Have you ever being in a situation to guess another person’s age? Well May be YES!! How about playing games like finding things in minimum time? or about finding the written character where your doctor wrote in the prescription when you are sick? Well everyone faced these problems in real life. How about asking your machine or your favorite computer to do the task for you. Isn’t it great? well computers actually do by using Machine Learning. so for doing this we actually need to train the machine by using some powerful datasets. The key to getting better most fields in life is practice. Practice on a variety of problem from image processing to speech recognition. Each of these problem has it’s own unique technique and approach. But how do you get this data? Working on these datasets will make you a better data expert and the amount of learning you will have will be invaluable in your career. We have listed a collection of high quality datasets that every Machine learning enthusiast should work on to apply and improve their skill. IMAGE DATASETS Car License Plate Detection Has around 500 images with car license plates marked as rectangular bounding boxes in images of cars on roads and streets. to dataset. Link Celebrity Face Key-Points A database of around 2500 images with faces of celebrities and important key-points like eyes, nose etc marked. to the Dataset. Link E-commerce Tagging for clothing Images from E-commerce sites with bounding boxes drawn around shirts, jackets, sunglasses etc. Has around 500 images manually tagged for item detection. to Dataset. Link Wound Dataset Around 300 medical surgery images with bounding boxes drawn around wounds. to Dataset. Link IMDB-WIKI dataset , an abbreviation of , is an of information related to world films, television programs, home videos and video games, and internet streams, including cast, production crew and personnel biographies, plot summaries, trivia, and fan reviews and ratings. An additional fan feature, message boards, was abandoned in February 2017. Originally a fan-operated website, the database is owned and operated by IMDb.com, Inc., a subsidiary of . Not very rare but the grand-daddy of all image datasets. IMDb Internet Movie Database online database Amazon Description: IMDB and Wikipedia face images with gender and age labels. Instances:523,051 Format: images Default task: Gender classification, face detection, face recognition, age estimation Created: 2015 by R. Rothe, R. Timofte, L. V. Gool Download link: https://data.vision.ee.ethz.ch/cvl/rrothe/imdb-wiki/ The ROOMS This dataset is an image classification dataset to classify room images as bedroom, kitchen, bathroom, living room, exterior, etc. Images from different houses are collected and kept together as a dataset for computer testing and training. This dataset helps for finding which image belongs to which part of house. Description: The dataset has 20001 items of which 4404 items have been manually labeled. Categories: bedroom, kitchen, bathroom, exterior, living room, other Default task: image classification, image captioning. Format: images Created by: DataTurks Download link : https://dataturks.com/projects/sheerun/rooms Visual Genome dataset Visual Genome is a dataset, a knowledge base, an ongoing effort to connect structured image concepts to language. Description: 108,077 Images 5.4 Million Region Descriptions 1.7 Million Visual Question Answers 3.8 Million Object Instances 2.8 Million Attributes 2.3 Million Relationships Everything Mapped to Wordnet Synsets Format: images, text Default task: Image captioning Created : 2016 by R. Krishna et al. Download link : http://visualgenome.org/api/v0/api_home.html CRACK Classification dataset This dataset is to classify the cracks on the walls. The dataset consists of wall images with or without cracks It has images with shadow of some wires also which exactly looks like cracks on the wall, we should train the system carefully so that it has to differentiate between cracks and shadow. This dataset is very challenging which will revamp your coding skills. Description: The dataset has 1428 items of which 1428 items have been manually labeled. Categories: crack, no-crack Format: images Default task: image classification Created by : Data Turks Download link: https://dataturks.com/projects/miaozh17/Crack%20Classification IIT-5K OCR dataset Has 5K labeled images of street signs cropped to just contain the portion that has the text. Quite a difficult dataset with even the best vision algorithms being at 80% accuracy rates. (Read: comparison of on this dataset) Google, AWS, Microsoft OCR APIs to the dataset. Link CARS dataset This dataset is to identify cars in the images. The set has different images which does or does not have cars in it. The main objective of this dataset is to identify even the small parts of the car in the images. This dataset is human labeled dataset. Description: dataset has 613 items of which 604 items have been manually labeled. Categories: cars, no-cars. Format : images Default task : image classification. Created by: Data Turks. Download link: https://dataturks.com/projects/dominique.paul.info/cars2 The FERET Dataset The Face Recognition Technology (FERET) program is managed by the Defense Advanced Research Projects Agency (DARPA) and the National Institute of Standards and Technology (NIST). Department of Defense (DoD) Counterdrug Technology Development Program Office sponsored the Face Recognition Technology (FERET) program. The goal of the FERET program was to develop automatic face recognition capabilities that could be employed to assist security, intelligence, and law enforcement personnel in the performance of their duties. The FERET database was collected in 15 sessions between August 1993 and July 1996. The database contains 1564 sets of images for a total of 14,126 images that includes 1199 individuals and 365 duplicate sets of images. A duplicate set is a second set of images of a person already in the database and was usually taken on a different day. Description:11338 images of 1199 individuals in different positions and at different times. Format: images Default task: Classification, face recognition Created : 2003 by United States Department of Defense Download link : https:// /.../ www.nist.gov face-recognition-technology-feret Face Detection Has around 1300 faces marked as rectangular bounding boxes in images. Images range from part pics to random people on streets. to Dataset. Link CALTECH-101 dataset is a of . The Caltech 101 data set was used to train and test several machine learning, computer vision recognition and classification algorithms. A set of annotations is provided for each image. Each set of annotations contains two pieces of information: the general bounding box in which the object is located and a detailed human-specified outline enclosing the object. Caltech 101 data set digital images A MATLAB script is provided with the annotations. It loads an image and its corresponding annotation file and displays them as a MATLAB figure. The Caltech 101 data set aims at alleviating many of these common problems. The images are cropped and re-sized. Many categories are represented, which suits both single and multiple class recognition algorithms. Detailed object outlines are marked. Available for general use, Caltech 101 acts as a common standard by which to compare different algorithms without bias due to different data sets. Description: Pictures of objects, detailed object outlines marked. Instances: 9,146 images, split between 101 different object categories, as well as an additional background/clutter category. Format: Images Default task: Classification, object recognition. Created: September 2003 and compiled by Fei-Fei Li Download link : http://www.vision.caltech.edu/Image_Datasets/Caltech101/ UXBOT Dataset This dataset is to classify uxbot pictures into dark, professional, Minimalist, Glamorous, etc.… uxbot is the platform for chatting now a days. This dataset is used to train computer with new technical skills. It is human labeled dataset. Description : dataset has 129 items of which 129 items have been manually labeled. Format : images Categories : Elegant, clean , fresh , light, Airy, cooperate, funky, Retro, Eddy, fun, etc.…. Default task : image classification Created : Data Turks. Download link: https://dataturks.com/projects/briannaorg/UXBot LABELME Dataset LabelMe is a project created by the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) which provides a dataset of digital images with annotations. The dataset is dynamic, free to use, and open to public contribution. The motivation behind creating LabelMe comes from the history of publicly available data for computer vision researchers. Most available data was tailored to a specific research group’s problems and caused new researchers to have to collect additional data to solve their own problems. LabelMe was created to solve several common shortcomings of available data Description : Large dataset of images for object classification. Format : Images, Text. Default task : Image Classification, object detection. Created : by 2005 MIT Computer Science and Artificial Intelligence Laboratory Download link : http://labelme.csail.mit.edu/Release3.0/browserTools/php/dataset.php You can find thousands of such open datasets here.

Amazon

Genome

Google

Microsoft

Rare Datasets for Computer Vision Every Machine Learning Expert Must Work With

A story of Galapagos birds and silicon chips Physics has changed human life a million times over. It regularly produces breakthroughs that seem science fiction: , , and . It’s amazing how scientists can unravel the mysteries of the universe. controlled star explosions putting a lab on a comet gravitational waves things too complicated to understand One would expect the human body and mind, which we all are familiar with, would pose little challenge. Yet psychology is going through a : less than 40% of tested research findings prove true. Then in medicine there is a marked lack of progress: despite incredibly large investments, . If you look at nutrition, it just a , with leading . replication crisis the war on cancer is being lost medical error might be third leading cause of death mess scientific research contradicting the official paradigm But why? We intuitively understand more about our minds and bodies than about quarks and bosons. Why can’t we develop this into a reliable science? And what does it have to do with Galapagos? There are 5 unavoidable causes for our incompetence is understanding ourselves: Galileo studied gravitational acceleration by dropping stones from the Tower of Pisa. In the 1940s the U.S. studied syphilis by infecting prisoners and individuals with mental illnesses with syphilis. Both are experiments, one in physics, the other in medicine. The gold standard of science is the experiment: a procedure undertaken to validate or refute a hypothesis. What Galileo did is admirable, while what the U.S. did is despicable. The inherent limitation of medicine and psychology: experiments can harm human beings. Because of this it is too costly to run most experiments for medicine and psychology. No potential finding is valuable enough to cost human lives. So medicine and psychology turned to observational studies. Scientists look at groups of people considered to differ in one specific variable (the independent variable). Then they measure various outcomes for these groups. The differences in outcomes between groups are attributed to the independent variable. Observational studies produce . They constitute the reason for much of the uncertainty in human understanding. uncertain findings But there are more basic limitations than the research design. They have a lot to do with the Galapagos and Darwin’s insights there. “All our knowledge begins with the senses” Immanuel Kant, Critique of Pure Reason Despite the complex and abstract nature of modern physics, it is rooted in sensory perceptions. Our sensory organs evolved to be great at perceiving the outside world. It was a basic requirement for survival. Physics is creating a model of the world out of what our senses tell us. When we turn our senses inward, the results are murky. You don’t know how fast your heart is beating. You cannot feel the health of your organs. You cannot even sense if you have foreign bodies inside (like parasites). Likewise we don’t perceive the processes in our minds, we can’t look at synapses or smell which areas of our brain are active. We can only feel the effects of our internal processes. Why are we in the dark about our own inside? Nothing about us is random. Ever forget to do anything? Maybe an email, or a chore? What if you forgot to breathe? Or did not send the right antibodies to fight an infection? We think of our conscious mind as our ultimate tool. But it cannot be trusted with the critical workings of our body. So they are automated. They are handled by an unconscious nervous system that does everything without our input. If we don’t consciously control these inner workings, then we do not need to know what is going on. Our senses do not tell us much about our own body. Except for low-res signals to prompt us to take specific actions (e.g. pain to not walk into walls or touch barbed wire). Without sensory input, we have to rely on indirect evidence of our own body’s processes. The first physics discovery might have been the intuitive understanding of gravity: objects fall. This is a repeatable, generalizable fact, e.g. all things fall, all the time, in any place on Earth. The first psychological discovery might have been that people get scared if you shout at them. But some people get more scared, some less, and others attack you. And all of them behave differently at different times. This does not fulfill any of the requirements for scientific fact. Plus, even without the sensory handicap, the human body is harder to study because of: “The first thing one learns about complex systems is that they are not a sum of body parts: a system is a collection of interactions, not an addition of individual responses.” Nassim Taleb In the physical world, if a stone falls besides you, you are mostly unaffected. As is most of the rest of the Earth. All events are correlated, but often the interaction effects are small. Scientist can study specific effects in isolation to the rest of the universe. In the human body, things are different. If you eat a bad egg, everything in your body is affected. Even your mood and thinking. The interaction effects are large and widespread. And it’s not just temporary events: , and your mind . This means it’s impossible to study any one event or one variable in complete isolation. Anything and everything interacts and affects the whole system. bacteria in your gut influence your emotions and behaviour health correlates with happiness can cause illness We don’t have the sensory organs to see these interaction effects. So we try to use math to model them. But our mathematical methods are quite . bad at understanding interaction effects in general In physics scientists use math to describe the world. Math is precise, clear, objective. It’s really hard to misunderstand math without showing you are an idiot. But because math is hard, and math for interconnected systems is especially hard, psychologists and physicians use words instead. Words are much less useful than math. It’s really, really easy to misunderstand words. Often it even makes you look smart. So: Physics = testable — easily observable — math (clear) Human body= not testable — hard to observe — words (unclear) Already looks hard. But it gets worse, the unclear words compound our: “Yet weren’t all human beings simply human beings no matter what name you applied to them[?]” Mind-reading robot Giskard, Robots and Empire, Isaac Asimov When the rock falls, I see it fall. And you see it fall. We agree that the rock fell. Even when we use imprecise words. When I eat a watermelon, I feel things (feelings, sensations, thoughts). When you eat the same watermelon, you feel different things. We also feel different things when someone shouts at us. Each of us has a big complex baggage of experiences that shapes our perceptions, emotions and thoughts. When studying the human body and mind, we bring all this baggage with us. When I research depression, I cannot help but think of my own experiences of being depressed. These biases cloud our judgements. I use poorly designed words to express my biased judgement. Your own biases then make you decode those words into something different than what I intended to express. Yet we might throw words at each other that make us believe we are in agreement. Maybe we need to design an A.I. that is free of all these biases so it can understand us for us. Did you see the connection with silicon chips yet? Hint: do our brains resemble computers? An A.I. would also benefit from not going through our biggest obstacle: evolution “Man is descended from a hairy, tailed quadruped, probably arboreal in its habits.” Charles Darwin Evolution is the root cause. We did not evolve to develop perfect medicine to live forever. We did not evolve to make ourselves happy and fulfilled. We evolved to survive and pass our DNA to offspring. That is it. We did not need an understanding of ourselves for this purpose. It might actually be detrimental. It’s hard to deal with the fact that there is no meaning to life beyond living. But it’s even more insidious. We tend to think of ourselves as brilliant minds trapped in flawed bodies. A popular metaphor is to think of the brain (and body) as hardware, and the mind as software running on top of this hardware. This is a bad metaphor. It confuses and obfuscates. A better metaphor is to think of the brain+mind as our Operating System and accompanying hardware. The brain and mind are interconnected. They have a continuous two-way relationship. They have evolved along a very long period of time to fulfill the survival function. On top of this O.S., we have the applications and files. The software. These are all the ideas, thoughts, models, emotions, concepts that we acquire and integrate in our own lifetime. They make up our individuality, our growth and our self. Files and applications can be added or erased. But we cannot change the O.S. (we don’t have admin). We cannot see the code behind it or go beyond it, to the BIOS. We have to work with what we have. Trying to create complex abstract theories without taking into account the specifics of the O.S. is akin to playing the latest videogame at high-res on Windows 3.1. It will crash and freeze. What I’m saying with my metaphor poorly constructed out of imprecise words is that the only clear fact about the human body and mind is it evolved for survival. If we can build a new model of ourselves, starting from that, then we might learn to improve our lives. We cannot have the science of physics, but we can have clarity. Help me help others Avantgarde Savage is creating a Radical Model for Better Life | Patreon I am creating Avantgarde Savage: a radical new model to create a Better Life for everyone. We live in a world where , in meaningless jobs, lost, complacently unhealthy, lonely and alone. All because we do not understand ourselves. most of us are unhappy is born from a multitude of sciences about human mind and brain and body. Based on how Homo Sapiens evolved in nature, we uncover how we can Avantgarde Savage make our lives in the modern world better. Discover Avantgarde Savage : did you find the link with Galapagos islands and silicon chips? Addendum we accept the theory of evolution inspired by Darwin voyage to the Galapagos island, yet we see ourselves as designed and abstract as silicon chips in computers we build. If not:

Waves

Discovery

Why Humans Rock at Physics but Suck at Psychology and Medicine

When you talk about cryptocurrencies with people unfamiliar with it, the question of volatility is among the first to pop up. Quite often, it’s a low effort way to dismiss the whole world of cryptocurrencies and recent progress made with decentralization. Volatility does not affect all cryptocurrencies: some of them are made to be stable (price wise) — we call them the “stablecoins.” Lire cet article en francais, sa langue originale. Before we get to that point, let’s state some necessary facts to understand this article: . There are more than 1500 tokens listed on CoinMarketCap, which is only the tip of the iceberg: in reality, there are thousands of cryptocurrencies circulating. The sheer amount of tokens can be scary, but it’s positive in the long run. It means that for every problem cryptocurrencies can solve you have dozens of different teams exploring as many different approaches — no stone is left unturned. Cryptocurrencies are much more than just the Bitcoin Whatever the initial vision of Satoshi was, . Its technical limitation such as its low transaction speed or the issues with mining (energy consumption, centralization, etc.) makes it inapt for this use case. However, its persistence ( is almost 10 years old!) as well as its reliability makes it a good “store of value”, just like gold. Bitcoin is unfit to become a mainstream currency Bitcoin Nevertheless, there are many projects with the ambition to become tomorrow’s mainstream currency. Those projects are called and they will be the topic of this piece. They answer the fundamental issue raised when considering Bitcoin as a currency: its volatility. stablecoins The question of volatility highlights a vital characteristic of a currency: price stability is a necessity to have a currency people feel confident using. Indeed, creating a new currency requires a thorough understanding of both technical mechanisms as well as social dynamics. In order to become a currency, a good must be recognized as such by its users and gain their trust. Are fiat going south? We observe a strong correlation between the adoption of cryptocurrencies and the “weakness” of the national currency (called fiat currencies) in a given country. It’s hard to measure the adoption of cryptocurrency, but we can easily get figures on how many people are interested in buying some using Google Trends: Geographical distribution of the query “How to buy bitcoin” using Google Trends The Ghanaian currency (the Cedi — GHS) went from roughly a dollar in 2008 to around 20 cents nowadays. Meanwhile, the Nigerian currency (the Naira — NGN) lost about half of its worth (measured against the dollar). The exchange rate of the dollar against the Cedi (left) and Naira (right) over the last ten years. Source: . TradingEconomics The devaluation happened fast enough for people living in these countries to see the impact of it. It could explain the strong interest we observe for cryptocurrencies in Ghana and Nigeria. However, the issue with the determining how much a given currency is worth is more complex than simply measuring it against the dollar. If we use the dollar as our reference point, we must question its stability: is the dollar itself even stable? Without thinking twice about it, the “yes” seems like the go-to answer. The dollar hasn’t changed over these last 50 years — a $1 bill emitted in 1968 would still be recognized as such today. However, what about its worth? It is changing all the time. Is one 1968 dollar still worth $1 then? The answer is simply no — it would give access to a purchasing power corresponding to roughly 7 present days dollar. Evolution of the buying power of the equivalent a 1968 dollar from this date until now. Source: . TradingEconomics The main driver of the spread is of course inflation. The topic is sophisticated enough to have theses written about it, so we’ll leave it to others. Now that we have some context let’s get back to stablecoins. What’s a stablecoin? The idea behind the stablecoin is pretty simple, the challenge lies in the execution. A stablecoin is a cryptocurrency that trades at a fixed price against a good (such as gold) or a common currency (most often the dollar). A stablecoin’s worth is not stable; it is merely its value that is pegged to a stable asset. Creating a reliable stablecoin is already a real challenge, so creating a fixed-worth currency is a wild dream. relatively However, the question of having a stable value is interesting to consider. How would it look like? That one unit of the given currency is always worth one unit? That’s DOGE’s specialty! The DOGE is a parodical currency created out of a meme which is now taken more and more seriously as cryptocurrencies develop Let’s go back to more serious matters: the stability of currency could be defined as the fact that one unit of it always grant access to the same value. This definition would be consistent with the definition of a currency proposed in a . previous piece Money is not a store of value, it’s a claim upon value. Keeping the link between the currency and the actual value it gives access to is elegant, in theory. However, in the real world, it’s close to impossible to measure the buying power of a given currency exactly. It is why we’ll consider stability defined as a currency keeping a constant price in regards to other goods or currencies which have a stable value. Before we get to the different stablecoin projects there are, let’s examine their utility. relatively What are stablecoins good for? We already have live and available stablecoins such as the USDT. They are mostly used by crypto exchanges and traders because they offer a way to avoid returning to fiat currencies and having to abide by all the legal framework that comes with it. Moreover, since their price is fixed against the dollar they are somewhat of a safe haven for traders: with them, they can get out of the market without really doing so. For now, stablecoins are mostly a traders’ gimmick used to get in and out of the market without bringing their money back into their bank account. Hence, . However, they can provide a real utility in the long run. in the short-term, they are mostly a speculative tool Since cryptocurrencies have high volatility compared to fiat currencies, they are not really used as currency. Almost nobody is choosing to get their pay in cryptocurrencies, or using them to buy their groceries. Having a stable cryptocurrency well integrated with the rest of the ecosystem would provide a credible solution to address these needs. At first, stablecoins would be the gateway between the fiat and the crypto world. In the long run, they could become the “fiat of cryptos” themselves: . It would provide a broader and more reliable access to the services offered by other (non-stable) cryptocurrencies. they would become cryptocurrencies which are stable enough to become a preferred payment method How to stabilize a cryptocurrency: live examples To fulfill their promise of stable exchange rate against a given fiat, stablecoins must offer a . As a result, several types of stablecoins emerged depending on the collateral mechanism they used: collateral Fiat collateralized stablecoins It is the most direct and simple approach to make a stablecoin, but also the most centralized. In brief, there is a central entity minting the stablecoin who owns a reserve of dollar corresponding to the volume put in circulation. This dollar reserve offers guarantees regarding the solvability of the coin and allow to exchange the stablecoin back to a dollar at any time. Even so, this approach comes with several issues. First, it’s since a single entity is emitting the stablecoin and backing it. There are no other choices but to trust it. Besides the approach is , especially at scale, as it requires to “freeze” an amount of money equal to the number of stablecoins in circulation. completely centralized economically inefficient USDT This mechanism is used by the USDT, the most widespread stablecoin. Despite its adoption, for its lack of transparency — there is no reliable method allowing its users to verify than Tether (the institution emitting USDT) actually possesses in its reserve a stock of dollar corresponding to the amount of USDT in circulation: around 2,8 billion currently. However, the ties between Tether and Bitfinex (the 4th cryptocurrency exchange regarding the trading volume) raise questions regarding the potential for market manipulation it allows. the USDT is widely denounced TrueUSD Created in answer to the lack of transparency of the USDT, TrueUSD uses the same mechanism while offering more guarantees. Rather than owning the token themselves, TrustToken (the institution emitting TrueUSD) works with a network of banks and financial institutions which keep the funds in escrow for them. One of them, Cohen&Company, publishes an twice a month to confirm that the financial partners of TrustToken do hold in escrow a number of dollars corresponding to the volume of TUSD in circulation. independent audit The guarantees offered seem more reliable than for the USDT, yet the problem is similar: this system requires its user to trust TrustToken and its partners, and particularly Cohen&Company to be honest. It works until it doesn’t. Crypto collateralized stablecoins While fiat collateralized stablecoins are fulfilling (so far) their promise of stability, they are doing a disservice to cryptocurrencies as a whole whose aim is to provide a decentralized alternative to the legacy financial and monetary networks. To answer this concern, another type of stablecoins emerged. They use other cryptocurrencies as collateral and a network of smart contracts to peg their exchange price to the underlying asset. Yet this type of stablecoins is facing issues too, the main one being the volatility of cryptocurrencies used as collateral. To address them, they necessitate , which hinders their potential large scale usage. a collateral superior to the value of the stablecoins emitted MakerDAO — Dai Dai is another stablecoin pegged to the dollar (1 Dai = $1). It has some interesting features, the main one being that it is fully decentralized. Indeed, the institution emitting the Dai is MakerDAO, a decentralized autonomous organization. It’s an , making them transparent and unchangeable. organization which governing rules are automated and enforced through publicly auditable smart contracts Dai works in pair with MKR, the governance token. Owners of MKR can vote to decide on the mechanisms used to stabilized the Dai. The system is quite complex and could deserve several articles by itself. Dai was launched on December 17th 2017 after three years of development. Since its launch, its price was in the same range of stability than fiat collateralized stablecoins (+/- ~10 %). Un-collateralized stablecoins These stablecoins are the most promising in the long run as the address the main concerns raised by the two previous types: they propose a decentralized solution which could be used at scale. These projects are still in the development phase. Most of them use algorithms which reduce or increase the amount of the given stablecoin in circulation to stabilize their price. If the price goes over the dollar, the circulating supply increases; if it goes below, the supply decreases. The two most prominent projects in this category are Basis and Carbon. Basis Basis wants to create an “algorithmic central bank” that would stabilize its price. Their system uses three different types of token which makes it quite complex. The goal is to reproduce a simplified version of the federal reserve system with the addition of investors. Carbon On the other hand, Carbon is trying to execute an approached theorized in 2014 already by Robert Sams called the . The solution is mostly theory for now and has yet to see a working implementation. seigniorage shares Stablecoins: a gift or a curse for cryptocurrencies? Stablecoins are a recent yet fascinating innovation. There are many projects currently in development ( ) using diverse approaches. The typology proposed above is simplified: some projects mix several types of collateral while others use radically different mechanisms. non-exhaustive list Today, stablecoins are speculative more than anything else, but they can become one of the critical elements of the mainstream adoption of cryptocurrencies in the long run. Looking at the volume, the USDT is currently obliterating its competition, which creates a systematic risk for the whole cryptocurrencies market. If USDT was to fail, it could lead to widespread panic. Thus tools speculation pushed the market towards a highly centralized tool lacking any real transparency: a gentle irony for the future of decentralization. Let me know if you liked this article and you can keep pressing 👏up to 50 claps. check out my wiki for more content like this. Don’t forget to clap it to help it spread: This article is the translation of an original (French) EcoCrypto.fr story. Sources / Deep Dive — Amy Castor on Bitcoin Magazine, July 2nd, 2018. Warning Signs? A Timeline of Tether and Bitfinex Events — Alex T. on CoinJournal, April 8th, 2018. The rise Of Stablecoins — Brian Quinlivan on Hackernoon, July 4th, 2018. Stablecoins — Carrying the Way to the Cryptocurrency Promised Land — Haseeb Quereshi on Hackernoon, Decembre 12th, 2017 Stablecoins: designing a price-stable currency — Preston Byrne on its personal blog, December 10th, 2017. Stablecoins are doomed to fail — MakerDAO’s blog, September 10th, 2015. What is MKR? Stablecoin projects mentioned in the article: USDT TrueUSD MakerDAO Basis Carbon

Tether

Dogecoin

Maker

TrueUSD

MakerDAO

Is stability a necessity for cryptocurrencies mainstream adoption?

Cryptojacking has been a recent buzz word in the cyber security world. It has been on the radar since August 2017 and had increased by an alarming rate of 8500% by December 2017. This noticeable increase is mainly due to the fact that cryptojacking is the easiest and most covert way to obtain money from a victim’s machine. It is far more discreet than ransomware, where access to the victim’s files are blocked or encrypted as part of the attack and then the attacker demands a ransom to decrypt the files. In the case of cryptojacking all the attacker has to do is infect a machine and it will generate money without the victim’s knowledge. cryptojacking and crypto mining are very similar, in fact crypto mining evolves into cryptojacking when it is installed or run on a victim’s machine without authorization. Cryptojacking is the use of computer power to mine cryptocurrency without the prior consent or authorization from its users. This can only be achieved through malicious activities or other means of deception, such as hidden code in the browser that runs undetected. Browser cryptojacking is more popular among cybercriminals than malware, mainly due to the fact that in-browser cryptojacking requires no installation to run, which makes it very easy to infect users machines and operate. HOW IT WORKS? The first method is deceiving the user to open a link or download a file (usually by social engineering), following which the script runs in the background unknown to the victim. The other method is in-browser cryptojacking which runs a code in the browser when a particular site is accessed. JavaScript is used to run the cryptojacking scripts that are usually concealed in an advertisement, which contributes to the silent nature of in-browser cryptojacking. Some attackers would incorporate both methods to maximize their profit from each victim. Cryptojacking scripts do not have the intention to harm the files of the infected machine, however they cause the machine to operate at a slower rate. STEPS: The threat actor compromises a website Users connect to the compromised website and the crypto mining script executes Users unknowingly start mining cryptocurrency on behalf of threat actor Upon successfully adding a new block to the blockchain, the threat actor receives a reward in cryptocurrency coins LOSSES January 2018: a crypto mining botnet infected computers in Russia, India and Taiwan. It is estimated that half-million computers were infected, the amount of mined cryptocurrency was valued at $3.6 million. February 2018: a cybersecurity firm that operates in Spain was victim to cryptojacking. WannaMine was the script used to infect the machines which was used to mine the cryptocurrency “Monero”. February 2018: the U.S. and the U.K. government’s websites were used for in-browser cryptojacking. The U.K.’s Information Commissioner’s office website was running the cryptojacking scripts, which also infected any visitor of the website. Furthermore, the American court system website had the same cryptojacking scripts. February 2018: Tesla Inc. had been affected by cryptojacking when its Amazon Web Services software container was compromised. Such attacks have been reported to have occurred to other companies and organizations dating back to october 2017. DETECTION METHODS The first sign is that computers perform abnormally slow for their processing power, so if machines have been identified with all of a sudden lower performance, this is red flag and may be an indicator of cryptojacking. Another indicator is overheating of systems due to the usage of CPU power, especially on mobile devices. Also check for CPU high usage spikes on PCs or mainframes, high CPU usage may also be an indicator of a cryptojacking activity. Specific network monitoring tools also help companies detect cryptojacking, and many agree that this is the best detection method for large corporates. PREVENTION METHODS Spread awareness about cryptojacking among users, so they are more vigilant towards malicious websites and links. Use extensions that block domains that associate themselves with cryptojacking scripts and restrict permission for unauthorized browser extensions from gaining access or executing processes. Use a mobile device management solution to better control what users can access or have on their devices. Restrict websites that deliver scripts to prevent in-browsing cryptojacking, so employees cannot access them. HOW TO MITIGATE A DETECTED CRYPTOJACKING INSTANCE Disable network privileges for any detected cryptojacking websites/scripts, since crypto mining requires sending data out as a proof of work, and isolate and scan the machines for these malicious files and remove them. If machine performance has decreased, then identify the process which is using the most memory and verify if it has any connection with mining processes. If it does, halt it and blacklist the process, so it can not be executed anymore. Common anti-malware detection methods as well as reputable internet security products are effective against cryptojacking script files, incorporating anti-malware methods to find and remove the malicious files. CONCLUSION Cryptojacking is a new money generating scheme that has taken over the threat landscape. Users must be aware of this new lucrative method, so they can prevent any cryptojacking from occurring on their machines. Furthermore users should follow the aforementioned recommendations for preventing a cryptojacking incident. Some individuals would argue that cryptojacking is a victimless crime, the foundation of this argument is that cryptojacking does not steal the confidential information of the target nor does it harm their files, this argument is unsound; crypto mining can harm your machine, in some cases it was reported there was physical damage to some of the victim machines due to the high usage of the processing power. In addition to that, the device will perform poorly and with a significantly lower remaining processing power.

Monero

Target

Tesla

CRYPTOJACKING?

Photo by on Mike Enerio Unsplash If you are one of the strong readers who made it to the end of my previous article, welcome back! If you are not, I strongly urge you to read the first part and get back after. _Today we will start building a Distributed Application from the ground up. It will use the Ethereum Blockchain…_hackernoon.com Dip Dapp Doe — Anatomy of an Ethereum distributed fair game (part 1) Today we are going to deploy the contracts and throw some color on the screen as we design and build a static web site to interact with the dapp. For speed and simplicity we will be using to render the page, to manage its state and to develop and bundle the static assets. React Redux Parcel Picture by publicdomainphotos As a reminder, the source code of the article can be found on the GitHub repo: _dip-dapp-doe - Distributed app featured in a Medium article_github.com ledfusion/dip-dapp-doe What are the main differences between a traditional web site and a Ðapp? Traditional web sites connect to a remote API, while Ðapps submit signed transactions to the using . Ethereum blockchain Web3 Traditional web sites are hosted in centralized servers and updated can be rolled out at any time, while Ðapps are mode of static files identified by an immutable hash. Traditional web sites need full net connectivity and HTTPS/TLS to trust the remote host and enforce data integrity/privacy. Ðapps could be opened from a local folder and transactions could be signed locally now and be relayed to the blockchain by an untrusted peer later. The key component to interact with the Ethereum blockchain is Web3. It is available in Javascript when using browsers like , when installing a plugin for / or when we manually import it in a Node script. Mist Chrome Firefox Before we use it to interact with any smart contract, the contract needs to be deployed to a public blockchain. And to do that, we need a wallet. Deploying the contract Among the alternatives to deploy contracts to the blockchain, we could use web wallets like , desktop wallets like or send a transaction with binary data from . MyEtherWallet Mist Metamask In our case we prefer and let a script compile, handle the bytecode, the linking and the transactions. not to depend on human interaction First, install the compiler globally on your computer. Go to the contracts folder and compile them: Solidity $ [sudo] npm install -g solc$ cd blockchain/contracts$ solcjs --bin --abi --optimize -o ../build *.sol For every contract, will generate an ABI file (JSON with the contract operations) and a BIN file (with the bytecode). solc Next, create the folder . Get into it, create an NPM file and install a couple of dependencies: contracts/deploy package.json $ npm init -y$ npm install truffle-hdwallet-provider web3 And then, create a couple of JS files that will do the work for us. A few of things to note: As we saw in the previous post, we need to deploy the library first, link it to the contract and deploy the contract after To link the library ourselves we need to replace a placeholder left by in the contract binary by the actual address of the library solc If the library is already deployed, we can use the existing address and skip deploying again The contract’s bytecode passed to needs to be prefixed with deploy() "0x" Our constructor expects a timeout parameter: by leaving it to zero, a 10 minutes interval will be used What are the and ? PROVIDER_URI WALLET_MNEMONIC A is the Ethereum API gateway that we will use to relay signed transactions to the blockchain. By now, our transactions will be sent to the and we are going to use . Web3 provider Ropsten test network Infura In order to prevent abuse, Infura requires to sign up for an Api Key but the service is free to use. Follow the process and copy the Ropsten URL for later. Getting a wallet are a friendly way for users to store and import crypto accounts among different wallets. They consists of plain English words that generate the public/private keys of a wallet and allow to restore it anywhere. . Wallet mnemonics More info here To create an account, open and generate a random set of valid words: this web site Copy the words and keep them for later. But if I use the seed they generate, they could go and steal my funds? To deploy to the Ethereum main net, you can generate your own seed off line by combining 12, 15, 18… of . Test ether is free to get so we don’t care at this point. We would obviously need to care if the contract creator had a special privilege on it. Otherwise, keep in mind that a normal contract deployment should cost less than 1–2$ on the main net. these valid words Let’s put the URL and the seed words we got into : blockchain/env.json {"PROVIDER_URI": " ","WALLET_MNEMONIC": "morning evolve industry guitar retire warfare wage wasp busy point dish globe spoil seven uphold"} https://ropsten.infura.io/--YOUR-API-KEY-HERE-- Our script is ready to run. However, if we do run it: $ node deployThe account used to deploy is 0x70cBDCC3C3A86c56deCBEcd3D4D5EE4017F65510Current balance:  0 Deploying LibString... Unable to deploy: insufficient funds for gas * price + value We created the account, but have no money yet to pay for gas. How do we get any Ropsten ether if no exchange works with it? By using the : Ropsten Faucet https://faucet.ropsten.be/ Copy the address above , paste it into the form field and submit it. After a few seconds, your balance should be more than enough to pay for transactions gas. 0x70cBD... Getting the contract deployed Drum roll… $ node deployThe account used to deploy is 0x70cBDCC3C3A86c56deCBEcd3D4D5EE4017F65510Current balance:  98802321000000000 Deploying LibString...- LibString deployed at 0x5184C01F47FAE8869C154771b7a3C16FcDae2b38 Deploying DipDappDoe...- DipDappDoe deployed at 0x232d56C9D7C8298f0ECFCf71a187CBAdDd85439c Yay! Our contract is now carved in stone on the Ethereum test network: LibString transaction DipDappDoe transaction , because that’s where our transactions will be sent to. Write down the DipDappDoe address Let’s check that they just work. Copy the following into a local script and replace the content of by your actual deployment address: blockchain/deploy/check-deployment.js CONTRACT_ADDRESS Above, we’re just creating a game and printing the current info of the game: GAME CREATED 0Result {'0': [ '0', '0', '0', '0', '0', '0', '0', '0', '0' ],'1': '0','2': '1000000000000000','3': 'James','4': '',cells: [ '0', '0', '0', '0', '0', '0', '0', '0', '0' ],status: '0',amount: '1000000000000000',nick1: 'James',nick2: '' } We get two versions of the same return data, but that contains the data that we care about. Cool! Let’s note one thing, though. If you recall the Javascript tests that we wrote with Truffle, you may notice that function invocations and transaction results look different. We need to be aware of this, because Truffle may be using a different Web3 version than the one we get from NPM. With that said, it’s time for a web to interact with the smart contract! Ðapp frontend architecture Unless our game players become expert developers, it would be unlikely for anyone to play DipDappDoe without a UI. Let’s craft a static web page so that they can play like human beings. First off, inside the folder, let’s start a NodeJS file and install a few dependencies: web package.json $ [sudo] npm install -g parcel-bundler$ npm init -y$ npm install react react-dom react-router-dom redux react-redux redux-thunk web3 Let’s create with the following starter HTML: web/src/index.html Then, create Javascript file referenced from with a dummy instruction to check that it works: web/src/index.js index.html alert("HELLO DIP DAP DOE"); Next, run the ParcelJS bundler: $ parcel src/index.html And visit http://localhost:1234/: Frontend overview Next, we need to define the global structure of our site: When Web3 is not available, we will ask the user to install MetaMask for Chrome/Firefox When Web3 is locked, we will ask the user to unlock it Otherwise we will show the main view, with the list of open games When we create or accept a game, we will jump to the game view The main view will allow to join available games, as well as creating a new one. The game view will display the Tic-Tac-Toe board, as well as the Withdraw button and the game timers. In a normal project, we will be working side by side with a , but in today’s article, our goal is to demonstrate the dapp’s technology. To the extent of my abilities, this is the design that I have come up to: designer Main view and creation view design (Mobile) — Background image by Max Pixel Main view with game and creation cards design (Desktop) — Background image by Max Pixel Game view design (Mobile) — Background image by Max Pixel Game view design (Desktop) — Background image by Max Pixel Let’s add some code to start shaping our layout: This is a very simple skeleton with basic detection of Web3 support to show dummy messages accordingly. than the example above. Detecting the current browser and guiding the user is a mandatory UI/UX step but this is a topic for another article. The full casuistry is much deeper As soon as our views spread into separate files and different components need to work with the same state, we will realize that passing the local state from to all of of its children is not viable. app.js That’s where comes into play. Redux provides global state management for React apps, so that any component can dispatch actions to modify the state and retrieve the subset of information that it needs. More about Redux . redux here Let’s import the library, create a store and add a couple of reducers to the project: Let’s get back to . We left the main component with a few local state variables. web/src/app.js Now that we have a global state manager, let’s refactor our code to dispatch actions as soon as we become aware of Web3 events. The wrapper injects the function and the reducers’ state into the component’s . As soon as its state changes, Redux will trigger a re-render of the component. connect(...)(App) dispatch accounts/status props If we refresh the web at this point, we will see the message . “Please, install Metamask for Chrome or Firefox” If you haven’t done already, visit and install the plugin in your browser. Feel free to import your wallet with the mnemonic you generated earlier. If you refresh again, you should see text, which is our main view’s placeholder. So there we are! https://www.metamask.io/ “Main View” A contract wrapper We compiled and deployed the contract before. Now we need to tell Web3 again about the contract. This time, we only need the deployed address and the contract’s JSON ABI. Since the ABI is already compiled, a clean thing to do is to link it from the already compiled file. Any contract operation added in the future will automatically reflect on the other side too. $ mkdir web/src/contracts$ cd web/src/contracts$ ln -s ../../../blockchain/build/__contracts_DipDappDoe_sol_DipDappDoe.abi dip-dapp-doe.json Also, let’s create and export a DipDappDoe contract’s instance: Our contract is ready to be interacted with and our web app is capable of handling global state changes. Now it’s time for UI. UI components Let’s do some tweaks to the UI by adding React Media and Ant Design. React Media will be useful to render content targeted for mobile or desktop browsers. Ant Design will provide us with nice widgets, dialogs and UI notifications. $ npm install react-media antd Our main view skeleton would look like this with them in action: As you can see: We are retrieving via the props we tell Redux to inject openGames We are using a media query to detect whether we will render the mobile components or the desktop ones We also use the grid system from antd After a bit of work, markup and styling, the main page will look like this: Desktop version of the main view Mobile version of the main view The main difference will be that in order to create a game in the mobile version, we will need to tap the start button to reveal the creation card. How would our first contract invocation look like? We have added a new action called in the reducer. When we create a game we need to store the random number and the salt in order to confirm it later. ADD_CREATED_GAME status Web3, MetaMask and events : Because of a limitation of MetaMask, we can not subscribe to events because they work on top of Web Sockets. But MetaMask only supports HTTP providers. Problem This leads us to using two Web3 instances in the meantime: One instance using Infura as the Web Socket provider to listen to the contract events The MetaMask instance to sign and launch transactions from our personal wallet : At the time of writing, Web3 version 1.0.0-beta-34 and beta.35 have issues in connecting to Infura via WS. Web3 beta 36 is expected to address the issue, but meanwhile, we will need to downgrade to Beta 33. Warning So we will have a Web3 instance and a one. Wouldn’t it be nice to wrap the logic into a helper and provide the to everyone? No need to open many Web Sockets if a single instance could do the same job. read-only read-write singleton We will need to refactor our code and make sure that event listeners and transaction senders use the appropriate instance. But let’s do it! Code where was used, now invokes either or instead. web3 getInjectedWeb3() getWebSocketWeb3() Similar thing with the contract instance. Contract event listeners can retrieve and signing transactions can be done with . getDipDappDoeInstance() getDipDappDoeInstance(true) Now our event listeners look like: const DipDappDoe = getDipDappDoeInstance() this.creationEvent = DipDappDoe.events.GameCreated({// filter: ...fromBlock: this.props.status.startingBlock || 0}).on('data', event => this.onGameCreated(event)).on('error', err => message.error(err && err.message || err)) As you see, we ask for events at the block number where we loaded the dapp or later. Updating a contract Shouldn’t we filter events to get only those about active games that affect us? Absolutely, listening to just all events would be inefficient. To remain minimal, we need to . And for this filter to work, we need to add it as a parameter to the existing events. filter by the opponent’s address What that means: Making the change (quick) Adding testcases (simple) Ensuring that testing works (automated) Compiling and deploying again (automated) Luckily for us, the effort spent into TDD pays off with a seamless update. Now our events look like that: GameCreated( gameIdx);GameAccepted( gameIdx, opponent);GameStarted( gameIdx, opponent);PositionMarked( gameIdx, opponent);GameEnded( gameIdx, opponent); uint32 indexed uint32 indexed address indexed uint32 indexed address indexed uint32 indexed address indexed uint32 indexed address indexed On we will just want to refresh the list of games, so there is no need to add any parameter. GameCreated After rerunning our deployment script, the new contract : is deployed $ node deploy The account used to deploy is 0x70cBDCC3C3A86c56deCBEcd3D4D5EE4017F65510Current balance:  93111123000000000 Deploying LibString...- LibString deployed at 0x5eF66e8df15eC7940639E054548a0d7B95535529 Deploying DipDappDoe...- DipDappDoe deployed at 0xf42F14d2cE796fec7Cd8a2D575dDCe402F2f3F8F After the update When we tell our frontend to attach to the new instance, what will happen? The current logic, transactions, games and money will remain in the old instance forever We have no special control over it; we could have coded a mechanism to allow the creator it or refund the money but it’s not the case kill() The only way now is to let timeout’s expire and withdraw any deposits from the player accounts As soon as we update the frontend, users will be directed to a new contract with fresh data There are ways to and allow the successor to take over the legacy state, but this would allow for a whole article on its own fork from one contract Since we are developing the dapp, this is no issue at all now. Event handling Back to the events! Now we can start setting the parameter and filter the emitted events. opponent How would it look now? let event = this.DipDappDoe.events.GameAccepted({filter: { opponent: this.props.accounts[0] },fromBlock: this.props.status.startingBlock || 0}).on('data', event => this.onGameAccepted(event)).on('error', err => message.error(err && err.message || err)) // later onevent.unsubscribe() In places like the main view, we will want to be notified in any case. When anyone accepts, the game will disappear from the available games and we will want to know when to reload. However, in the game view, we will want to be notified only when a game of ours is accepted, started, updated or ended. Only in this case, we will make use of the filter. UI recap After the tools to launch transactions and receive event notifications are ready, what’s to be expected from the frontend? In the main view: Whenever someone creates a game, we want to update the list of open games Whenever a game is accepted, we need to update the list so that it doesn’t appear anymore When we create a new game, we want to jump into its corresponding game screen When we accept an existing game, we want to jump to the game screen as well So, in the game view: When an opponent accepts our game, our dapp should update the game status, retrieve our random number and salt and confirm the game for us When our opponent confirms, the game status should be updated and the player that can start should be told to mark a cell When our opponent marks a cell, the game status should be updated too, and we should be notified that it is now our turn When the game ends, the game status should be refreshed and both players should be notified Whenever a withdraw can be made, the button should be visible to the affected users Summary At this point, we have a clear image of the frontend’s architecture, how we should deal with events and what needs to be done at every stage of a game. We have also seen a few changes that the smart contract should include in order to improve the user experience. What happens to Test Driven Development on the frontend? So far, we have been exploring how to assemble a React+Redux application in order to get Web3 to connect to the blockchain and have the relevant state updated. Now that the skeleton is ready, again, we should write the specs first and work the functionality after. Shouldn’t you have started with TDD’s at the very beginning? Sure, but in this case it’s quite helpful to showcase the frontend architecture instead of jumping into it with no introduction. In normal frontend testing we can simulate clicks and events, but how can we simulate and automate MetaMask confirmations? This is what we are going to explore in the third part of the article, along with bundling the app and publishing it with IPFS. This concludes part #2 on our journey to build a distributed application running on the Ethereum blockchain. The code featured on the article can be found in the GitHub repo: _dip-dapp-doe — Distributed app featured in a Medium article_github.com ledfusion/dip-dapp-doe If you found the article useful, clap 👏, smile 😊, like 👍 and share with the same 💚 as I wrote it. After you take a well deserved break, part 3 is awaiting you, here: _Today we are going to follow the Test Driven Development methodology on our frontend, along with Web3…_hackernoon.com Dip Dapp Doe — Anatomy of an Ethereum distributed fair game (part 3) Photo by on Mike Enerio Unsplash

Dapp

Mozilla

Dip Dapp Doe — Anatomy of an Ethereum distributed fair game (part 2)

is one of core values. Excellence over Mediocrity Quovantis’s I know what you’re thinking- Duh! Shouldn’t it be obvious? Who wouldn’t want to do their best work every day? Otherwise, why would we choose to dress up in the morning and brave a long commute to come to work to only do mediocre work? Unfortunately, mediocrity exists. It not only exists but thrives all around us. We even do mediocre work at our workplaces, which are built to . put the proverbial dent in the universe I understand that every day isn’t created equal. There are days we have to carry the emotional burdens of our lives and we end up carrying fragments of those emotions to our workplace because, sometimes, they are too precious to be left at the doorstep of our workplace. I totally get it and its ramification on the quality of our work. But I’m talking about the other normal days when the skies are blue, birds are chirping their favorite song, our in-laws love us and there is a feeling to do shit that matters. perceived And you know what, sadly even in those days mediocrity happens. It happens so much that it sometimes ends up blinding us to see what great work looks like. If left unchecked, it builds like a termite infestation that eats our passion and makes us hollow from inside. So why does it happen? Why do people willingly choose the dark cloak of mediocrity when they can shine like the stars? What propels people to continue laboring under its weight, without wanting to break free? In my quest to find these answers, I have found the following of mediocrity that are universally applicable- causes Sense of Futility The person feels that whatever they create doesn’t matter and would go unnoticed. It may be because they have never heard a word of appreciation from their peers or leaders. Thus what difference does it make if it’s a work of art or worthless mediocre crap, right? This kind of thought process is generally either self-inflicted misery or cultural conditioning passed on as a virus by people around us. Too Much Effort Who on the earth has the time to put in those long days and years to create a masterpiece when something mediocre or ‘just good enough’ could be passed as work? The person feels that they have made it till here and no one has objected so far! If anyone needs a better version of it, there would be enough people around to take that work forward and make it ‘perfect’. One doesn’t know what great work looks like Someone hasn’t experienced the craft of the highest caliber and doesn’t even know what it looks like. It’s the case when we don’t know what we don’t know — . Unknown Unknowns The person is just happy with whatever comes their way because no one has ever questioned their methods and they’re too happy with their bare minimum standards. Not connected with the cause “Why do we need to put in so many efforts?” “What good will it do to others?” “Why can’t someone else do this?” One generally wonders about such things when they aren’t aligned with the purpose of the work — the “why” of it, like Simon Sinek would say. Bottom line, the person isn’t emotionally vested in a job to do an excellent job. You know the best part in all the above scenarios is that we have an option. The option is to have an ego. No, not like an egotistical maniac but – Yes, Ego! Ego from a perspective that I don’t care if I’m appreciated but my work got to be great because my name is written on it. Ego from a standpoint that I’ll push my limits and go an extra mile to make my work look like a piece of art. Ego to commit that even if I have to plunge into my unknown unknowns to discover new things, I’ll do so without batting an eyelid. Ego to make efforts to find meaning in the work that I do. That’s the kind of attitude to build anything worth talking about and that’s the only way to leave a legacy. This is a part of our Friday newsletter about professional growth. Here’s the archive if you would like to read the previous newsletters.

Seeing Mediocrity all around? These could be the reasons.

He gave birth to beautiful paintings on a desert canvas. He gazed at them for hours which turned into days; locked in his workshop, he only prayed to the powers of his vision. He never shared his work. It was his to create, his to admire and only his to abandon. Creating only for and out of himself was his principle ideology. The word about the artist spread through the rustic village where the artist had migrated. Villagers hastened by his workshop in the evening to experience what they thought would be a pleasant escape from their routines. The artist threatened them, “If any one of you ever visits me again, I will burn each one of you alive and cast my paintings in the same fire to fuel it. You can then look at the paintings for as long as you want”. Under the soft light of the moon, his words blazed through his stormy eyes; no one spoke a word further. The villagers declared the man insane. They realized he was just a deceiver, he was no artist; just a drunkard who had lost his mind to some shattering truths of life. The artist continued painting through the years. On one evening when the sky was a particular , a wee boy was returning home with a sachet of oil paints and a sheet of paper. A mighty gust brought the fragile body of the boy onto the ground and the wind stole the sheet from his fingers and spit it in a big, hollow chamber of what looked like emptiness. The boy ran for it. He climbed the gate, ran further through an open door and steeply landed right in the middle of the chamber. He picked up his sheet of paper and smiled with utmost joy after taking another glimpse at the result of his favourite hobby. His eyes gently settled on another canvas. Not on the one in his hands. The canvas was stained with almost a replica of what the boy had painted a few hours ago on his own sheet. A shade to shade cast, a stroke to scene mimicry. The artist appeared with a glass of water in his hands and a bucket of rage in his eyes. rare shade of midnight blue intertwined with a hue of olive “How dare you enter my temple?” he screamed with all the weight of the words he hadn’t spoken in years. The artist’s eyes fell on the kids painting. The buckets of rage were now upside down and he remained still for an unknown period of time. Encouraged by the calmness in the atmosphere, the kid asked him, “Where did you learn to paint like that, sir? The sky, the river? I did not see you by the bank of the river”.The artist replied with explicit calm and broken thoughts, “Years of training. For years, many years’ worth of paint has dropped by as I have crafted thousands of skies until my vision, my devotion, my patience, and my intellect molded in harmony to give birth to this shade of sky. Through my own trials and my own learnings, I came up with this particular shade of sky. I learned to paint only from myself, and only for myself. How many years have you been painting for? How did you learn to paint that shade, kid?” The kid turned red with joy. “Oh, I wish I could learn to do it like you one day, sir. It is beautiful. Why don’t you ever share it with the villagers? The villagers say you’re egoistic, some say you can’t dissolve criticism. My mother thinks you’re hurt and so you’ve closed. My sister told me that maybe your ideas are so ahead of its time that they won’t be accepted and so you rejected the society before even giving it a chance.”“Enough with the questions. Where did you learn to paint like that? Who helped you imagine that? You copied my work, didn’t you, you lying filth?”, the artist demanded.The kid immediately chirped, “Oh no, sir. I have no powers of imagination like you, sir. I have no magic to think ahead of my time, I can just see what is now. I just went outside by the river and the sky shared with me its . It’s just what the evening looks like today, sir. Had the sky not shared itself, how could I have ever known that the sky is blue and that it is above and not below?” he giggled and ran towards the gate hopping in the wind, trying to hug the clouds. shade of midnight blue with a hue of olive The artist who never shared, shared the ache of seeing the sun behind the moon. He saw that the sun had all along shared how much he had blinded himself in his pursuit of excellence and being the best version of himself, but truly, just being a better version than others. The dark of the sun had always emitted his intoxication in trying to create a simulation of the very world he lived in, trying to compose a perfect symphony, that he had drowned deep enough to not be able to breathe the world around him, to breathe nature and what it already has to offer, because nature has always been willing to share. A tale by Rohan Dave.Late Summer ’18. San Francisco.

Harmony

The Artist Who Never Shared His Art

“In God we trust, all others must bring data” ~ W.E. Deming The rise of as a to package & run software has been nothing short of an urban legend coming true. In this article I’ll use data points to showcase the sheer magnitude of the success of the technology & highlight the trends of container adoption across enterprises. docker tool This article is the first part in a series of posts, in the next i’ll discuss the challenges that organizations face while trying to adopt containers. article is an open platform to build, ship, and run distributed applications on laptops, data center VMs, or the cloud. It makes it possible to separate the applications from the infrastructure and also to package & run an application in a loosely isolated environment called a container. These attributes enable teams to deliver software faster & more efficiently. Docker The best way to understand the growth of the docker platform is to look at Figure’s 1.0 & 1.1. These charts provide a direct indication of the users interacting with the docker platform, by plotting the cumulative sum of images downloaded from the official docker store over time. As is evident from Fig 1.0, it took about two years (somewhere in the mid of 2015) for the technology to reach the , post which the adoption has grown in an exponential manner. Tipping Point At the time of writing this article about images were being downloaded from the online docker store every 2 weeks & around images have been downloaded till date. In the last one year more than new developers jumped onto the docker bandwagon. 1 Billion 38 Billion 1 Million Adoption rates such as these are almost unprecedented in the software industry & were partly fueled by the founder’s strategy to the technology early in its life cycle. Today a large active community is contributing to the success of this platform open source (source: blog.docker.com) Fig 1.1 We would have to go way back to the beginning of the millennium to find another software that met with such super success early in it existence. It was in early 2000 that Salesforce flew out of the blocks after the maverick & controversial advertising campaign and has never looked back since. “End Of Software” The great thing about docker is that it alleviates pain points for technologists across all spectrums of the SDLC. As a result Docker found early promoters in the best of breed architects, developers, testers as well as operations engineers. These early adopters were keen on sharing the knowledge they had gained and desired to collaborate with like minded individuals to enable the success of the technology. Hackathons were organized, meet-ups were planned, blogs written, online webinars created. In short everything possible was done to share the knowledge! The wider community came on board and used the technology to resolve their existing pain points, experienced the benefits and pushed for adoption within their organizations. What followed was a classic technology adoption across organizations of all sizes and businesses, as depicted below in Fig 2.0. Pull based Fig 2.0 As the the container technology became exceedingly popular the eco-system around it also began to grow at an astounding pace. “ with docker that all the technology behemoths & bellwethers of the day jumped into the fray and started investing in creating their own platforms for providing docker based services, products or managed platforms. The force was so strong” (source: blog.docker.com) Fig 2.1 By now most enterprises had already started experimenting with the technology in non-production environments in some form or the other. As the eco-system matured, more and more production workloads were being moved over to containers, traditional applications were being containerized & a very high percentage of new applications were designed to be docker compatible. As of 2018, docker has proved its production capabilities and based on a report by Datadog around 25% of their enterprise customers have adopted Containers in production & only 8 % have abandoned the technology . Most of the organizations that are dabbling with the technology also become adopters in the long run. The adoption trend line has consistently stayed positive and shows no signs of tapering, at this rate around of all enterprises would be using containers in production by 2021. (Fig 3.0) 45–50 % (source: Datadog) Fig 3.0 What is extremely important to note here is that once an Enterprise decides to go with containers, the footprint of the workloads in containers keeps on growing at a breakneck pace, with the deployment rate increasing by 75 % in the last one year alone as depicted in . Also noteworthy is the fact that the deployment rate has been increasing year on year. This data point is in itself a testimony to the value that containers bring to an enterprise. Fig 3.1 (source: Datadog) Fig 3.1 While the use cases for containers can differ from enterprise to enterprise, the benefits accrued tend to be consistent if the container strategy is well thought through. Use of docker brings substantial improvements to MTTR , Developer Productivity & reduction in the no of VM’s & server costs These improvements would lead to much faster time to market, better user experience & lower costs. (~ 60 %) (~65%) (~ 45%). (source: state of App Development survey) Fig 3.2 Another important aspect of the adoption of the containers is the fact that most Enterprises are using managed container platforms to deploy containers in production across on-Premise and cloud based setups ( It can also be inferred from the below data that most enterprises are choosing to use the managed container platform of the cloud vendor that they are already engaged with. While Kubernetes & Docker Swarm have emerged as the dominant orchestrator engines for running production workloads. Fig 3.3). (source RightScale ) Fig 3.2 As the container adoption grow and the eco-system keeps on evolving it would become increasingly important to learn for the missteps and create blue prints for successfully moving to a containerized environment & that is what the focus of the existing IT teams should be. In the next article I discuss the challenges faced by organizations while adoption containers.Find the link here . Challenges in Container Adoption

Community Is

Salesforce

Super

Democratization of Container Technologies

This is going to be controversial. I believe “ needs to die, for and crypto to flourish." Bitcoin blockchain Here is why I say this. We have wasted a good part of the last 10 years running after bitcoin. It is just infatuation driven by the rising valuation of bitcoin. Bitcoin has achieved something quite remarkable. It has shown the world how a digital currency can truly work. It has demonstrated how transactions can work without any need of an intermediator to ensure trust. Bitcoin has pioneered trustless systems. It has achieved its purpose. Bitcoin could not become a digital currency. It was initially projected as something which would replace fiat. It failed. Quite understandably so. When it failed to replace fiat, people - infatuated with bitcoin, started pushing the theory of "store of value" or equivalent of digital gold. It’s going to fail at this too. ( I believe it has failed already.) Not to mention, it is moving towards centralization. 15 mining pools control more than 75%(or more) of the mining. I think, Bitcoin is never going to touch 100k or 50k or even 20k again, not in the next 3-5 years. It has already peaked. It is going to stay range bound, and keep making new lows, and lower highs. In other words, the store of value, will continue losing value. And this is a good thing for blockchain and crypto. A blessing in disguise ! Sooner than later, people will start understanding this. Good blockchain and crypto projects will come out of the shadow of bitcoin. They will innovate and solve real world problems. One of this will replace fiat, or may be multiple of them. Some will solve real issues like universal basic income, corruption, taxes, marketplaces, true decentralization, etc. And no, projects promising million transactions per second, are not going to succeed either. They are not solving any real world problems. They are building a solution, looking for a problem. Anyways, I digress. Good blockchain projects will come out of shadow of bitcoin. Sooner than later. The technology will evolve. It may not be blockchain, but a derivative of blockchain, with similar underline core features. Trustless, immutable, and decentralized. They will see larger adoption. 100x better than bitcoin is today. (Is bitcoin used by 20+ million unique users, or a fraction of it? ) The new tech will become mainstream and ubiquitous. Governments will not ban or block them, but adopt them en-mass. Bitcoin will be dead. Well, not exactly dead. Not in the top 50 (or even 100) cryptos in terms of market cap. No one will bother about its value anymore. It will be as good as dead. The sooner this happens, the better it is for the community. This is definitely going to happen in the next 3-5 years. The technology has to be evolve and come out of its shell that it is in today. Think of the first successful airplane by Wright Brothers. It had to go through multiple iterations for commercial air travel to be possible. (Would be like to fly, from one city to another, in the first airplane invented by Wright Brothers?) Similarly, Satoshi invented the first decentralized, trustless, immutable digital currency, bitcoin. The technology needs to evolve. It needs to get ready for commercial flights. Under the shadow of bitcoin, this evolution is suffering. Real use cases and application will lead the technology evolution. Not the other way round. As I said it before, Bitcoin needs to die for blockchain and crypto to flourish! Sooner or later, it should happen. (hopefully sooner) The future is bright. Sit tight!

Bitcoin needs to die, for blockchain and crypto to flourish

The State Of The Market: The total market cap went below $200 Billion for the first time in 2018, to a low of $192 Billion. However, towards the weekend, the recovery was swift and the market is now at $215 Billion. Most cryptocurrencies went down by more than 10% during the crash, except Bitcoin. Bitcoin (BTC) remained stable above $6,000 as Ethereum (ETH) touched a low of $250. The recovery was led by Ethereum, which is now back up at $302. Ethereum Classic (ETC) is finally live on Coinbase and Coinbase Index Fund. The exchange took two months to test ETC after initially announcing support for it in June. Right after it went live on Thursday, the price of Ethereum Classic went up from $12 to $15. Ethereum Classic (ETC) is currently priced at $13.38, losing 3.87% in the last 24 hours ( ). read more After months of teasing, Binance finally launched its first Fiat-to-Crypto exchange in Liechtenstein, in partnership with Liechtenstein Cryptoassets Exchange (LCX). Liechtenstein is a German-speaking country in Europe which shares its borders with Austria and Switzerland. Binance will provide technical support and infrastructure to run the exchange, while LCX will take care of compliance and other regulations ( ). read more Fidelity Investments, which manages $2.4 Trillion worth assets began offering Bitcoin exposure to its clients through Bitcoin Exchange Traded Notes (ETN). The ETN is offered by XBT provider on Nasdaq on Fidelity’s OTC market against US Dollar. This is very different from ETFs because XBT buys and holds Bitcoin purchased by its clients ( ). read more Subscribe to the Berminal Brief Newsletter Download Berminal for Free The Berminal Website Berminal Official Telegram Berminal Twitter

Ethereum Classic

Binance

Coinbase

Twitter

Weekend Berminal Brief (8/19/2018): Ethereum Classic Goes Live On Coinbase And Binance Launches…

The crypto markets are still on a recovery phase after a short dip in the last few weeks. The end of next month, however, holds a lot of possibilities for the crypto space as some significant events that would affect the direction of the industry to a great extent have been scheduled to take place. Invitations for ICO Regulation A recent has revealed that numerous big players in the cryptocurrency industry have been invited to a meeting in order to deliberate on issues regarding the regulation of (ICOs) in the United States. report Initial Coin Offerings Thirty-two crypto-related companies, including Ripple, Kraken, CME Group, Coin Center, ICE, and Circle received the invitation from US Congressman Warren Davidson, a man who is known to have a keen interest in the crypto space with many positive comments relating to ICOs and virtual currencies. The congressman intends to “lead” the ICO legislation. He believes that the lack of regulation is the primary reason why the US crypto market looks risky and unsafe; hence giving rooms for fraudulent coin offerings. According to him during an , a “light touch regulatory framework” would provide “certainty for the ICO market.” interview with CNBC The meeting which is scheduled for September 25th, 2018, will cover different crypto-related issues including how to protect investors, the responsibilities of regulators, and how to provide a friendly ecosystem for startups to raise capital. SEC final Decision On CBOE Bitcoin ETF Application A second rejection of the BTC ETF proposal submitted by the Winklevoss brothers on July this year, sent the crypto markets into a short-term dip. However, as always, members of the crypto community remained confident that the ongoing market correction would follow. The SEC rejected the application stating that the ETF is not in line with the Securities Exchange Act. The regulators also referred to a that claimed that the price of BTC could be subjected to manipulation. Nonetheless, SEC Commissioner Hester Peirce publicly disagreed with the decision. recent academic research paper ” Peirce stated. “I take the position that actually the change that was put before us was consistent with the Exchange Act, However, a recent signed by Eduardo Aleman and published on the commission’s website revealed that the SEC had scheduled September 30 as the date to make the final decision concerning the Bitcoin ETF proposal. The report also stated that the reason for extending the date is that the SEC finds its appropriate to do so. report ” the report reads. “The Commission finds that it is appropriate to designate a longer period within which to take action on the proposed rule change so that it has sufficient time to consider the proposed rule change, Conclusion Indeed, September is a critical month for the cryptospace. If after the meeting with the US regulators and a new legislative framework for the US ICO market is published, then more businesses would be able to seek for capital through ICOs. And if the SEC approves the BTC ETF application of the Winklevoss twins, then investors would be able to trade Shares of SolidX bitcoin, which will eventually attract more institutional investors into the crypto space, and hopefully, . We can only hope that things go well. change the crypto market Originally published at cryptopotato.com on August 19, 2018.

September Could be a Critical Month For The Crypto Industry

What breakfast cereals, FOMO3D and a Shakespeare Doge all have in common… Introduction In my previous posts ( , ), I looked at various token designs that could be useful for capturing value. Basically, the aim of a crypto asset should be to become a Store of Value, charge fees or confer control. I believe that whilst the SoV argument is fairly well known, the ability for a crypto system to charge fees or deliver valuable control is often disputed (“just fork it”). part 1 part 2 crypto Q: How can a crypto system sustainably charge fees in a world of interoperable & open-source tech? A: The end-user is king. In this somewhat familiar B2C landscape, a focus on brand loyalty, switching costs, and network effects of scale could be key. For projects that don’t charge fees, governance can have a role in value creation. Climbing the Stack When thinking about where value may accrue across the crypto landscape, I found it helpful to think of the ‘web3.0’ tech stack and the type of user at each level. Stephan Tual’s Web Stack There has been an understandably large focus on the infrastructure level of the stack, with developers and investors working together to create potentially lucrative building blocks. However, if the technology becomes interoperable (the big assumption) then these building blocks become modular and interchangeable rather than ecosystem specific. These modularised components begin to lose their ability to extract fees and instead, value is only found at the top of the stack where the system interacts with the end user. This is an application of ‘aggregation theory’ first coined by Ben Thompson of Stratechery and later applied to crypto thinking here and here. With crypto, the differences between the users interacting with the technology is key. For example, a dev interacting with systems at the base/mid of the stack will be optimizing for ‘performance at a price’ and it’s their full-time focus to scour the landscape for this. Whereas at the top of the stack, if a ‘retail’ user of a fee charging dApp encountered a similar looking free version, would they bother to switch? How different is a retail user’s decision process compared to a professional developer’s? This is key to understanding the potential for sustainable crypto value capture. Hint: I believe the strength of the relationship between the end user and the crypto system will dictate this. The dev team in action Augur: dApp or Protocol? How can a dApp or protocol strengthen the relationship with its end user? Which aspects of a UX will be important to generate loyalty? I find it helpful when thinking about this to consider the popular example of the decentralized prediction market, Augur. This project charges fees for resolving prediction markets and these fees are distributed to the participating reporters via the REP token. Augur can be viewed as both a dApp and a protocol. When operating as a dApp, the Augur platform is facing the bet-placing end user. This is done through their proprietary . Whereas when operating as a protocol, Augur’s reporters and prediction marketplace provide a backend service for a competing 3rd party website. web app How could the fee potential of Augur’s REP token vary under these differing scenarios? Operating as a dApp: an exercise in consumer loyalty The user interacts directly with Augur through a proprietary web interface. Frequent, favourable experiences will reinforce classic consumer behaviours such as familiarity (saving the website to favourites), word of mouth (desire to use a service recommended by a friend) and ironically for a decentralised system, brand loyalty and trust. Why trust? Because when websites appeared, consumers were initially hesitant to enter their credit card details into these new systems. Likewise, it’s not unreasonable to expect similar scepticism from new crypto dApp users when presented with a new platform or forked ‘pirate’ version, if the user’s original crypto dApp is operating smoothly. A familiar use-case for crypto? And branding? Due to the complex nature of the service technology, mainstream users may be unwilling or unable to evaluate the potential pro/cons for a competing service, even if offered at a lower price. Instead of focussing on the ‘tough’ technical questions, the consumer may look for easier ways to resolve the consumption decision- a phenomenon referred to as ‘ ’. It’s possible that in this scenario, a strong dApp brand could dominate the decision process for a consumer. attribute substitution Likewise, ‘crypto’ itself can be a powerful brand when competing against the incumbents. The popular 'one liner' of the public is that a crypto system is ‘decentralised’. If a crypto project’s service is neither cheaper nor faster than the incumbent it may still attract users by offering something . Through changing the conversation to one instead about ‘decentralisation’, a consumer loyalty could be developed. centralised different The result of this could be that ‘lazy’ consumers stick with a service that is more expensive but perceived to be better (or different). This prevents the platform from leaking users to competitors and so maintaining its network effect of a critical operating size. The power of successful branding Operating as a protocol: the cost of becoming a module The user interacts with a 3rd party application that is built upon Augur’s prediction market. For example, a website specialising in financial market bets. The potential for Augur to charge fees has now been reduced for two reasons. Firstly, if there is another way for this financial prediction website to get the outcome data (e.g. conventional centralised service or an Augur competitor, Gnosis) then Augur could be modularised and replaced by the informed and fee conscious developer building the finance website. Secondly, if the Augur fees are excessive but there are no alternatives to its service, the 3rd party could consider forking the protocol. Kyle examines this scenario for ‘AugurCo’ . here However, it’s worth noting that there are limitations to protocol forking. Firstly, the new fork must directly contend with the established original. As per above, the consumers may be brand conscious, technically unskilled and ill-informed, making them unaware or reluctant to try the cheaper fork. Secondly, due to this competition for users it may be hard to achieve the required network scale and so worsen the UX, especially if it’s a marketplace service. And third, the security of the forked protocol may suffer if insufficient token value is achieved. To B2C? I think this example shows that the end consumer has the potential to be less fee-sensitive than a professional developer. Given this, I think that if a project’s interactions drift from direct user contact, from dApp to a protocol, the potential to be modularised increases. “To B2C…or not to B2C?” This is a similar argument to the relative merits of a B2C vs B2B business model. Since I am imagining this debate to be taking place in a future world with interoperable and technically replicable protocols, a B2C model seems to be more lucrative. However, this is clearly a crucial assumption. If you have a strong view or insight into this, please comment! blockchain Fattening the protocols How can a dApp/protocol’s fee and value capture potential be increased? If it doesn’t charge fees, apart from a SoV, how else can its value capture be increased? The ideas that I discuss above suggest that for a fee charging protocol, a first action may be to avoid providing mid/base level services and instead focus on the top of the stack, serving the more malleable consumer. The next action is to keep this user happy by i) satisfying competitive niches and maintaining a superior UX. This superior UX could be achieved by ii) generating switching costs for the providers of the platform’s service, making it harder for competitors to launch in a sufficient scale, and so concentrating the participants to one network. For protocols that do not charge fees, iii) governance functions may be able to increase value capture. Lastly, all of these actions to fatten the protocol/dApp should hopefully be reinforced by the iv) size of the network required to operate the service. i) Competitive niches Here’s an example of how to proactively fill competitive niches. If Augur was to launch a dedicated web interface (using a unique brand?) for each of the key prediction markets, users may be less likely to switch to a specialist competitor- an approach known as ‘brand proliferation’. This classic was seen in the breakfast cereal market in the 1970s and studies have attempted to explain why we have such a huge choice of cereals (and often provided by only a few companies). A common theory is that it is a strategy to ‘cover all the bases’ in order to prevent a competitor from entering. case study Bitflakes…c_oming to a store near you_ ii) Switching costs By introducing switching costs to providers of a protocol’s service, competitors or forks may find it harder to achieve the necessary operating scale for a good UX. This is one of the bootstrapping benefits of ‘work tokens’ such as Augur, which operate like a taxi license for platform service providers (see ). Further to this, work projects could increase their supplier switching costs by taking a page from Uber’s book. Uber offer services to their drivers such as car and discounted in an attempt to reduce churn. Similar to this, Augur could provide financing to new reporters, allowing them to pay for their required REP tokens over time. This becomes particularly effective when identity (and legality, see ) is added to the crypto system- making any default from this finance model increasingly costly. here financing insurance Mattereum A less conventional method for creating switching costs for service providers could lie within the popular dApp, . I am totally fascinated with the psychology and economic paradox behind this seemingly simple game (with a of 21.8k ETH and growing!). So simple that at first glance, I initially dismissed FOMO3D as an outright ! What I find particularly neat is a) using perpetual dividends to entice the initial users. This leapfrogs the backward induction obstacle whereby it’s usually not logical to begin playing these types of . And b) showing that it’s not a scam via open-source immutable smart contracts. This dApp is leveraging many blockchain powers…albeit for a fairly irrelevant purpose! FOMO3D prize pot scam games I think that transferring elements of this game to another dApp or protocol could harness its mechanisms in a more productive way. For example, Augur could integrate it in order to keep reporters loyal to the REP token system. Augur could have a mini version of the FOMO game, whereby a pot of ETH is awarded to the last reporter to have participated on the platform and so the lure of this pot is used to ‘jump start’ the reporting/betting. Forked versions of Augur would not have this prize pot of ETH available to reporters. These prize pots could even be partly funded by community resources eg from the Augur treasury, as it could be argued that it is enhancing overall platform liquidity and transaction fees. ‘The Donald’ distributing his prize pot to Puerto Rico As a general counter argument to creating switching costs, a platform could attempt to reduce these frictions by to own a participatory work token. This ‘free to work’ system would require the service providers to stake a SoV coin such as ETH or BTC. It could reduce the overall cost of service provision by removing the upfront capital cost that a new provider would have to amortise. removing the requirement However, although the service could be cheaper, without the community-coordinating tool of a work token the ability to provide a consistent service of similar quality/liquidity as before is unclear. If the service providers don’t have an incentive or coordination device to encourage them to return to the same platform (what the work token ensures), then their skills/liquidity could be fragmented across multiple competing platforms throughout the market, resulting in a poorer UX for all. iii) Governance The introduction of an on-chain governance function, to a network with a high portion of active participants, may help to fatten a protocol/dApp. I’m going to set aside the debate on the relative merits of on-chain governance and instead focus on the mechanical implication that this function could have upon a token’s value capture. A system that introduces on-chain governance via coin voting (ie the more coins you hold, the more votes you get) permits a degree of control over the protocol. If there are valuable dApps which are generating fees, the protocol on which they are operating can be considered a piece of mission-critical infrastructure. By owning some of the voting power, the dApp can attempt to influence the evolution of the critical protocol to ensure that its fee generation is not inhibited. “First comes the bitcoin, then comes the power” However, what number of votes should the dApp acquire? It’s been by that the of a protocol change could be a logical amount to spend in acquiring 51% of a governance token. Apart from the noted logistical problem of doing this, in practice a dApp is not operating alone on top of a protocol. If several dApps coordinated their voting, they could share the cost of this control and so spend less than they would if acting alone. At a higher level, it’s that a community of dApps and institutions that benefit from the protocol may wish to pay for this control and security certainty, in the form of a decentralization ‘tax’ perhaps. suggested Phil Bonello opportunity cost suggested This outcome may not be ideal, as it would mean returning to the conventional (off-chain) informal system of trust and politics to organize this. However, it gives a sense of the potential value that may lie in the governance function despite the difficulty in explicitly valuing it. This function may be an important source of value for protocols that don’t charge fees and don’t expect to become a SoV. iv) Coordinating the critical size For a rival network charging lower fees, in order to provide a compelling UX it must quickly bootstrap a critical number of users upon launch. The coordination required to bootstrap a network is a problem that crypto tokens are known to help solve. However, this ability may be lessened when it’s one crypto system vs another, especially if some of the earlier suggested defences have been actioned. Due to this difficulty, I believe it’s possible that the risks posed by fork networks can be overstated and that requiring a critical size can help reinforce protocol fattening. Caveat Emptor In order for these token mechanisms such as work or governance to release their powers, they must be owned by the network participants- not just by the investors. Although this is a common battle cry from the crypto community, by conducting the above thinking on how these systems can generate and sustain value, it helped me to better understand the importance of this. Network participants need to own the majority of the tokens, not just to ‘show usage’ to the investors, but in order for the token incentives to operate. If successful, these incentives should reduce the threat of forks and hopefully allow for larger, more holistic networks to grow. This happens when token investors HODL without active users? This creates an interesting predicament for the token acquiring investor. One solution may be to develop a new type of investment strategy, with participation in addition to capital being contributed to these networks. Ben Sparango is calling these ‘ . Another solution, that I described earlier, could be for an investor to lease their portfolio tokens to active network participants. Both solutions attempt to enhance the incentives to participate and invest. delegated work entities’ For fee-generating networks that don’t rely upon the token for coordinating the interactions, then the amount held by passive investors is less relevant. For example, see dividend distribution tokens like (KCS). KuCoin Shares Conclusions from my 3-part value capture series Over these three posts I’ve tried to outline some of my thinking around the potential for value capture to emerge within the crypto token universe. This was initially motivated by Pfeffer’s fantastic of the non-bitcoin space and the seemingly dire potential for tokens to charge fees or accumulate value. critique I started by looking at the drawbacks of the popular native payment token. Then I looked at the broader token universe, highlighting which token mechanisms could be helpful for capturing value. Finally, and discussed here, I did some thinking around how fees and value could be sustainably generated by these token systems. In a way, I find it somewhat reassuring that in a new and technically complex asset class an important driver of token value may be the familiar strategic challenges surrounding consumer behavior. Taking measures to entice users onto a platform, create a sense of loyalty, and increase the cost of their departure, . Although the distinction between a protocol or dApp can blur, if a service sits close to the end-user, it’s possible that these value-capturing measures will be easier to implement. a service can command valuable fees despite the threat of competition Perhaps less familiar, are some of the possible solutions that I think may be helpful to incentivise participants to remain on a platform such as . And other solutions to overcome the double-edged sword that passive token investment can create, like token financing and FOMO-style games token leasing and investors participating in the network. I hope that this 3-part series has shed some light on the relevant drivers of crypto value capture. I found that through writing these posts, I was often forced to rethink ideas or concepts that I had previously assumed to be true. There are still many (many!) different ways to approach the topics that I’ve focused on, so please get in touch with your thoughts and feedback. Many thanks to , & for your inital feedback on this post! Nathaniel Whittemore Joanna Hubbard Arek // Follow me on Twitter @CryptoBonsai for more debate :)

KuCoin Token

Chain

fees

Fomo

Uber

Capturing Crypto Value (Part 3/3): Fattening the protocols & dApps

Disclaimer: this article represents the author’s opinions only and should not be considered investment advice. Stablecoins are the talk of the town this year. Let’s dig in and then tear most of em’ apart. While a formal definition remains elusive, most people will likely define a stablecoin as a crypto which remains stable by virtue of being pegged to the value of another currency or asset. This can be anything — another crypto, fiat, gold, etc. The original stablecoin is of course, Tether (USDT). Tether is pegged to the price of USD and users of Tether trust to maintain a fiat reserve equivalent to the amount of tether in circulation. Tether Holdings Limited Put quite simply, USDT is kinda like a digital arcade token that can be exchanged for (not fiat mind you). The original motivation behind the concept was to provide a ‘safe’ alternative to volatile crypto assets for traders without requiring them to pay expensive fees to retreat into true fiat. Bitcoin The vast majority of stablecoins are iterations of Tether. Some promise better transparency than Tether (TrueCoin for example). Other, more ambitious projects aim to peg their coins without a fiat reserve as collateral. Maker proposes to peg to the price of USD using smart contracts, while Basis is attempting a peg based on indexes of asset classes. There are many articles on Medium and elsewhere that discuss the various types of stablecoins out there, so I won’t go into any more detail. At the end of the day, these stablecoins nearly all fail when it comes to at least one of the two following parameters: economic autonomy (functionally meaning: can the coin function in a vacuum?) and decentralization. This might not at first glance seem like a problem. After all, if the purpose of a stablecoin is to provide a retreat for traders, autonomy doesn’t much matter. Even decentralization might not be too critical if we can trust the governing body. However, many stablecoins have bigger ambitions. Bitcoins price has historically, been volatile in the short term and increasing in the long term. This has made Bitcoin a great long term , but a questionable short term . Why spend Bitcoin when you can hodl it? store of value medium of exchange This has naturally led many stablecoin projects out there to view themselves as having the potential to be better than Bitcoin because they are, well, stable. Some even view themselves as being a strong alternative to fiat currency. mediums of exchange This is an admirable goal. But in this light, economic autonomy and decentralization are big frickin’ deals. Also, questions about scale, security and how these stablecoins address credit, money creation and other features seen in the fiat economy should be considered carefully. To illustrate what a stablecoin must be to truly live up to that kind of hype, let’s pretend we are experiencing an apocalypse. We’re talking killer robots, people. For arguments’ sake, lets assume that during the apocalypse, the internet is conveniently still up and power has not gone out. But governments are no longer functioning and fiat currency the world over is dead. Humanity has retreated to the only safe places they know: precious metals and Bitcoin. Also, it’s dark and toxic and stuff. Like in Blade Runner. Or Mad Max. Bitcoin has become the de facto medium of exchange because it is far easier to transact with than gold. Due to the amazing advances provided by lightning network, Bitcoin is able to scale to the needs of the entire globe. However, not all is perfect in this Bitcoin dominated world (beyond the toxicity). Bitcoin’s fixed supply causes it to behave a lot like gold standard currencies once did in pre-apocalyptic countries. This means that it experiences devastating short term economic shocks alongside long term price increases. It remains the roller coaster we know and love. If the last paragraphs made you angry, take a deep breath and read article I wrote which compares Bitcoin to gold standard USD. If you still think I am completely wrong afterwards, I welcome you to abuse me in the comment section. this In this world of scarcity and economic instability a single heroic figure makes himself known. A programmer, economist and researcher, he is hell bent on delivering a stablecoin capable of providing for the worlds needs. He determines that the currency must be: : must provide the 3 key characteristics of money: and Read for more. Money store of value, medium of exchange unit of account. this : capable of functioning in a world without fiat or most other asset classes. Pegging prices to gold won’t work because that will be just as volatile as Bitcoin. Autonomous Bye bye USDT, TrueCoin, Maker, Basis, etc. : necessary for a currency to be widespread for a large population. Divisible : after the fall of governments no one is going to make the mistake of creating a centralized currency again. This also means no solutions that rely on a council of decision makers. Decentralized Bye bye USDT (again), TrueCoin (again), Havven, InitiativeQ, etc. : no duh**.** Secure : Must be able to process enough transactions to serve the entire population. Scalable Rewind. It’s the summer of 2018. The sun is still shining. Toxicity levels are normal(ish). Trump is in the Whitehouse. Bitcoin is still facing an uphill adoption battle. The crypto market is very much in a funk (sigh). And yet, perhaps an apocalypse-ready stablecoin is more than just a thought exercise. The heartbreaking situation in Venezuela is the perfect example for how things can go wrong very suddenly. The US Federal Reserve is fairly successful and USD is therefore a stable currency. But who is to say what the future may bring? Should a stablecoin be able to live up to apocalyptic requirements, it could truly be a competitor for fiat. This would finally help us get rid of those pesky banks once and for all and provide safety from economic crises caused by bad regulation. Let’s redefine what a stablecoin is according to apocalyptic requirements. “A stablecoin is an autonomous, secure, decentralized and scalable cryptocurrency which minimizes economic shocks and successfully provides the three key features of money: , and .” store of value medium of exchange unit of account A bit of a mouthful but I think we covered all our bases. To achieve this, a stablecoin should probably try to learn from fiat currencies. Or at least from the things that fiat does well. Fiat is better at dealing with economic shocks than gold standard currency (and Bitcoin) in large part because it has a dynamic money supply. To use USD as a (grossly simplified) example for how this works: the fed loans newly created dollars to commercial banks which then loan out that money at a 10:1 ratio according to the fractional reserve system. In essence, every dollar created by the fed is multiplied at a factor of 10 in the economy. This enables the fed to enact economic policies aimed at minimizing shocks and promoting a healthy economy. Of course the ugly side of this is that we are reliant on a central bank to make good decisions. When they mess up, hyperinflation can occur as well as other problems. Also, the system is rigged against the little guys. Commercial banks get access to juicy federal loans and to the big business of the 10:1 ratio we talked about earlier (they of course loan out that newly created money at higher interest rates to the population). Read more about money supply. here Disclaimer #2: I am not involved with the Rya project in any way, but I do intend on participating in their airdrop. The only crypto that I know of that might have a chance of functioning in an apocalypse today is a. Ry The main focus of this project is to create a cryptocurrency with a dynamic money supply rate that adjusts based on the price of money in the system, minimizes economic shocks and is an ideal . medium of exchange In fiat, a central bank has a on money supply, meaning that they dictate the price of money by creating the amount that suits regulator strategy. Rya proposes a system in which the price of money is the result of free market economics, or simple supply and demand. The gist of it to cut out the fed and commercial banks from the money creation process and to make a fairer type of money. monopoly In order to achieve this, Rya has a two token architecture. Rya tokens function as exchangeable money and Trust tokens serve as a credit score of sorts. Trust tokens cannot be transferred from account to account. An account can increase its Trust token balance by engaging in peer-to-peer loans (by lending money and by paying back loans with interest). Accounts with higher Trust token balances are able to create more money when they mine Rya, so every miner is like a mini central bank. The Trust balance also functions as a nifty credit score for an account as it reflects economic behavior. When the price of money and interest rates are low, more people are likely to take out P2P loans and successfully pay them back with interest. This will result in many accounts with high Trust balances, causing an increase in the money creation rate in the short term. “every miner is like a mini central bank” On the other hand, when the economy slows, the price of money will increase and less people will engage in loans, causing Trust to be lower across the system. This will result in a slower rate of money creation in the short term. This concept is elegant and seems to promise money supply rate changes as offered by fiat, but in a completely decentralized way while being a better than fiat. Also, P2P loans replace the need for institutional loans seen today. Economically speaking, Rya could exist in a vacuum and be perfectly fine. store of value From a technical perspective, Rya is based on a clone of Nxt source code. When a miner generates new money, they also secure the . This process is coined as Proof of Trust, a twist on PoS that uses the Trust token as the stake parameter instead of the actual token stake as seen in traditional PoS. blockchain The core blockchain tech is very very similar to Nxt, which is probably a good thing as Nxt is a mature project. Rya is currently in testnet, so I have little doubt that changes will still be made down the line. I know the project is looking for more devs. If you have more questions I recommend reading the whitepaper at the Ryo website — it addresses economic aspects of the project in more detail and also drills down into technical/cryptographic questions. The important thing to note, is that Rya seems to have the potential to provide a secure, divisible and scalable solution that fits the apocalypse-ready template very well. Lastly, Rya is not an ICO. The anonymous team has been developing the product for about a year and are close to launching mainnet, which is refreshing after so many projects that have nothing beyond a whitepaper. They plan on distributing the first chunk of coins via airdrop to Nxt owners, however final details are not yet published at time of writing. Philosophically speaking, they believe that Bitcoin is the ideal and view Rya as a in synergy with Bitcoin. They intend on open sourcing the code alongside the launch. store of value medium of exchange To summarize, I believe that there is value in an apocalypse-ready crypto. Venezuela has shown us that. It seems to me that projects aiming to do so will only become more important as crypto gains wide stream attention and begins to truly challenge fiat. I think that Rya has the potential to deliver, and am keeping my eyes open for more projects that fit this mold.

Alongside

Rewind

A stablecoin for the apocalypse

Ahhh, date and time. It’s such a fun thing to deal with during development. In situations when we need more than what the Date object can offer us in terms of convenience, we turn to date wrapping libraries such as . Moment.js Moment.js is a JavaScript library designed to work on both the client side (in browser), and the server side (in Node.js). It has a lot of functions to parse, display and manipulate dates. This makes dealing with dates less painful. I would agree it makes things less painful, but let’s take a look at what happens in the source code. Let’s say hypothetically that I have to work out how many days there are between my birthday and the last day of the tax year in the UK. Let’s use Moment.js to do this. Consider the following code: Moment example There are 166 days between the last day of tax year and my birthday Okay great, we now have the count in days. Simple stuff, and a very convenient method from moment. So what is happening during the execution for this result to be obtained? I am profiling a production build of my application and it’s being viewed in incognito mode. Please note: Photo by on Fabian Grohs Unsplash Okay well first off, my main bundle is a total of 66KB. Moment is 51KB of that, 77% of my bundle. The library is pulling in around 4500 lines of code. Using the JavaScript Profiler in the Chrome Developer Tools, I could see 23 functions were invoked in the moment source code for the 3 functions I used in my example. If I created a single Moment object for the last day of the tax year, a total of 18 functions are called. 5ms was recorded by aggregating all of the function calls on a slow 3G connection. The internal functions of moment that took the longest were: createFromConfig (0.9ms) createLocal (0.9ms) createLocalOrUTC (0.8ms) Now I know what you are thinking, this is so minimal. The dreaded micro-optimisation. Of course it depends on the circumstances of the problem and how complex your date logic is, but it’s always worth reviewing if there is the need to import a package to do such a thing. In my case I would say it isn’t needed. So I decided to turn to a library called just to see how it compares. date-fns The date-fns library is different as it’s modular, meaning you only need to import the functionality you need. This will of course reduce the final bundle size. It doesn’t try and create its own object like Moment does, and it still makes use of the native Date object provided by JavaScript. Date FNS example Total time spent using this differenceInDays function from date-fns was 0.1ms. It required 3 other internal modules to accomplish the task, but is much lighter in comparison to moment. The file size of my bundle was 18KB. This means the functionality of date-fns was 3KB to achieve the same thing! Photo by on Ian Stauffer Unsplash Take a moment.js to observe the sizes of your packages and improve your overall page performance. Sorry for the pun. Thanks for reading, and please drop a few claps if you found it useful. Expecting happy birthdays mid-September time in the comments :) Man… I’m getting old. cringe-able Shaun Michael StoneTwitter: @shaunmstone

Wait a Moment.js!

reach out on twitter @ethanruhe TL;DR: Collecting fees on trading volume is Coinbase’s source of revenue. Coinbase’s YoY trading volume grew >50x in 2017, to $80B. did $1B in revenue last year, implying a 1.25% effective mean fee rate. Reports suggest Coinbase They starkly segment fees: ~2.5% on the app, ~0.25% on “Pro” trading platform. at 0 fee is emerging. Retail competition Other pro trading platforms generally than Coinbase. charge significantly less Large financial exchange comparison: for $0.5B in fees ( , data), an effective fee rate of 0.0035% (_i.e.,_Coinbase’s rate is ~350x higher). NASDAQ exchanged $14.8T in H1 2018 Q1 Q2 Volume is positively correlated with prices; both have been falling since the start of 2018. In the first half of 2018, Coinbase did $77B in volume. But, they only did $4.7B in June, 18.6% of January’s volume. Conclusions: In 2017, volume was split about 50/50 between retail (app +website) trades and pro platform trades. If Coinbase maintained the same retail/pro split, they booked about another $1B in revenue in the first half of 2018. Current fee rates are unsustainable in the long run, and competition will drive them down. Exchanges simply don’t support big-tech-scale businesses; put another way, Apple makes more setting the default search engine in Safari than NASDAQ does with its entire exchange business. Success of Coinbase’s ambition will depend on building an ecosystem of complimentary products that achieve scale, not exchange fees ( ; ). CEO Brian Armstrong lays out their broad 4 step strategy here their product news is here Coinbase seems to be in a unique position: 10x growth may depend on their current core business shrinking ( , exchange with less friction and lower fees). i.e. The gravity of Coinbase’s talent pool + on long-term problems + a mountain of cash all mean that Coinbase is well positioned, but its breakout path still isn’t clear (like Facebook before mobile, Instagram, and WhatsApp). sober focus . Code for pulling Coinbase data and replicating my analysis is available here If this is interesting to you, let me know, and I can do a deeper dive on what the data tell us. A longer version follows below… Why is it interesting? The strong case for cryptocurrencies as the biggest platform/protocol shift in a long time. Coinbase could become a juggernaut, but their current revenue model isn’t what will get them there. The company is unusually long-term focused (reflected in thoughts like and the fact that they give employees a ). this 7 year exercise window on options Coinbase’s revenue growth is rumored to be almost historically fast. ( ) The widely-reported rumor is that they did $1 billion in revenue last year. Allegedly the bulk of that came during the price surge in the waning months of 2017. With this data point in mind, we can estimate what proportion of Coinbase’s revenue comes from casual consumer speculators and how much is from their more sophisticated professional trading platform. We can then estimate how 2018 revenue is shaping up for the company. This analysis is possible because data are unusually accessible for a private company: running an exchange, Coinbase’s near-exclusive revenue source to-date, requires publishing price and volume data. These data are key drivers of revenue and therefore give a unique view into the growth of the private company. Coinbase has never aimed to be an exchange. Brian Armstrong, the CEO, lays out their broad strategy . They’ve been , especially in the past year. only here aggressively launching and buying products In order to be a ≥$100 billion company, Coinbase will need both successful new product bets and for institutional cryptocurrency adoption to happen sooner rather than later. The closest analogy to what Coinbase has to execute over the next few years is Facebook circa 2012–2014: a huge platform shift at a favorable time (consumer adoption of mobile in Facebook’s case, institutional adoption of cryptocurrencies in Coinbase’s), one or more successful bet-the-company moves to capitalize on the opportunity, and possibly homerun product acquisitions (_e.g.,_Instagram and WhatsApp). How does Coinbase make money? On an exchange like Coinbase’s, trading involves matching buyers and sellers. Coinbase takes a cut of each transaction. Limit orders involve a trader offering to buy or sell as much as a set amount of an asset at a specific unit price ( , “I’ll buy as many as 10 at a price of $5,000 per bitcoin,” or “I’ll sell as many as 2.5 bitcoins at a price of $5,010 per bitcoin”). Limit orders underlie the market and “give liquidity.” Those who place limit orders are called price makers since they literally name their price. To date, limit orders have been free on Coinbase in order to encourage liquidity growth. e.g. bitcoins A market order involves specifying a buy or sell against outstanding limit orders. In a market order, a trader specifies an amount of one asset to be exchanged, on the best possible terms, with outstanding limit orders on the other side. For example, “sell 100 bitcoins” results in selling 100 bitcoins to limit buy orders in descending order of price. Depending on the size of the market order, it may fulfil as little as part of one limit order or as much as multiple limit orders. The fees Coinbase charges for market orders are starkly segmented across what they refer to as retail traders (those who use the mobile app and coinbase.com) and professional traders (those who use pro.coinbase.com, which was formerly called GDAX). Anyone can perform retail trades with the app or on Coinbase.com. To date, the pro platform has also been easy to join. It just has a slightly more complicated interface. (Note: ) in May 2018 Coinbase launched an institutional trading platform called Prime, but this analysis will not attempt to separate that from Pro as there is no benchmark for how to do that cleanly. Coinbase Fees Retail trades (mobile app and coinbase.com) are subject to a fairly complex fee schedule that can vary with Coinbase’s discretion depending on market conditions when executed ( ). For use in the model, we’ll simplify it to the following: documentation is here Conversion fee of 1.49%: For every $100 in retail cryptocurrency transaction, Coinbase takes $1.49 for converting between fiat and cryptocurrency. Exchange fee of 0–2%: Coinbase says “Your exchange rate for buying or selling digital currency through our Conversion Service is calculated as the market rate of the digital currency on Coinbase Markets platform, plus a spread between 0 to 200 basis points determined by the size of your transaction, market volatility and length of time using Coinbase.” So, in addition to the conversion fee, they take a 0–2% exchange spread relative to their underlying exchange price. This is analogous to the “taker fee” we’ll see below. At the time of this writing, I initiated a $3,865 bitcoin purchase when the price was $7,730 and was quoted a purchase volume that implied a 50 basis point, or 0.50%, exchange fee. I have a relatively long track record on the platform and short-term volatility was relatively low, so I’d guess this is on the lower end of the spectrum. For the sake of simplicity, we’ll assume a 100 basis point, or 1.00% mean exchange fee for retail orders. Increased conversion fees if using a credit or debit card. In general, this is a 3.99% conversion fee instead of 1.49%. For the sake of simplicity, we’ll assume that additional 2.5% just covers the cost of payment processing and is therefore revenue neutral. So, under the simplification above, we’ll assume that for every Z dollars in retail trading volume, Coinbase generates: Z * (1.49% + 1.00%) = 0.0249*Z dollars in revenue. Coinbase Pro (previously GDAX) Fees Coinbase Pro charges 30 basis points, or 0.30% in exchange fees for price takers. Accounts that trade $10–100m or >$100m over a 30-day period receive a 10 or 20 basis point refund, respectively. ( .) Documentation is here The market data we’ll use doesn’t attribute volume by-trader, so we have to assume the relative proportion of Pro traders’ volumes in order to calculate a mean exchange fee. Because it is very easy to create a Pro account and even the discounted taker rates are prohibitively high for high-volume traders, it seems reasonable to assume that most volume is subject to the full 30 basis point fee. For the sake of easy comparison, we’ll assume the mean Pro fee is 24.9 basis points, or a tenth as much as the Coinbase retail fee. As the market matures, both the absolute rates and the degree of retail vs. pro price discrimination seems unsustainable. As is, any savvy trader should never use retail Coinbase. To-date, the segmentation has been quite successful, however. Many consumers either don’t realize the difference, or put a big premium on the convenience of the mobile app (as evident by the fact that it ). reached #1 in the app store for a while in late 2017 Expect serious downward pressure on Coinbase’s fee rates in the future. Competitors generally charge for “Pro” style exchanges. , expect pressure towards zero fees for investor-facing exchanges as firms seek to capture market liquidity. significantly lower fees Ceteris paribus Of course, everything else is not equal and being a leader in governmental and institutional regulatory compliance has been and may remain a competitive advantage for Coinbase. Coinbase will face similar retail fee headwinds. already offers “free” exchanges though custody isn’t completely transparent. are sure to pop up as well. Robinhood Other entrants Exchange Numbers Proportion of Revenue from Retail and Pro We’re going to make another simplifying assumption that the relative proportion of retail and pro traders has stayed constant because we don’t have better data to suggest otherwise. Given the assumptions: that proportions of retail and pro are stable a 2.49% revenue rate on retail volume a 0.249% revenue rate on pro volume; and, $1 billion in 2017 revenue we can calibrate the proportion of volume from retail and pro traders as follows: $1 billion = (X * V * 0.0249) + ((1-X) * V * 0.00249) where X is the proportion of volume from retail traders, (1-X) is the proportion of volume from pro traders, and V is the total volume traded in 2017. Based on Coinbase’s API data, they did about $80 billion in trading volume in 2017. Plugging that into the function as V, we find that X is approximately 50.2%. So, Coinbase appears to have done about half of their volume at retail and half at pro rates. $1 billion revenue on $80 billion in volume. These Numbers in Perspective NASDAQ, the second-largest stock exchange in the world, represents about $12 trillion in market cap. In the first half of 2018, . They claim to do about 1 in 10 security share sales in the world. Their exchange generated $517 million in trading fees based on this volume ( and financial results), which includes not just cash equities but more expensive products like derivatives. That’s a mean fee rate of about 0.0035%, orders of magnitude less than Coinbase. And, NASDAQ even has the advantage of near-exclusivity for many of its traded assets. NASDAQ traded $14.777 trillion Q1 Q2 An exchange is a good business, but margins go down as liquidity goes up. In the long run, Coinbase want a world in which their current fee rates are sustainable. That is indicative of an inefficient and speculative market, not the frictionless economic protocol they hope cryptocurrencies become. doesn’t An exchange simply doesn’t support big tech scale businesses. Put another way, Apple makes more setting the default search engine in Safari than NASDAQ does with its entire exchange business. Coinbase’s Volume Trend This is the cumulative volume traded on Coinbase (note the first years of the company’s existence aren’t shown because you wouldn’t be able to see it anyway): Annual totals are as follows: Finally, monthly totals are as follows: Appendix: Potential Material Fallacies These are some of the reasons my analysis above could be wrong: Fundamental misunderstanding of how trade volume is counted. It’s possible Coinbase obfuscates or does not include trading in the data available on their API. For example, they may subtract retail trades. I don’t have any insider information, so I can’t rule this out. all Historical fee rates may have differed significantly. This analysis applies current fee rates to historical data. Coinbase has certainly played with these rates, but I wasn’t able to find a detailed historical fee schedule, so I’m applying the current rates. To the degree these differed, this analysis will be inaccurate. We don’t know the % of pro trading that qualified for high-volume discounts. This would affect the retail vs. pro volume proportions calculation. $1 billion in 2017 revenue may not be accurate. The retail vs. pro mix has shifted in 2018. That would affect my estimate of Coinbase’s H1 2018 revenue.

Apple

Facebook

Instagram

Robinhood

Coinbase's exchange made $1B in 2017. It might have to make less to succeed.

ORANGE

Yellow

Fundamental Python Data Science Libraries: A Cheatsheet (Part 4/4)

Customer service from our robot overlords Update I had a friend who is a software engineer at google file an internal ticket to address the situation. He said it made it past the first layer of customer service. So far so good right? Well, he told me today (2018/11/09) that the ticket has been killed internally. If even an internal employee can’t fix this situation, I can only conclude that this really is permanent. It’s a real shame because I planned on using AdWords to buy up advertising on common error messages for which I thought would be a great way to fund open source developers and solve problems for corporate developers, but this channel is closed. Open Source Answers Recap I recently quit my hedge fund job to pursue entrepreneurship. My hypothesis is that in the long term entrepreneurship outperforms high paying jobs, and that most risk of failure can be minimized when starting a business by choosing conservative (easy) problems, and applying rational thinking as well as creativity to business problems. I started chronicling my success and failures , this is article 3. here My first project is a . I’m going to go into how I tried to advertise and how I accidentally (and permanently) got myself banned from Google Adwords. group email business GroupMail Marketing GroupMail So I had just finished building GroupMail, and I wanted to start marketing it. I was lost, I posted links on hackernews, facebook, got some clicks but not much. I started reading about , and . SEO is split into on-page SEO (setting up html properly on your website to make google happy), and off-page SEO (getting links from other well established domains so that you rank higher in google searches). There used to be some shady tricks for SEO that . Conventional wisdom is that creating good content, which people naturally link to, is the best way to get good SEO. However SEO takes time, It takes time to produce content, it takes time for people to link to your site and for search engines to trust you. So SEO isn’t really good for rapidly testing whether new ideas will work. SEO long tail keywords no longer work The idea behind long tail keywords is that as a startup, you’ll never rank for really popular phrases like “group email”, however more specific phrases that are less competitive like “free group email from gmail contacts” you can probably win. Fewer people will search for long tail keywords, but there are more of them, so by casting a wide net you can get traffic. This also carries over into adwords, the less popular phrases are much cheaper per click. Advertising on Adwords I don’t have time to wait for SEO, so I decided to get started¹ with adwords. I created this advertisement. It was approved in a few hours and began running. If you’ve done online marketing before, the no no with this ad is obvious to you. It didn’t even register to me. I had a few questions about when my ad was going to start running, so I contacted customer support. Google assigned me an account representative! A real human that would help me get started and help me optimize my campaign. He called me the and seemed very helpful, but I decided I wanted to play around a little bit before asking him for help. Thins were looking pretty good. That day, I got some clicks and I went to bed. Trouble Starts When I woke up in the morning, I got 2 emails about my account being suspended. One for “suspicious payments”, and one for circumventing systems. When I logged into adwords, this is what I saw. woah This is strange — my ad was approved, and I was using a credit card I had used forever, so I thought it had to be a mistake. If you click on CONTACT US you are taken to a page where you can appeal to google. I had no idea what could be wrong with my ad — it was approved before I began running it, and I knew my credit card was good, so I figured there was a glitch in Google’s automated system, and that when a human reviewed it, I’d be all set. I got the response: Hello, We’re sorry to hear that you’re experiencing difficulties with your Google Ads account. What happened? We’d like to let you know that your account is currently under review and we’re prioritizing this over other issues you may have written to us about. Our Specialists already know about this issue, and an investigation is underway. We periodically review accounts for security purposes and to verify billing information. What should you expect? We’ll send you an update after the investigation is complete, typically within three business days. Once we provide you with an update, please feel free to reply to this email if you have any questions regarding your Google Ads account or contact us through https://support.google.com/google-ads/contact Sincerely, The Google Ads Team Ok, that sounds pretty good. Wait! I have an account representative. Surely talking to a human will be able to sort this out! Calling my representative went straight to voicemail every time. Once your suspended, Google does not permit anyone to talk to you. I called 10 times!. I left messages explaining the situation — no response. I tried calling the support line, if I entered my account number, I would get disconnected. If I act like a brand new customer, I get a customer service agent, but once I brought up the suspension, customer support says they cannot discuss it, and cannot help you. Oh well, hopefully our new robot overlords will fix things for me. Later that day, I got back a response from Google. Dear advertiser, Thanks for your continued patience during this process. We’ve confirmed that your account is in violation of our Google Ads policies. Since this decision is final, the account will not be reinstated. Please avoid creating additional Google Ads accounts, as they will be subjected to the same suspension. Our support team will not be able to give you any more specifics on the suspension. We appreciate your understanding and cooperation. Sincerely,Elizabeth You may receive an email survey regarding your interaction with Google Ads support. We value your feedback. I knew nothing about online marketing. Ironically back when I was looking into affiliate marketing, and looking at Amazon, I was very careful. I always thought of Amazon as a cutthroat master who would discard affiliates at the first opportunity. I read many articles on how to avoid getting . kicked out of Amazon’s affiliate program I thought Amazon was the scary one Consequences I had always thought of google as a vaguely positive, don’t be evil, omniscient mega corp. It turns out google is just as hard to deal with. . I am evidently a moron. I asked reddit what I did wrong Dude you can’t put an email in the ad hahaha and I am surprised that they even approved to start with Ok — I can understand that, I admit that I made a mistake and committed a no no, and I’ll never do it again. However now I’m banned from google advertising forever. It’s not just that account. It’s not just GroupMail. A week later, I decided to do a bit of experimentation in real estate, so I bought the domain watsonvillehousesforsale.com, threw up a landing page and bought some ads. Within 3 days, my account was suspended due to “suspicious payments”. I used a different credit card, but that credit card was also in my name. I’m cutoff, and I can’t even talk to a person to fix the situation. There’s actually out there who is an artist. Sucks to be you! you’re probably banned too! I hope you never need to advertise on google! another Hugo Shi How I Feel This made me realize just how much I don’t like one entity having so much power on the internet. You may be thinking, well if you weren’t an idiot in creating your ads, you wouldn’t have this problem. That’s true, and if you guarantee you’re never an idiot in the eyes of google, then I guess it’s fine for you. But consider the following Maybe you hired a cheap SEO firm, and they employed some black hat strategies, and now your page is from google search banned Maybe you hired a bad marketing firm and they did some bad things with AdWords What if your gmail account is compromised, someone uses it to send spam, and your gmail account is shut down. Bing only has 10% market share, and it’s the top contender for #2. So in terms of getting paid search traffic — there really are no alternatives. I’d encourage everyone to think about what would happen if you lost your GMail account Conclusions In summary, I’m super uncomfortable with Google controlling for all intents and purposes what the internet is. But it’s kind of too late now. Maybe the new . I do think that you should have a serious look at your corporate dependencies, and figure out what you can do to minimize your risk. For me, I’ll probably just have to figure out how to hire a company to advertise for me on AdWords, or focus on businesses where other advertising channels are more worthwhile. Or maybe one of these companies can get me back on to adwords decentralized internet will save us I also think it’s kind of crazy that you can buy AdWords advertising for a company to get people back onto Google AdWords², but what I did, got me banned for life. Thanks for reading!. My past projects are , and getting . My current project provides . In the next article of this series, .. In the next article of this series, I write about how I with the help of Mechanical Turk. about simple easy email groups help for open source projects cloud hosted Jupyter notebooks I write about building GroupMail optimized the GroupMail landing page Footnotes Ok so I actually created my first adwords account in 2013. I spent 5$ on it before I got distracted and started working on something else. That account is still there, but I bet if I started using it, it would get shut down. If anyone can get me back on, my contact is hugo.r.shi@gmail.com

I’m Permanently Banned from Google AdWords

How research is being exploited by corporations and academic institutions in a society of shifting incentives It all starts with the question we’ve been told we have to decide in order to make the right career choice, when the time comes to start being financially independent: should I choose a career in or ? The problem here is that the question itself is rigged, being based on the assumptions that 1) a person can only be good at one thing and one thing only, and 2) that in order to make a living, one has to put all their efforts toward the advancement of a larger organization. What were not usually told is that there are to leverage ones skills to make a living in through independent, self-determined efforts, and more so now than ever before. industry academia several ways The result of such a limited way of thinking is that highly skilled or educated people choose one of the two options, thereby concentrating most of the innovations of society within large organizations: corporations or academic institutions (many of which are privately owned, blurring the distinction between the two). And along with those innovations, most of the research funding is concentrated there as well. Such phenomenon is should not be taken for granted as the natural order of things; they are merely the result of various decisions and policy-making which are especially favored by the free market, unregulated economy of Western nations over the past two centuries, leading to the exploitation of science primarily as a means for monetary gains, rather than to improve people’s standard of living or otherwise simply . understand how the natural world works But the trend is starting to change. With technology becoming miniaturized, cheaper, and more accessible to the average person, there could soon be a major shift in the mindset of society that innovation has to stay in the confines of large, profit-driven organizations. Some of the signs are becoming apparent, with the rise of research crowdfunding sites, the open-access movement, and the trend of more and more young people choosing the startup route, bringing technological innovation directly to consumers, rather than them having to filter through the drawn-out bureaucratic process of technology transfer and R&D in universities and corporations, leaving a large and unnecessary gap between the time of development and the time public actually gets to see the end product. With such factors on the rise, there are are clear reasons to define why the trend toward decentralized, independent research will be a good thing for our society and researchers themselves. The economics of research are corrupt and skewed Because universities are funded in proportion to its research output, academic departments and laboratories are willing to spend exorbitant fees on journal subscriptions just for the opportunity to publish. This leads to, first of all, the incentives for doing research becoming skewed, from one of scientific to monetary motivations, and also serves to distance the ownership of the research away from the researcher and to the university as a whole. Another problem with this economic system is that journal publishers are disproportionately profiting off of the fees that institutions have to pay in order to get published. The economics are wildly favorable for the publishers, who essentially take the free labor of academics and researchers (the authors and peer-reviewers of scholarly work are unpaid for their contributions), and sell it back to the universities that employ these authors, often at a very high cost. — Quartz It follows that a university has to pay to get published, and needs to get publish to get paid, resulting in a situation wherein the journal publishers are the ones who practically . dictate the success of a lab or entire institution As the cycle of publishing and funding continues, costs rise, and revenue for the publishers . far exceed their cost in editing and maintaining the journals The current research ecosystem has been built up over the years into a vicious cycle wherein: funding leads to innovation, innovation leads to reputation, reputation leads to talent, talent leads to more funding and innovation, leaving any opportunity for independent research without collaboration from a large organization practically impossible. This leads to the integrity of the journal business becoming by the incentive for financial gains it poses, with some academics having gone so far as to call for a of one of the foremost academic publishers, Elsevier, for their high subscription prices and other purported unethical practices. corrupted boycott This misincentivized economic cycle also results in a , in exchange for higher quantity of output, as well as a growing trend of or highly unreliable publishers coming onto the scene, as it’s seen as a lucrative business opportunity. decrease in the quality of research fake But the publishing dilemma is not the only one. There is also the question of who has the intellectual property rights to — and therefore any future profits from — any technology developed in a university lab, and that answer is usually — yup, you guessed it — big business. Corporations gain control over the technologies developed through research While research laboratories are preoccupied with publishing, they may overlook the of their research. Although for many researchers, making profit off of an invention is not a primary motivation, that’s no excuse to be informed of the possibilities to gain financially from one’s efforts. When corporations sponsor academic research, it is in their interest to take the rights to any inventions that are developed through said research, which they can then commercialize and make a profit off of (considering that the primary goal of a corporation is — unless some major change has occurred in the fundamental structure of society that I haven’t heard about — to make profit). Thus, the researcher is satisfied merely getting published, while the corporations get the rights to their technology, effectively creating an oligopoly of innovation and eliminating any chance of competition from smaller startups in the same industry. Thus, another cycle is formed in which researchers have little incentive to commercialize their technologies, widening the gap in incentive between the financial beneficiaries and those who put in the effort, and corporations get ever bigger and maintain their grip on new technologies (not to mention the cultural implications of such financially-driven innovation). intellectual property considerations not Research is also big business — at large universities like Stanford , sponsored research often exceeds revenues from tuition. * To sum it up, researchers are , whereas the larger organizations (universities and corporations) are the ones who are benefiting most_._ Such correlation is the basis of a functioning economy, and without it it’s a wonder the research industry has managed to sustain itself so long. not being compensated in a way that is commensurate with their efforts People start pursuing PhD for the wrong reasons, eroding their value and leading to their impending obsolescence Oftentimes, nowadays, many people are going into a with the sole intention having a better looking resume, or to get a marginally higher salary at their future job. But this begs the question: are 4 to 6 years of research either necessary or relevant to performing a narrow corporate job description better than someone with a bachelor’s or master’s degree? The skills one gains in research are incredibly different, and arguably to many positions, as compared to those gained through pure work-experience over the same duration of time. This should not come as a surprise, considering that a PhD was not designed as a qualification tool, or as a level-up from a Bachelor’s or master’s degree, but rather, a period of time in which a curious person can delve into a field of interest to broaden the body of knowledge about a particular field. PhD less relevant The benefits of a PhD a not as significant as one might expect (Image copyright: ) Nature.com From this perspective, it seems to be a misalignment of incentives for a student to go into a PhD solely with employment-related aspirations. Whereas that student will gain a degree in a field she may or may not have more than a vague interest in, the university sponsoring such a degree reaps the benefits of government and corporate funding to sustain research operations, as well as maintains their academic standing through the publication output and reputation boost it implies. As such, universities and corporations are both benefited, but the researcher herself gains nothing but a degree. The disheartening result is that many PhD students are graduating into a oversaturated pool of qualified individuals, most of who end up working for a small handful of large corporations, ultimately feeding back into the cycle of , and further leading to the devaluation of higher education qualifications, including the PhD. oligopolization of innovation The future is not be as bleak as it seems Expanding the scope of research outside of a handful of academic institutions would be an incredible step, one that could allow part-time researchers to contribute their knowledge as well.* The current trend of academic research is not exactly promising. But sooner or later, it will become evident that a degree is not the only means to career success, and more will come about. With crowdfunding platforms and open access publications, research could be re-envisioned as an independent practice, unreliant on the corporate or university structure to make it a sustainable occupation. As researchers become aware of the potential commercial applications of their innovations, they may more readily pursue startups as a means of not only sustaining their research operations, but also bringing their technologies out of the lab, and into the real world, making a direct, positive impact on people’s lives. Inquisitive minds will be empowered to freely experiment without the exhausting pressures of academic life nor the restrictive obligations of corporate employment. Corporations will be pressured to compete on a fair playing field with independent research and startups, unable to count on their large budgets to buy new innovations to maintain their market share. creative ways of sustaining a living Could this be the lab of the future for independent researchers? ( Photo by on ) Vital Sinkevich Unsplash With such a large scale revision of the existing ecosystem, the integrity of research will naturally be restored, and the incentives for pursuing scientific investigation will be realigned — all based on the motivation of an individual to fulfill their innermost potential… and for the sake of a higher salary or a more competitive resume. not *Quotes from article by Techcrunch. this For more musings on the intersection of technology, education, and entrepreneurship, follow me on ! 🐦 Twitter

Be Informed

Funding

Techcrunch

Click here to follow my continued adventures

Read My Stories

The PhD-ilemma

Probably most of us have been in a situation in which we have sent an email that we did not want to send. We might have not finished formatting, or double checking it before the actual sending process, but we hit the button. Most of the times, it can be unintentionally, and as a result, we may write one or two follow-ups as replies to that email, adding or sending clarifications about certain things that have been mentioned in that email. During this week, I learned a small trick from a friend of mine that can help you in such occasions. Send PS As I was replying to an email, this friend told me to delete the email of the recipient, as this way, you will not be able to send this email. You may not be ready to send that email, but may accidentally press the button. When you have not mentioned a client for your email, your email client will not know where to send it. As a result, you may save up a lot of time and keep yourself away from getting embarrassed for sending an email half finished. Send This may sound as something really simple and that may not be that much of a husle, but I believe it can be very helpful to you.

Keep

A quick tip on how to stop sending emails you are not done writing yet

According to Rock Health, the digital health funding in the first quarter of 2018 reached a whopping $1.6 billion. This is the highest first quarter on record as it surpassed the much-hyped $1.4 billion funding in venture capital during Q1 2016. AI is being applied in 5 different areas of healthcare to improve its overall operations. It is helping in administration, care provision, clinical decision-making, big data analysis, and remote patient monitoring. We consider the top 5 startups which are relying on AI to provide the best healthcare services to their patients: Ucare.AI Founded by Neal Liu, an MIT graduate, the company started receiving funding in 2016. Its AI algorithms and the neural network is used to predict the healthcare patterns in patients. The aim is to prevent hospitalization so that the costs and the time can be saved. The sole purpose of this app is to provide the medical practitioners with an opportunity to focus on only such patients who are at a higher level of risk. Bay Labs, Inc This is a San Francisco based startup which has been able to use AI in cardiovascular imaging. This helps in the diagnosis and treatment of heart disease. The data which is produced through cardiovascular imaging can be hard to understand for some. Nonetheless, AI technology uses such algorithms and videos from the huge library that help the medical personnel in proper interpretation and cure. Talkspace Mental health is a growing issue and needs to be addressed properly before we lose a number of precious lives around us. People like Michael Phelps have also suffered from the worst form of so, consider this epidemic to be as real as possible. Lately, Phelps partnered with Talkspace to increase awareness on mental health. The company was founded by Oren and Roni Frank in 2012 and has already treated around half a million patients. It helps you connect with licensed therapists through their app or website. depression Vitagene Until now, the company has raised $6.6 million and is a renowned name in genetic sequencing. This app will run your DNA model through its AI platform comparing it with other samples in their database. The results help the app in recommending a healthy diet, an exercise routine, and a supplement plan. Not only will you be able to get the DNA health test but it will also present you with the ancestry report. VaGenie Childbirth crazily affects a woman’s body and weakens the pelvic floor. As per the research, 2 out of 3 women suffer from a weak pelvic floor and this causes various problems in them including back pain, incontinence, and prolapse. VaGenie provides a combination device and app to help women in strengthening their pelvic floor through Kegel training. It teaches the women how to perform the exercises correctly while making them fun for them. These startups are disrupting the health industry and using AI to improve the care which is provided to the patients. There are plenty of more startups working in the field and in near future, AI is going to bring more ease to this industry.

Routine

5 Health Startups Using AI to Improve Daily Wellness

Less than a month ago, the U.S. Securities and Exchange Commission (SEC) delayed a decision to approve a Bitcoin-based ETF. In a few days, it will decide on another proposed Bitcoin ETF. This article briefly explains what an ETF is, different types of Bitcoin ETFs, challenges to Bitcoin ETF approval, and which Bitcoin ETFs are likely to be approved first. What is an ETF? A lot of people have been talking lately about the creation of Bitcoin ETFs_._ An ETF (“exchange traded fund”) is a security that tracks some underlying assets. these assets could be equities, bonds, commodities, or cryptocurrency. An ETF takes custody of the underlying assets. A custodian (bank or other financial entity) is responsible for safeguarding the fund’s assets. Then it issues a number of shares that represent ownership of those assets. Types of Bitcoin ETFs There are two basic types of Bitcoin ETFs: ETFs that “physically” hold BTC, for example, (supported by the Chicago Board Options Exchange). VanEck/SolidX ETFs that purchase Bitcoin derivatives. These ETFs are not based on actual BTC holdings but on trading Bitcoin futures, options, swaps, money market instruments, and other investment vehicles. Two examples of Bitcoin derivative ETFs are the and ETF Direxion ETF Proshares The SEC was set to decide whether to approve the VanEck/SolidX Bitcoin ETF in early August but delayed its decision. In a few days, it will decide the fate of Proshares. However, unlike with VanEck/SolidX, the SEC cannot delay this decision. Concerns about Bitcoin ETFs ETFs are traded on exchanges, which means they are traded in a way similar to public equities. The regulatory requirements to having an ETF approved are stringent because being listed on an exchange lets anyone invest in them. When the SEC evaluates a proposed ETF it is concerned with minimizing conflicts of interest, errors in tracking, manipulations, and that investment risks are reasonable and properly disclosed. Custody Risk One big worry about ETFs that hold Bitcoin is custody risk. Custody risk is essentially the risk that an ETF custodian might go bankrupt and no longer support the ETF. This is a particular concern with ETFs that hold physical Bitcoin. Bitcoin is largely traded on unregulated international exchanges that risk being hacked or going out of business. The value of Bitcoin is also highly volatile. Non-transparent Data Sources However, reducing custody risk is still easier than mitigating the risks of errors in tracking Bitcoin’s value. This concern applies to primarily to derivative Bitcoin ETFs. Compared to regulated equities, cryptocurrency markets are less transparent and more costly. Further, small price changes can have a huge impact on the ETF’s value because of leveraged trading. The result is investors need to have a lot of trust in a derivative Bitcoin ETF manager to properly track the performance of Bitcoin. Conclusion This year, there has been progress on custody solutions for cryptocurrency assets. The BTC in VanEck & SolidXs proposal are now insured, which may improve its chances of approval in the future. ETFs based on physical Bitcoin holdings are also the most transparent in terms of value-tracking. However, derivatives-based Bitcoin ETFs carry less custodial risk, which is a point in their favour. In any case, with the SEC having yet to approve any Bitcoin ETFs and cryptomarket volatility persisting, the upcoming decision on the Proshare ETF will be watched closely and will certainly have a impact on Bitcoin and the broader crypto market. Subscribe to my channels Medium and Twitter if you like my articles and want to learn about blockchain and cryptocurrency projects. If you have any questions about this article, please comment in the comment section below. To support my writing: 1PUzB7FtURwikfUg4zgdULyyeQkcK6o9fhThank you!

Winner of Noonies 2020 for the award Hacker Noon Contributor of the Year - INVESTING.

2020 - Hacker Noon Contributor of the Year in INVESTING

Check out Mina's blog

Buy Me a Coffee

What are Bitcoin ETFs and Why are They Controversial?

This summer I had the pleasure of interning at USV this summer to support the . Network Team Through my community management work in Slack as well as in attending USV events, I noticed that one of the most common questions that comes up in the network is what are used by other companies. network tools To embark on this research project, we gathered over 450 software tools used across 64 companies, representing approximately 86% of our active portfolio. We began by tagging each tool based on department, the type of tool, and then explored how company size impacted usage. Below are some observations and trends that we discovered. Types of Tools Reported As you can see below, developer tools ranked #1 among the categories of reported tools for this project, with 164 total tools reported. We also saw large numbers of tools used for communication, business intelligence, and design Most Popular Tools The top reported tools were G Suite, Slack, GitHub, Jira, Sketch, Salesforce, DropBox, AWS, Excel, and InVision. These top tools (in particular G Suite, Slack, and GitHub) were used across multiple departments or tended to skew toward developer tools. Developer Tools in the Network The chart below show the most popular developer tools overall in the network. After Github and Jira, there was a drop off which may be because there are many free/open source options, causing individuals and companies to explore and favor different ones. Size Breakouts Company size creates a stratification among developer, hiring, and communication tools. As an example of how this plays out in engineering, you can see that all of the size brackets use either Github or Jira, but after that, the differentiation develops. For example, smaller companies are less likely to use security software. On the communication side, smaller companies are more likely to use internal tools like Calendly, GoToMeeting, and Rocketchat, whereas external tools including Docusign, Facebook, Medium, Sendgrid, and Twitter persisted across all size brackets. As for hiring tools, LinkedIn, Lever, and Greenhouse are the most popular in the network, but these are primarily only used by companies with more than 30 employees. Tools like Guru, Textio, DiscoverOrg, and Checkr were used with companies larger than 50 employees, while Jobbatical, Entelo, Jazz, and AngelList were used by our smaller companies. Market Dominant Tools by Category While AWS still appears to be the favored cloud platform used among our portfolio, Google Cloud in particular appears to be gaining more traction. As you can see below, ¼ of companies are using either Google Cloud or Microsoft Azure. Other tools with dominant market share include Slack and Salesforce. While 92% of our companies report using Slack, some companies opted to use Hipchat (recently acquired by Slack), Beekeeper, and Rocket Chat. For sales management CRMs, 42% of our companies use Salesforce, however, some used alternatives like Google Streak, Insightly, or Close.io. It does not seem as though size played a role in a company’s choice in any of these circumstances. Closing Thoughts One of the most fun parts about working on this project was learning about new tools used in the network. In fact, 52% of the tools reported were used exclusively by one company in our portfolio, which exposed a few new tools like and (both workflow management tools) as well as (an HR platform built for G Suite). It was also great to see how 78% of our portfolio companies use tools built by our portfolio companies (including and , .) Jell Perdoo Sapling Code Climate Cloudflare among others While the average company in our portfolio is using 7 tools, one of our larger companies reported using 77 different tools, which made me realize that transitioning between tools may be a source of growing pains for startups. As our portfolio continues to expand, we hope that by aggregating and sharing this list internally (we’re even building a Slack Bot for our network to search these tools), we can make it easier to choose the best tool for each use case.

Dropbox

Slack

Top Tools Being Used in the Union Square Ventures Network

The Scalability Problem The scalability problem lies in the fact Ethereum does not scale on the base-layer. This is why even halfway popular decentralized applications (dapps) can cripple the network (e.g. Cryptokitties or past popular ICOs). The scalability problem is a barrier to mainstream adoption. As it stands, Ethereum could not handle even one dapp going mainstream. The Ethereum blockchain size has also . The problem with this is larger blocks require . The computational power needed exceeds the average user’s hardware and bandwidth. As a result, there are fewer and fewer full nodes that update the network. Fewer nodes mean more centralization. Thus, Ethereum’s growing blocksize is resulting in a more centralized network. exceeded 1 TB more powerful machines to verify Scaling Solutions and Why They Won’t Work Three main scaling solutions for the Ethereum Network have been proposed: Proof of Stake (PoS) verification, sharding and . Each of these have some serious limitations, which I describe below. Layer2 solutions . Unlike Bitcoin’s PoW, PoS verification is not anchored in material value because it does not need an external resource. With PoW, this resource is electricity. The need for electricity ensures it is not economically viable to attack a PoW blockchain. Using energy to back a block also gives its immutability objectivity. Non-energy-based verification ultimately requires a subjective interpretation of immutability. In contrast, attaching energy to a block gives it a material existence in the physical world. Proof of Stake (PoS) has no material basis . Increasing blocksize means power is concentrated with those who can afford to have full nodes, which is to say, those with the most money. Currently, the entrance barrier to staking a node on the Ethereum Network is 32 ETH. This is different from PoW where miners compete through the use of energy. PoS is centralizing is a technique where a blockchain is split in multiple parts (“shards”) with different validating responsibilities. The idea is to lessen the workload on any individual node so there can be more of them. Yet, this solution only prolongs the problem. Sharding takes a single blockchain and turns it into multiple blockchains but does not eliminate growing blocksizes and the increasing the cost of staking. In short, sharding does not fix the underlying problem. Sharding are the only viable scaling option . New projects such as Raiden and the Loom Network are offering such solutions. Loom is especially promising as it enables sidechains on top of the Ethereum base-layer. However, Layer2 solution have security issues. Moreover, if Ethereum adopts PoS it will have an insecure base-layer, and it not possible to build a secure second-layer on top of an insecure base. Layer2 solutions in my view Smart Contracts Smart contracts lack the potential for mass adoption. Here is why. They are (see the Parity and the DAO hacks for example) and they are (i.e. not real contracts). Securing a smart contract is like proving a computer program doesn’t have bugs. It is very difficult as every computer program in existence has bugs. Moreover, to write traditional contracts takes years of study and passing a bar exam. To have real-world utility, smart contracts need at least that level of competence. Yet, people who write them typically don’t understand the law and how secure a contract must be. hard to write not legally binding There is also the so-called Smart contracts rely on the accuracy of outside data. But where does this data come from? What guarantees these oracles or not wrong or even corrupt? oracle problem. Other Issues with Ethereum There are other issues with Ethereum. One is its . The issuance of ETH is not capped. This means there is a constant flow of newly minted Ether, which makes Ethereum an inflationary asset. A second issue is the The number one of which is tokenization, used almost exclusively by ICOs. As the ICO trend declines, the price of ETH will decline. With this, Ethereum’s second use case (a capital raising platform) also dies. Tokenized securities and cryptocollectibles (ERC721) are potential future markets, but they will get nowhere as big as some expect (a topic for another article). Finally, with (smart contracts) and (side chains) issuance policy limited use cases for Ethereum. BTC also increasingly cuts into these two use cases Rootstock Drivechain Final Thoughts Blockchains have advantages in very specific areas, namely censorship-resistance and . Implementing a blockchain makes sense in contexts where immutability and the prevention of centralized control are important. The best use case of this is money. I can’t imagine a future where everyone holds an Ethereum wallet with 345 different tokens for 345 specific services. I also doubt Ethereum’s capability to handle this kind of data load. Ethereum is not leaderless (Vitalik Buterin and dev-team), not entirely immutable (e.g. DAO-hack and subsequent fork), not a good store of value because of an uncapped supply, and does not scale on the base-layer. This is why I am bearish on Ethereum. immutability Subscribe to my channels Medium and Twitter if you like my articles and want to learn about blockchain and cryptocurrency projects. If you have any questions about this article, please comment in the section below. To support my writing: 1PUzB7FtURwikfUg4zgdULyyeQkcK6o9fh Thank you!

Flow

Oracle

I am Bearish on ETH and Here is Why

And How That Increases Competition in FinTech In a recent survey conducted by PwC, 544 C-suite executives were interviewed on their views of blockchain and the future of banking. The study revealed an astonishing 95% of executives in banking believe elements of their business are at risk to various FinTech startups- many of which are powering their solutions through blockchain technology. Of the executives surveyed, 56% of them claimed to recognize the importance of blockchain. While financial services such as banking become concerned with FinTech taking hold of their market share, FinTech companies have concerns of their own. Is adopting blockchain technology a necessity to survive in this fast changing industry? While executives in traditional financial service companies may be losing sleep over the rise in FinTech, the benefits blockchain provides to financial technology is becoming more and more difficult to argue with. Increased Security Photo by on Felix Koutchinski Unsplash Centralized databases, the likes of which are used by banks, stock exchanges, and money transfer services are a vulnerability to hackers. These can be exploited to manipulate an entire database and steal user data. The very nature of blockchain and distributed ledgers make it difficult to penetrate. Distributed Ledgers are . decentralized databases What this means is: A hacker attempting to hack a centralized database has to break into one house to get what he or she wants. A hacker in a decentralized database has to break into every house in an entire neighborhood to be able to get what he or she wants from any individual house within the neighborhood. Faster and Secure Transactions It’s difficult to pinpoint an industry with more intermediaries than Finance. The transfer of capital within a country is inefficient, when the scope is broadened to include international money transfers, the problems are more abundant. The main issues contributing to these inefficiencies are the time spent transferring the money, and the security in which its transferred. With Blockchain, money can be transferred across the globe with a few clicks. The transfer time with smaller blockchains is often instant, and with Bitcoin and Ethereum can be anywhere from 5–30 minutes. There are no intermediaries involved, and the fees are minimal: smaller than found in any other money transfer alternatives. The sector that is especially ripe for disruption within Finance is money remittances. These companies charge a premium for transferring funds to other countries. If they do not change their fee structure drastically, they will soon become obsolete to blockchain fueled alternatives. Universal Customer Identification Processes Photo by on Kyle Glenn Unsplash Financial institutions currently spend around 500 million USD on processes for their customers. The process of performing this requires cooperation between several different financial institutions to verify information. Know Your Customer (KYC) Through the use of blockchain, an individual can verify their identity and KYC themselves through one platform. There are already projects on the market that aim to be immutable digital records of identity. Once a market leader has established itself, there can be an immutable ledger of pre-KYC’d individuals that financial institutions can securely procure data from. If you want to learn more about trading, technical analysis, or even just chat about the market Stop by and say hello, my Discord handle is RezaJates#5824 Cosmic Trading _Join over 10,000 members to talk about cryptocurrency, trading tips, signals, and other blockchain related topics!_discord.gg Cosmic Trading

Ledgers

BLOCKCHAIN DISRUPTCEPTION — How Disrupting Banking Disrupts FinTech

August 20, 2018 By , I’m referring to introductory content that is intending to teach a concept succinctly. I’ve avoided including chapters of books, which have a greater breadth of coverage, and research papers, which generally don’t do a good job in teaching concepts. Why not just buy a book? Tutorials are helpful when you’re trying to learn a specific niche topic or want to get different perspectives. tutorial I’ve split this post into four sections: , , , and . I’ve included a sampling of topics within each section, but given the vastness of the material, I can’t possibly include every possible topic. Machine Learning NLP Python Math If there are tutorials you are aware of that I’m missing, please let me know! I’m trying to limit each topic to five or six tutorials since much beyond that would be repetitive. Each link should have a different material from the other links or present information in a different way (e.g. code versus slides versus long-form) or from a different perspective. (Source via — Robbie Allen) good Machine Learning (medium.com/@ageitgey) Machine Learning is Fun! Machine Learning Crash Course: , , (Machine Learning at Berkeley) Part I Part II Part III (toptal.com) An Introduction to Machine Learning Theory and Its Applications: A Visual Tutorial with Examples (monkeylearn.com) A Gentle Guide to Machine Learning (sas.com) Which machine learning algorithm should I use? Activation and Loss Functions (neuralnetworksanddeeplearning.com) Sigmoid neurons (quora.com) What is the role of the activation function in a neural network? (stats.stackexchange.com) Comprehensive list of activation functions in neural networks with pros/cons (medium.com) Activation functions and it’s types-Which is better? (exegetic.biz) Making Sense of Logarithmic Loss (Stanford CS231n) Loss Functions (rishy.github.io) L1 vs. L2 Loss function (neuralnetworksanddeeplearning.com) The cross-entropy cost function Bias (stackoverflow.com) Role of Bias in Neural Networks (makeyourownneuralnetwork.blogspot.com) Bias Nodes in Neural Networks (quora.com) What is bias in artificial neural network? Perceptron (neuralnetworksanddeeplearning.com) Perceptrons (natureofcode.com) The Perception (dcu.ie) Single-layer Neural Networks (Perceptrons) (toptal.com) From Perceptrons to Deep Networks Regression (duke.edu) Introduction to linear regression analysis (ufldl.stanford.edu) Linear Regression (readthedocs.io) Linear Regression (readthedocs.io) Logistic Regression (machinelearningmastery.com) Simple Linear Regression Tutorial for Machine Learning (machinelearningmastery.com) Logistic Regression Tutorial for Machine Learning (ufldl.stanford.edu) Softmax Regression Gradient Descent (neuralnetworksanddeeplearning.com) Learning with gradient descent (iamtrask.github.io) Gradient Descent (kdnuggets.com) How to understand Gradient Descent algorithm (sebastianruder.com) An overview of gradient descent optimization algorithms (Stanford CS231n) Optimization: Stochastic Gradient Descent Generative Learning (Stanford CS229) Generative Learning Algorithms (monkeylearn.com) A practical explanation of a Naive Bayes classifier Support Vector Machines (monkeylearn.com) An introduction to Support Vector Machines (SVM) (Stanford CS229) Support Vector Machines (Stanford 231n) Linear classification: Support Vector Machine, Softmax Backpropagation (medium.com/@karpathy) Yes you should understand backprop (github.com/rasbt) Can you give a visual explanation for the back propagation algorithm for neural networks? (neuralnetworksanddeeplearning.com) How the backpropagation algorithm works (wildml.com) Backpropagation Through Time and Vanishing Gradients (machinelearningmastery.com) A Gentle Introduction to Backpropagation Through Time (Stanford CS231n) Backpropagation, Intuitions Deep Learning (nikhilbuduma.com) Deep Learning in a Nutshell (Quoc V. Le) A Tutorial on Deep Learning (machinelearningmastery.com) What is Deep Learning? (nvidia.com) What’s the Difference Between Artificial Intelligence, Machine Learning, and Deep Learning? Optimization and Dimensionality Reduction (knime.org) Seven Techniques for Data Dimensionality Reduction (Stanford CS229) Principal components analysis (Hinton @ NIPS 2012) Dropout: A simple way to improve neural networks (rishy.github.io) How to train your Deep Neural Network Long Short-Term Memory (LSTM) (machinelearningmastery.com) A Gentle Introduction to Long Short-Term Memory Networks by the Experts (colah.github.io) Understanding LSTM Networks (echen.me) Exploring LSTMs (iamtrask.github.io) Anyone Can Learn To Code an LSTM-RNN in Python Convolutional Neural Networks (CNNs) (neuralnetworksanddeeplearning.com) Introducing convolutional networks (medium.com/@ageitgey) Deep Learning and Convolutional Neural Networks (colah.github.io) Conv Nets: A Modular Perspective (colah.github.io) Understanding Convolutions Recurrent Neural Nets (RNNs) (wildml.com) Recurrent Neural Networks Tutorial (distill.pub) Attention and Augmented Recurrent Neural Networks (karpathy.github.io) The Unreasonable Effectiveness of Recurrent Neural Networks (nikhilbuduma.com) A Deep Dive into Recurrent Neural Nets Reinforcement Learning (analyticsvidhya.com) Simple Beginner’s guide to Reinforcement Learning & its implementation (mst.edu) A Tutorial for Reinforcement Learning (wildml.com) Learning Reinforcement Learning (karpathy.github.io) Deep Reinforcement Learning: Pong from Pixels Generative Adversarial Networks (GANs) (nvidia.com) What’s a Generative Adversarial Network? (medium.com/@ageitgey) Abusing Generative Adversarial Networks to Make 8-bit Pixel Art (aylien.com) An introduction to Generative Adversarial Networks (with code in TensorFlow) (oreilly.com) Generative Adversarial Networks for Beginners Multi-task Learning (sebastianruder.com) An Overview of Multi-Task Learning in Deep Neural Networks NLP (Yoav Goldberg) A Primer on Neural Network Models for Natural Language Processing (monkeylearn.com) The Definitive Guide to Natural Language Processing (algorithmia.com) Introduction to Natural Language Processing (vikparuchuri.com) Natural Language Processing Tutorial (arxiv.org) Natural Language Processing (almost) from Scratch Deep Learning and NLP (arxiv.org) Deep Learning applied to NLP (Richard Socher) Deep Learning for NLP (without Magic) (wildml.com) Understanding Convolutional Neural Networks for NLP (colah.github.io) Deep Learning, NLP, and Representations (explosion.ai) Embed, encode, attend, predict: The new deep learning formula for state-of-the-art NLP models (nvidia.com) Understanding Natural Language with Deep Neural Networks Using Torch (pytorich.org) Deep Learning for NLP with Pytorch Word Vectors (kaggle.com) Bag of Words Meets Bags of Popcorn On word embeddings , , (sebastianruder.com) Part I Part II Part III (acolyer.org) The amazing power of word vectors (arxiv.org) word2vec Parameter Learning Explained Word2Vec Tutorial — , (mccormickml.com) The Skip-Gram Model Negative Sampling Encoder-Decoder (wildml.com) Attention and Memory in Deep Learning and NLP (tensorflow.org) Sequence to Sequence Models (NIPS 2014) Sequence to Sequence Learning with Neural Networks (medium.com/@ageitgey) Machine Learning is Fun Part 5: Language Translation with Deep Learning and the Magic of Sequences (machinelearningmastery.com) How to use an Encoder-Decoder LSTM to Echo Sequences of Random Integers (google.github.io) tf-seq2seq Python (kdnuggets.com) 7 Steps to Mastering Machine Learning With Python (nbviewer.jupyter.org) An example machine learning notebook Examples (machinelearningmastery.com) How To Implement The Perceptron Algorithm From Scratch In Python (wildml.com) Implementing a Neural Network from Scratch in Python (iamtrask.github.io) A Neural Network in 11 lines of Python (kdnuggets.com) Implementing Your Own k-Nearest Neighbour Algorithm Using Python (machinelearningmastery.com) Demonstration of Memory with a Long Short-Term Memory Network in Python (machinelearningmastery.com) How to Learn to Echo Random Integers with Long Short-Term Memory Recurrent Neural Networks (machinelearningmastery.com) How to Learn to Add Numbers with seq2seq Recurrent Neural Networks Scipy and numpy (scipy-lectures.org) Scipy Lecture Notes (Stanford CS231n) Python Numpy Tutorial (UCSB CHE210D) An introduction to Numpy and Scipy (nbviewer.jupyter.org) A Crash Course in Python for Scientists scikit-learn (nbviewer.jupyter.org) PyCon scikit-learn Tutorial Index (github.com/mmmayo13) scikit-learn Classification Algorithms (scikit-learn.org) scikit-learn Tutorials (github.com/mmmayo13) Abridged scikit-learn Tutorials Tensorflow (tensorflow.org) Tensorflow Tutorials (medium.com/@erikhallstrm) Introduction to TensorFlow — CPU vs GPU (metaflow.fr) TensorFlow: A primer (wildml.com) RNNs in Tensorflow (wildml.com) Implementing a CNN for Text Classification in TensorFlow (surmenok.com) How to Run Text Summarization with TensorFlow PyTorch (pytorch.org) PyTorch Tutorials (gaurav.im) A Gentle Intro to PyTorch (iamtrask.github.io) Tutorial: Deep Learning in PyTorch (github.com/jcjohnson) PyTorch Examples (github.com/MorvanZhou) PyTorch Tutorial (github.com/yunjey) PyTorch Tutorial for Deep Learning Researchers Math (ucsc.edu) Math for Machine Learning (UMIACS CMSC422) Math for Machine Learning Linear algebra (betterexplained.com) An Intuitive Guide to Linear Algebra (betterexplained.com) A Programmer’s Intuition for Matrix Multiplication (betterexplained.com) Understanding the Cross Product (betterexplained.com) Understanding the Dot Product (U. of Buffalo CSE574) Linear Algebra for Machine Learning (medium.com) Linear algebra cheat sheet for deep learning (Stanford CS229) Linear Algebra Review and Reference Probability (betterexplained.com) Understanding Bayes Theorem With Ratios (Stanford CS229) Review of Probability Theory (Stanford CS229) Probability Theory Review for Machine Learning (U. of Buffalo CSE574) Probability Theory (U. of Toronto CSC411) Probability Theory for Machine Learning Calculus (betterexplained.com) How To Understand Derivatives: The Quotient Rule, Exponents, and Logarithms (betterexplained.com) How To Understand Derivatives: The Product, Power & Chain Rules (betterexplained.com) Vector Calculus: Understanding the Gradient (Stanford CS224n) Differential Calculus (readthedocs.io) Calculus Overview Hope you like this article !! Don’t forget to subscribe and like my blogs. Share some love too! Share: Press This Telegram WhatsApp Mohit Sharma(themenyouwanttobe&Co.) Connect with me Telegram My Latest Blog: THEMENYOUWANTTOBE

NVIDIA

Data Science Resources

Even though the majority of the world countries is embracing blockchain as the next step in technological evolution, Some countries do not share the similar viewpoint. You can read more about the “ World’s view points of crypto currency” . here Instead, some countries feel that Bitcoin or crypto assets are a threat to the existing system and they have put across stricter ban or controls to curb the crypto trading. If you hail from one of those countries, Like India, China etc and still want to be a part of the crypto revolution, or Over The Counter (OTC) products can be of a great help. ‘Peer-to-peer’ (P2P) Let’s first understand what is an Over The Counter (OTC) product? ‘Peer-to-Peer’ (P2P) or Over The Counter (OTC) or Peer-to-Peer’ (P2P) To understand an OTC market, Let’s first understand how an Exchange works. I believe most of you have traded in trusted online exchange platforms like Huobi, Hadax etc. In an Online exchange ecosystem, Exchanges which are in fact an organization acts as an intermediary. Sellers who want to sell crypto assets, quote their price and place their sell order in the common ledger commonly termed as the ‘order book.’ Buyers look for a buying price as per his liking or he can create his own buy order with his price and can place the buy order in the same order book. Once the exchange matches buy and sell orders it processes the trades. One of the downsides of online exchange is the time that may take to execute the order. There is a high chance that another seller or buyer can place a sell or buy order outmatching yours which may delay is executing your own order. Else if you are in a hurry you may have to execute a market order which can lead to losses. Another downside is that since online exchange platforms are registered firms they have to adhere to the law of land in which they have registered. For e.g In India when the banking Ban was imposed by the Reserve Bank of India(RBI), all the banks stopped banking relationship with Indian Exchanges which affected all the deposit and withdrawal of local currency. Whereas in P2P or OTC setup, all these won’t matters. A P2P or OTC exchange allows the sellers and buyers to trade directly with each other without any third party to process trades. In simple terms It’s like an open market, You can post your price of asset and the way you want to be paid( Bank Transfer, WU, Paypal etc). If a buyer feels that is a good deal, he can buy the asset from you by transferring the needed currency to your account or the mode of payment you specified. Once you confirm the payment received, the assets will be released to buyers wallet. Some of the advantages of P2P exchanges are Censorship resistance Almost Zero Fees Privacy Control over your fund The greatest challenge in P2P exchange is two things. Liquidity Seller Trustworthiness. When you select a P2P platform, It important to make sure you check all the above said points. Since I am from Indian working in Kuwait, I felt Huobi OTP is a great platform considering the above said points. Platform Analysis: Huobi OTC offers an exchange bridge between fiats and digital assets. Huobi OTC is an affiliated website of Huobi Global , is a platform devoted to over-the-counter trades of digital assets. One can buy/sell your digital assets by posting a buy/sell advertisement on Huobi OTC and then complete transactions with the other party so as to exchange your digital assets with fiats. Huobi OTC offers an aggregated display of buying or selling info and the actual payment is transferred and completed offline.（In other words, Huobi OTC provides escrow on digital assets . Huobi OTC Unless other OTC platform which are one country specific, Huobi OTC offers P2P trading in 6 countries which includes China and India. All the services offered are with Zero transaction fees, High security as well as 24*7 customer service which helps to resolve any dispute if any. Unlike most of the P2P platforms which are web based Huobi OTC offers various versions to trade. It has option of iOS, Android, Windows and Mac platforms. You can to download as per your preference. click here My Personal experience with Huobi OTC As I have already mentioned before, I am an Indian staying in Kuwait. In both countries converting BTC to local currency is very challenging . I had an option to use another P2P platform “localbitcoins.com” which gave me options of buyers in Kuwait. But one thing I have noticed was the vast difference in pricing. Most of the time the price difference was around 25–30% which mean a seller will have to endure heavy losses. The second option “localbitcoins” gave me was to visit a physical location of the buyer where I can personally visit and complete a trade. Again the issue was the same. The price difference and security of the transfer as in Kuwait a Bank transfer will take place in a day or two. Whereas the buyer was demanding transfer of my BTC to his wallet immediately. The chance of fraud was too high. Whereas in Huobi OTC most of the transfer happens in a 15-minute window. The sellers and buyers were mostly verified with the star rating which helps you to identify a genuine seller or buyer. One more advantage was the price. Even though the buy price was on a higher side, The selling price was on par with the market which ensures that you won’t suffer a loss. Also if there is a conflict you have the option to raise the same with the customer service team, which adds more security to the entire process. Also if you are buying BTC, you have more option to convert into either USDT or any other crypto assets with the help of Huobi Global, which is one of the most traded platform. For any one using an OTC platform I would recommend USDT. The greatest advantage of USDT is that it is a stable coin which ensures the price fluctuation is extremely narrow. This is because its is held in reserves in a 1:1 ratio. So if you are in a country having restrictions, OTC is a great way out. 100% backed by the USD If you don’t have a Huobi account, You can sign-up using this link !!! here Disclaimer : The material provided herein is general in nature and does not take into account your objectives, financial situation or needs. While every care has been taken in preparing this material, we do not provide any representation or warranty (express or implied) with respect to its completeness or accuracy. For the avoidance of doubt, this article is solely intended to be for general information on the usage of the Huobi Platform and does not in any way constitute as professional advice or financial advice. This is not an invitation or an offer to buy or sell cryptocurrencies, nor is it a recommendation to buy or sell specific types of cryptocurrencies. Trading cryptocurrencies on margin carries a high level of risk that may not be suitable for some. Consider your investment objectives, level of experience, financial resources, risk appetite and other relevant circumstances carefully. The possibility exists that you could lose some or all of your investments, including your initial deposits. If in doubt, please seek independent expert advice.

HackerNoon Tech Stories Archive August 19th, 2018

@dataturksDataTurks: Data Annotations Made Super Easy

@dataturksDataTurks: Data Annotations Made Super Easy

@victorrotariuVictor Rotariu

@BBerdahBrice Berdah

@CTM360CTM360

@ledfusionJordi Moraleda

@tarun_kohliTarun Kohli

@rohandave31Rohan Dave

@junaid.waniJunaid

@rashmiranjanRashmi R Padhy

@BerminalAppBerminal - The Crypto News Platform

@ricrubenRic Ruben

@duncanchiahDuncan Chiah

@TG34TG34

@shaunmstoneShaun Michael Stone

@ethanruheEthan Ruhe

@laurenjglass9Lauren Glass

@hhuuggooHugo Shi

@nicksukieNick S

@FatosMorinaFatos Morina

@hackernoon-archivesHicks Crawford

@mina.downMina Down

@jennifer.greenbergJennifer Greenberg

@mina.downMina Down

@rezaReza Jafery

@thems18Mohit Sharma

@hackernoon-archivesCrypto Pineapple

@dataturks
DataTurks: Data Annotations Made Super Easy

@dataturks
DataTurks: Data Annotations Made Super Easy

@victorrotariu
Victor Rotariu

@BBerdah
Brice Berdah

@CTM360
CTM360

@ledfusion
Jordi Moraleda

@tarun_kohli
Tarun Kohli

@rohandave31
Rohan Dave

@junaid.wani
Junaid

@rashmiranjan
Rashmi R Padhy

@BerminalApp
Berminal - The Crypto News Platform

@ricruben
Ric Ruben

@duncanchiah
Duncan Chiah

@TG34
TG34

@shaunmstone
Shaun Michael Stone

@ethanruhe
Ethan Ruhe

@laurenjglass9
Lauren Glass

@hhuuggoo
Hugo Shi

@nicksukie
Nick S

@FatosMorina
Fatos Morina

@hackernoon-archives
Hicks Crawford

@mina.down
Mina Down

@jennifer.greenberg
Jennifer Greenberg

@mina.down
Mina Down

@reza
Reza Jafery

@thems18
Mohit Sharma

@hackernoon-archives
Crypto Pineapple