the generated polynomials, and outputs the pairing of a updated version of proving key as a proof by taking a linear combination of the proving key.</p><p>Lastly, a verifier that has access to the verification key can validate the proof, and this can be done in a smart contract. Zokrates can directly export a verification contract, which takes both the proof and the public input and return a true or false value.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bdac7f32b48f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Public blockchain security ultimately relies on endowment effect in behavioral economics]]></title>
            <link>https://medium.com/@chenxicai1/public-blockchain-security-ultimately-relies-on-endowment-effect-in-behavioral-economics-2e630c13efc6?source=rss-9bdad0704145------2</link>
            <guid isPermaLink="false">https://medium.com/p/2e630c13efc6</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[behavioral-economics]]></category>
            <category><![CDATA[bitcoin]]></category>
            <dc:creator><![CDATA[Chenxi Cai]]></dc:creator>
            <pubDate>Sat, 15 Sep 2018 08:27:52 GMT</pubDate>
            <atom:updated>2018-09-15T08:27:52.898Z</atom:updated>
            <content:encoded><![CDATA[<p>It has become conventional wisdom in the blockchain circle that Proof of Work provides a Nash Equilibrium that incentives a majority mining power to secure the blockchain, and thereby securing the value of the network. Arguments arise when it comes to whether Proof of Stake provides the same security guarantee.</p><p>What I would like to point out is that this is only the cooperative Nash Equilibrium. There is also the non-cooperative Nash Equilibrium where an attacker convincingly commits to destroying the network, and miners give up mining on the blockchain, which results in a plummeting value of the network.</p><p>This non-cooperative equilibrium was pointed out in <a href="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2393940">a past article</a> before in the context of Proof of Stake. However, it is actually universal to all kinds of bootstrapped consensus mechanisms that does not have a “real asset” backing. There is nothing special to Proof of Work relative to Proof of Stake in terms of the existence of this equilibrium.</p><p>This leaves the question of why this non-cooperative equilibrium has not come into play so far, at least for Bitcoin. From my observation of miners’ behavior during a bear market, I hypothesize that endowment effect from behavioral economics might account for part of the story.</p><p>The endowment effect refers to the irrational tendency of people to hold onto their existing possessions as if they value them at a premium to market prices. For example, when one group of people are asked to choose between being given a pen or a mug, they would behave differently from when the same group of people are randomly assigned a pen or a mug and then asked to trade amongst themselves. Always, people do very few trades because of the endowment effect.</p><p>This seems to also be happening to Bitcoin miners on a large scale. When they acquire Bitcoins through mining, they tend to hold onto them rather than selling them immediately. This means that for an attacker to convince the miners to stop mining, the attacker would need to demonstrate a much higher commitment to destroy the network than she otherwise would if miners were rational economic agents.</p><p>For PoW versus PoS, it is worth asking whether the endowment effect is diminished if no “real” resources were expended during mining, so that the attacker costs to demonstrate a commitment becomes lower for PoS. Now it depends on another cognitive illusion called “sunk cost fallacy”. This is where people use forgone costs to measure the value of a current commodity. Note that this is also an irrational behavior as the cost of investing in mining gigs is “sunk” and, therefore irrelevant to current decision making about whether or not to continue to mine.</p><p>If the sunk cost fallacy is true in this case, it would mean that for PoW, since miners already invested in mining equipments, they will have a higher willingness to continue to mine for the network when an attack occurs, than for PoS, where mining is virtual and requires no illiquid fixed investments. Again, this would be a purely behavioral argument rather than a rational economic argument.</p><p>In conclusion, the security of blockchain consensus protocols ultimately relies on behavioral economics. This fact has been overlooked and deserves serious attention from people developing these protocols.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2e630c13efc6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My Key Takeaways From Crypto 2018’s Cryptocurrency Workshop]]></title>
            <link>https://medium.com/@chenxicai1/my-key-takeaways-from-crypto-2018s-cryptocurrency-workshop-a6968a8d3595?source=rss-9bdad0704145------2</link>
            <guid isPermaLink="false">https://medium.com/p/a6968a8d3595</guid>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[crypto]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Chenxi Cai]]></dc:creator>
            <pubDate>Mon, 20 Aug 2018 06:03:01 GMT</pubDate>
            <atom:updated>2018-08-20T06:03:01.814Z</atom:updated>
            <content:encoded><![CDATA[<p>Crypto 2018 is the annual conference for cryptographers worldwide. This year, for the first time, it added a two-day pre-conference workshop on cryptocurrencies. High profile speakers include <a href="https://medium.com/u/587a00dbce51">Vitalik Buterin</a>, Dawn Song of Oasis Labs, <a href="https://medium.com/u/ebd2a5a0fc">Silvio Micali</a> of Algorand, Rafael Pass of Thunder and many more. They not only offered detailed insights about their respective projects, but also provided exclusive scoop about new directions that they are taking on.</p><p>Vitalik revealed a couple of economic theory based proposals to select optimal block size as well as to determine transaction fees. I think these proposals are well researched and long overdue to be implemented.</p><p>It appears that one of the greatest consensus among attendees were that they don’t trust Intel’s SGX virtualization technology, which was proven to be vulnerable a few days ago. According to Mic Bowman from Intel, they will release a 2.0 version of the SGX technology, which hopefully will deliver more convincing security features.</p><p>In light of this, Dawn Song made a striking announcement that Oasis Labs will not rely on SGX technology, or any TEE anymore. Instead, they will be working toward using purely software based solution to achieve verifiable computing. The exact details are still to be released, however.</p><p>Silvio Micali presented on a new-and-improved Algorand 2.0 that reduced the number of communication rounds from 25 to just 6, which is an impressive achievement. Algorand was thought by some to be impractical for use in a public blockchain, precisely because of the high communication costs. It remains to be seen whether that number can be further reduced in future iterations.</p><p>Rafael Pass talked about two future directions for the Thunder project near the end of his presentation. Thunder Cash is a tailored version of Thunder that removes the need for a leader, and will be suitable for transactions only (no smart contracts). Optimal BFT is an optimized BFT protocol specifically designed to accelerate the already fast confirmation time of Thunder.</p><p>Besides the cryptocurrency workshop, there was a concurrent one-day workshop on Post Quantum Cryptography which also drew my interests. One talk reviewed submissions to NISTPQC, a crowd-sourced competition to select the most hopeful candidates for quantum-computing resistant algorithms. Of the 64 submissions, many have already been hacked and others weakened. It appears that for signatures, lattice based and hash based are in general harder to hack, while almost all the code based submissions were hacked. Also, a factoring-based called pqRSA deserves more attention.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a6968a8d3595" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Stablecoin Projects (Besides Tether) Are Bound To Fail]]></title>
            <link>https://medium.com/@chenxicai1/why-stablecoin-projects-besides-tether-are-bound-to-fail-344652596717?source=rss-9bdad0704145------2</link>
            <guid isPermaLink="false">https://medium.com/p/344652596717</guid>
            <category><![CDATA[stable-coin]]></category>
            <category><![CDATA[bitcoin]]></category>
            <category><![CDATA[usdt]]></category>
            <category><![CDATA[tether]]></category>
            <dc:creator><![CDATA[Chenxi Cai]]></dc:creator>
            <pubDate>Sun, 01 Jul 2018 15:41:18 GMT</pubDate>
            <atom:updated>2018-07-01T16:27:11.229Z</atom:updated>
            <content:encoded><![CDATA[<p>Stablecoin (cryptos whose values are pegged 1:1 to fiats like US Dollars) is a hot investment area for Silicon Valley VCs right now. (Curiously, this is not the case for those in Beijing here at all. We simply know better :-)). Supposedly, stablecoins will increase blockchain application adoption by non-crypto users without them having to learn the in-and-outs of the crypto market. Currently, the most popular stablecoin is Tether (USDT), which has been under attack for being centrally managed by the exchange Bitfinex and lacking transparency (although they did conduct a full auditing of funds lately). A lot of new projects are spun out every few weeks this year to address this supposed shortcoming of Tether.</p><p>However, I would like to argue that building a stablecoin that is not secured 1:1 by actual fiats is fundamentally an impossible mission, analogous to building a perpetual motion machine, or turning iron into gold. Here is my argument, based on physics.</p><p><strong>Market Prices Fluctuations Is Entropy</strong></p><p>When two assets are traded on the open market, their relative prices follows a path that is uncertain to all participants. This inherent uncertainty is the entropy of the market.</p><p><strong>Stabilizing Prices Requires Outside Intervention</strong></p><p>Following the Second Law of Thermodynamics, stabilizing prices requires costly intervention from outside the system. This will take the form of arbitraging slippages and fees for “Algo central bank” type proposals and margin calls and default losses for “Collateralized debt” type proposals. Therefore, in order to properly operate, stablecoin systems require outside fiats to be constantly fed. This will introduce a burn rate to the fiats that is used in the system.</p><p><strong>Burn Rate Is Proportional To Velocity Of Money</strong></p><p>Now comes the most crucial piece. We know for a fact that due to much reduced settlement time, the velocity of money is much, much higher in the crypto world than that in the outside world. This means that to stablize the price of a stablecoin, you would also have to burn fiats much faster than if you just use USD!</p><p><strong>All in all</strong>, stablecoins are thermodynamically impossible to operate without either burning a large amount of cash constantly or being backed 1:1 by fiats.</p><p>In fact, one of the major sources of<strong> hidden</strong> inefficiency in the fiat world is exactly the desire of central banks and governments to stabilize prices through various financial and regulative means such as currency anchoring, inflation targeting, rent control, anti-predatory lending laws and etc. These policies, likewise, require tremendous of outside resources (tax revenue) to sustain and places a constant burn on our economic system. This was the true reason why the free floating cryptocurrencies were liberating in the first place.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=344652596717" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Smart Contract Formal Verification: An Overview]]></title>
            <link>https://medium.com/@chenxicai1/smart-contract-formal-verification-an-overview-9228fb56c83b?source=rss-9bdad0704145------2</link>
            <guid isPermaLink="false">https://medium.com/p/9228fb56c83b</guid>
            <category><![CDATA[smart-contract-security]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Chenxi Cai]]></dc:creator>
            <pubDate>Wed, 27 Jun 2018 03:46:23 GMT</pubDate>
            <atom:updated>2018-06-27T03:46:23.281Z</atom:updated>
            <content:encoded><![CDATA[<p>As DApps platforms grow in popularity and the number of smart contracts deployed increases exponentially, smart contract security becomes the focus of investors and developers. Several ERC-20 ICO projects this year have been hacked because of loopholes in the smart contract code, leading to huge losses for investors. In order to prevent the occurrence of similar events, exchanges, wallets and ICO projects are investing more in smart contract security, so the ecosystem surrounding smart contract security has become an investment hotspot.</p><p>Smart contract security can be roughly divided into 4 sub areas: contract development template, contract auditing, syntax tools and bounty system.</p><p>(1) Contract development template. Templates such as Openzeppelin, Etherparty and so on provide a number of standardized templates that are supposedly intensely audited and battle-tested.</p><p>(2) Contract auditing. It can be further divided into human audit and machine-assisted audit.</p><p>Machine-assisted audit can be further divided into rules-based verification, semantic verification and formal verification, and formal verification is the focus of this report.</p><p>(3) Syntax tools. A number of new public blockchain projects use Haskell or OCaml language as smart contract programming language, because it is easier to write the developer library for formal verification in functional languages.</p><p>(4)Bounty system. It is more of an ad hoc system to avoid large hacking losses after contracts have been deployed.</p><p>Formal verification refers to the use of mathematical formal methods to prove the correctness of a program. So far, formal verification is mainly used in military, aerospace and other areas requiring a high level of security. Since traditional Internet application and blockchain DApps are fundamentally different, their development process should be adjusted accordingly, the most important point being how much emphasis is placed on security.</p><p>In traditional Internet applications, because of the universal use of centralized server + light clients model, if the application contains security risks, the owner only needs to modify the server-side code and easily roll back faulty transactions to prevent user losses. Therefore, the process of traditional Internet application development pays more attention to the fast iteration, at the expense of security in exchange for efficiency.</p><p>In block chain DApps, because the block chain is immutable, once a bug in the program is exploited, the loss is huge and irreversible. Once a hacking event occurs, it takes a community-wide consensus to roll back the transaction, so it is unrealistic to roll back the deal every time an attack occurs. Therefore, the process of development of block chain application requires much higher level of security testing, which in turn will sacrifice the speed of iteration.</p><p>Due to the scarcity of block chain developers not catching up with the increase of the number of intelligent contracts, human audits will be prohibitively expensive, so machine-assisted verification is the inevitable trend at present. The implementation of rules and semantic verification is relatively easy and the technical threshold is lower, but only shallow errors can be caught. Only formal verification method can really discover deep logical issues within a program.</p><p>At present, there are three kinds of products in the block chain industry related to formal verification: Vaas (Verification-As-A-Service) platform, public chain, and language. I briefly list them below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UTp1mC9FUOSzhZ2I7D_MxQ.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9228fb56c83b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why the Weiss Crypto rating is kind of ridiculous and should not be trusted]]></title>
            <link>https://medium.com/@chenxicai1/why-the-weiss-crypto-rating-is-kind-of-ridiculous-and-should-not-be-trusted-9f22f13a3563?source=rss-9bdad0704145------2</link>
            <guid isPermaLink="false">https://medium.com/p/9f22f13a3563</guid>
            <category><![CDATA[crypto]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Chenxi Cai]]></dc:creator>
            <pubDate>Thu, 25 Jan 2018 07:45:07 GMT</pubDate>
            <atom:updated>2018-01-25T07:45:07.682Z</atom:updated>
            <content:encoded><![CDATA[<p>The Weiss company just released their crypto rating, touted as the first ever by an “official” ratings agency. However, after reviewing their rating, I gasped at some of their judgements.</p><p>Someone clipped the ratings here in a tweet:</p><p><a href="https://twitter.com/CryptoGeisha/status/956173006547558400">https://twitter.com/CryptoGeisha/status/956173006547558400</a></p><p>The most striking observation is that Bitcoin, the granddaddy of all cryptos, only gets a C+ rating, below that of many “hot” coins of the day. How is this ever reasonable? Yes, the transaction fees was somewhat visibly higher, but does it not instead testify to the high demand? The low rating assigned to Bitcoin puts into question whether the rater understands the security trade-offs involved in crypto architecture. Bitcoin’s ostensibly slow speed exactly guarantees a high degree of security.</p><p>Ethereum and EOS both get a B rating, the highest rating given to any crypto. While I do not question the utility offered by Ethereum, and the great future potential of EOS, the fact is that Ethereum suffered several instances of serious hacking over its life time in which hundreds of millions of dollars in funds were stolen or frozen, and that EOS still take years to develop and many unknowns are involved.</p><p>The most baffling positive rating is for Cardano, which receives a B- rating. From what I understand, Cardano is a copycat project trying to learn (or steal) from the envious success of NEO (previously Antshares). It was hyped in a hurry in November 2017 without a working product or developer community, and is therefore a classic pump and dump scheme.</p><p>In sum, the rating given to the cryptos appear to not take into consideration of many important material dimensions, such as security, existence of a working product, and even originality. This seriously reminds me how we got to the subprime mortgage meltdown.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9f22f13a3563" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Bitcoin is still cheap at $10,000+]]></title>
            <link>https://medium.com/@chenxicai1/why-bitcoin-is-still-cheap-at-10-000-14d8ebc7839c?source=rss-9bdad0704145------2</link>
            <guid isPermaLink="false">https://medium.com/p/14d8ebc7839c</guid>
            <category><![CDATA[bitcoin]]></category>
            <dc:creator><![CDATA[Chenxi Cai]]></dc:creator>
            <pubDate>Sun, 21 Jan 2018 04:15:52 GMT</pubDate>
            <atom:updated>2018-01-21T04:24:59.186Z</atom:updated>
            <content:encoded><![CDATA[<p>Yes, one bitcoin costs more than 7 ounces of gold, a used car in decent conditions, and a year of rent in all but the most expensive cities. No, it is not expensive (yet). This is for the following reasons.</p><ol><li>My price target for 2018 is between $40,000 and $100,000.</li></ol><p>Comparing bitcoin to tulip is simply ridiculous, because a) the supply of bitcoin simply does not respond to an increase in price, unlike tulips, which can be grown in relatively short amount of time b) tulips are not a new tech that has a mass potential. As a result, when bitcoin has a price increase, no other force than an drop in demand will slow it down. For a nascent technology, a drop in demand will only occur long after mass adoption already take place(think radios, landlines, desktop computers).</p><p>The price of bitcoin roughly follows the Metcalf’s law and roughly grows in proportion to the square of its user base. Therefore, it is only normal for it to exhibit a parabolic trend. Sometimes, it hits the top of the parabolic channel as it did in 2014, which necessitated a downward adjustment. Overall, however, the parabolic pattern holds over its 9-year lifetime. For a simple comparison, just think of Facebook stocks pre-IPO.</p><p>2. Bitcoin transaction fees will go down significantly</p><p>Recently, transaction fees became a huge concern for many. However, the lightning network, a technology that allows very low fee bitcoin transactions, is making fast progress. Once adopted, one can comfortably get a cup of coffee with bitcoins. This means bitcoin will start taking market shares away from payment network such as Visa, Paypal and etc, each of which has tens of billions in valuation.</p><p>3. The altcoin boom will ultimately benefit bitcoin</p><p>Bitcoin is the reserve currency of the crypto world. Even though its market share has declined over the past year, it remains the go-to unit of accounting for most veteran traders. As the altcoin market develop, it will generate more demand for bitcoin as the “gateway coin” between fiats and cryptos.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=14d8ebc7839c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Trump’s Tax Reform Merely A Bargaining Chip Against China]]></title>
            <link>https://medium.com/@chenxicai1/trumps-tax-reform-merely-a-bargaining-chip-against-china-7d61a9fc2a13?source=rss-9bdad0704145------2</link>
            <guid isPermaLink="false">https://medium.com/p/7d61a9fc2a13</guid>
            <category><![CDATA[china]]></category>
            <category><![CDATA[trump]]></category>
            <category><![CDATA[politics]]></category>
            <category><![CDATA[tax-reform]]></category>
            <dc:creator><![CDATA[Chenxi Cai]]></dc:creator>
            <pubDate>Sat, 18 Nov 2017 19:22:03 GMT</pubDate>
            <atom:updated>2017-11-18T19:31:20.337Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PYDPrDkpijK_73YwCl4BWQ.jpeg" /></figure><p>Much ado has been given to Trump’s tax reform. However, it is apparent to me that Trump did not anticipate the bill’s major clauses to pass the Senate without substantial modification in the first place. Both the timing and the content of the bill aim at getting China to make some of the biggest trade concessions in recent decades.</p><p>The central piece of the reform that gets Chinese officials really nervous is the proposed corporate tax cut. According to Chinese researchers’ estimates, billions of dollars of corporate profits parked in emerging markets will flow back to onshore accounts once the reform goes into effect. China has already tightened capital controls during the past year to prevent a stampede of capital flight, and such an outcome, probably to Trump’s knowledge, will be nightmarish for the Chinese financial sector, which is already battered by tightening of the housing market and cooling of the manufacturing sector.</p><p>Given the contentious split even among many Republicans regarding many items in the tax reform, the House’s version is very unlikely to pass the Senate super-majority. However, Trump correctly anticipated that the Chinese officials, unfamiliar with the American legislative process, did not see through his shenanigans. As a result, Trump secured unprecedented concessions at the negotiation table, among which we count <a href="https://www.ft.com/content/d4a85422-c5d5-11e7-b2bb-322b2cb39656">opening financial sector to foreign ownership</a>, <a href="http://China Considers Opening Up to Foreign Electric Carmakers">opening up to foreign electric car-makers</a>, and possibly <a href="https://www.cnbc.com/2017/09/19/facebook-to-enter-china-next-year-analyst-predicts.html">allowing Facebook to re-enter China</a> under a censored version.</p><p>As much as many people hate Trump, he does excel in two things really well: shameless self-promotion and bluffing in negotiations.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7d61a9fc2a13" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>