...by Daniel Szego
"Simplicity is the ultimate sophistication."
Leonardo da Vinci

Tuesday, October 31, 2017

Notes on the need of Ethereum privacy

The Ethereum network suffers at the moment from a serious of privacy problem, every transaction is visible to the whole world. Perhaps, it might be an advantage in the direction of simple transaction validation, but it is a huge drawback in the direction of privacy. Even if public keys are not bound to an identity by default, if the identity is confidently leaked, the whole transaction network of an individual will be visible. It is simply not an acceptable. 

On solution might be the introduction of the Bitcoin style HD wallets, that simply creates a new address at each transaction. Certainly, it works well as we need only to transfer Ether or coins, however the situation might be a little be complicated as soon we need general smart contracts and access rights of the contracts . It is an open question how the HD wallet concept can be extended to cover general smart contracts with access rights as well.

A second possibility is to use mixers or similar functionalities. One way might be to use directly mixers as on the top of Ethereum blockchain, the second one is to build mixers indirectly into the protocol. However at both cases the problem is not exactly same as at the Bitcoin blockchain: as we simply transfer ether everything works fine, however as soon we have more complicated use-cases and general smart contracts, the model will be again difficult.

The third option might be to generally change the open style of the platform to a more private one. A simple solution might be to use Ethereum in a consortium Blockchain scenario. A more complicated is perhaps to build zero knowledge proof into the platform and consensus algorithm itself, implying more privacy by default. Certainly, the way of doing this is pretty difficult not only from a practical but a theoretical one as well.

Sunday, October 29, 2017

Notes on digital transformation

Digital transformation used to have a useful meaning and a couple of techniques and tools to deliver meaningful change into the society or business making. 

However, as the word "digital transformation" become popular outside the hard-core technology area, mostly in sales, marketing and consulting, it has been lost its meaning. Nowadays "digital transformation" is mostly a marketing buzzword to sell every piece of shit that has somehow something to do with IT.

Micro-economical considerations of the cryptocurrency mining

Cryptocurrency mining especially proof of work can be analysed from a pure micro-economical point of view. First of all, the market is about transaction validation. The mining economy is split into two parts: the miners provide the supply for transaction validation. Other actors, like users, merchants, exchange services that send transaction into the system provide the demand for transactions. 

It started somehow in a way that everyone could  run mining on a simple laptop, implying practically a perfect competition between suppliers with minimal fix costs and entry barriers. As the hardware requirements developed further the fix costs of the miners were getting bigger and bigger, implying centralization efforts which was manifested for example in the appearance of mining pools. It means that the transaction validation market moved from the perfect competition to the monopoly direction, having at the moment somewhere at the stage of oligopoly. 

In this sense we might as well define centralization and decentralization in the sense of market structure, meaning that in a total decentralized application the transactions of the application are executed on a supplier market that is globally distributed and is in the state of perfect competition.    

The rebirth of classical numerical economic theories at cryptoecomomics

Classical numerical economic theories, like microeconomics or game theory were once vital elements of every economic analysis. However in the meantime, even if these theories are part of the general university education, they seem to be less popular comparing to more "human-oriented" areas, like behavior based models. However, it is foreseen that these theories will have a rebirth in the area of public Blockchain and other distributed ledger technologies, implying the theoretical area as cryptoeconomics. 

In this area, we can actually work pretty well with the homo-economicus assumptions: we basically dealt with actors that are either humans, or simply computer algorithms, that do not know and do not trust each other, so the major motivation of the cooperation is simply the profit maximization.   

Supply demand analysis of the Bitcoin transaction fee

Recent events showed an extreme increase in the bitcoin price especially at the end of summer, implying situation in which the transaction fee itself was sometimes bigger than the amount to be transferred. Certainly with the SegWit and SegWit2x activations, the situation will be better, however the basic problem will not be solved. 

Transaction fees are basically priced on the market based on supply and demand (Figure 1). Demand curve seem to be relative simple, if the transaction fee is high, people tend to send as little amount of transactions as possible. If the price is low however, than a big number of microtransactions can be carried out without a huge cost implication. Supply curve however is a little bit special. On on hand, it might make sense to carry out transactions even if the transaction fee is zero or negative, because the miners reward is not only based on the transaction fees, but on a block reward as well. On the other hand, because of the block limit, the possible number of transactions is limited to around 7, meaning that the supply curve will change to vertical at this point, meaning that the supply is perfectly inelastic.     

Figure 1, supply demand curve for the Bitcoin transaction fee.

Figure 1, shows that cause of the fast increase of the transaction fee pretty well. As Bitcoin is slowly reaching the mainstream adaptation,  the demand curve will generally increase, slipping to right and high. As soon as the demand curve reaches the vertical part of the supply curve, even a small amount of change in the demand results a huge increase in the transaction fee.

The situation will be better, but actually will not be solved by the SegWit activation, as the vertical point of the supply curve will only be at a higher value but do not disappear totally. As a consequence, we will face with the same problem probably in a year as the general demand for Bitcoin transaction increases. 

Saturday, October 28, 2017

What is the difference between a real estate and a cryptocurrency

This was actually a general question in of of the community blog, however it is interesting generally, so I try to answer it generally.

Both cryptocurrencies and real estates are assets. The only difference is that a cryptocurrency is digital asset meaning the ownership is guaranteed by network protocols and cryptography. On the other hand, a real estate is a physical asset, meaning that even if they are administrated by a Blockchain protocol, the value of the ownership of the asset is guaranteed by the government.   

It is important to note however, that in both cases only the ownership of the asset is guaranteed, the value itself is not guaranteed. In both cases the value of the asset is based on pure market mechanism, namely on supply and demand. It is certainly true that at the moment the price volatility of a real estate is much smaller than at a cryptocurrency, however this might change on a long run because of the network effect, as cryptocurrencies will be reaching mainstream adaptation.

Blockchain as a management or leadership philosophy

"Agile leadership", "Agile management", "Data-driven leadership", "Lean management", "AI in leadership", "How AI will redefine the management", "Lean Leadership" .... Business and management philosophies seem sometimes lack of the original new ideas, instead they borrow ideas from the technology field and try to adapt somehow to an absolutely other field. Classical nowadays example is the Agile leadership and Agile management, which has got the original root in the software development, however it is widely used from classical project management to leadership philosophies with more or less success. Similarly lean, or 6 sigma were actually rather special techniques for production until they were adapted to a broader audience. AI (Artificial Intelligence) and machine learning have been starting to appear in the technology hype phase, however there are already initiatives to define data-driven leadership, and there is brainstorming as well to define AI in leadership.

Surely, there will be initiatives to adapt Blockchain or rather in broader aspect decentralization as a management, business or leadership philosophy. Actually, we have already seen the first initiatives, the DAO (Decentralized Autonomous Organisation) that were trying to replace some of the core management functions with smart contracts. Unfortunately the project was hacked, as a consequence the initiative did not really reach the Hype phase. However, the DAO is only one part of the spectrum. As it is possible to replace most of the core management administration and coordination with smart contracts, it is still pretty much an illusion to automate everything from leadership to doing business. On the other hand, management and leadership philosophies are simply products that are sold among the others to managers and leaders, so automating everything might not be the best selling argument for some of the target groups. 

As a consequence, I am sure that Blockchain or Decentralization will appear as a general management or leadership philosophy as well that has got the roots in the technology field itself but will adapt most of the ideas. How will this new philosophy look like ? Well, we can only guess: It will be probably prefer platform thinking, it will automate most of administration tasks with smart contracts (and hopefully rather with gamification than with some boring administrative tasks), it will be exponential, it will consider non-humans as well both and employees or as customers .... 

If it generally make sense to have decentralized management or leadership philosophy? 
That is actually a pretty good question but considering the recent Blockchain hype, the general philosophy side will appear, independently if it makes sense or not. Just as Lean, Agile and Data-driven have been already appeared on the management and leadership philosophy side. 

Notes on Initial Coin Offering

Well, ICO (Initial Coin Offering) it a topic that is pretty much highly discussed at the moment. On the one hand the whole are is basically a wild west, apart from the good intention start-ups, there are a lot of Ponzi schemes, Pump and Dump schemes, or simply tokens that provide nothing in exchange to the value of the token. 

On the other hand, the whole ICO market shows that the idea is actually a pretty much working idea, summarizing the advantages in three major points:

1. The area is booming of innovations: there are a lot of experimental initiatives to try out in which fields is it possible to use the decentralized model, how they might work and how they do not. 

2. There is pretty huge demand on the market: If I am someone that wants to invest some money into a start-up, it is pretty difficult to do it otherwise. Certainly I can buy some stocks on the finance market, however on the one hand stocks are investments of big companies, not exactly start-ups, on the other hand there is a financial border for most people throughout the world of investing in stocks. ICO on the other hand is available for everyone on a self-service basis. 

3. There is a huge supply on the market: for every start-up, there is not necessarily an easy way to collect funds. To get venture capital, is usually not an easy process and the result is pretty much questionable as well. On the other hand, latest news have been demonstrated that raising money with the help of an ICO can be pretty fast and pretty successful.    

Does Cryptocurrency mean the end of Fiat money ?

Perhaps the question itself is a little bit provocative, however we might as well consider the appearance of the cryptocurrencies simply from an evolutionary perspective. As barter was replaces by primitive, following coinage, then paper money and fiat currencies, perhaps cryptocurrencies are simple the new stage of the financial technology evolution. Certainly for the first run Bitcoin and other technologies provide rather a simple money transferring technology. However, there not seem to be any reason why the technology could not extend to a broader area, providing general financial services besides money transfer. One example for that is the appearance of cryptograhicaly funding services, like ICO-s that have the chance to replace the whole venture capitaslim. In this sense, we might as well easily imagine that in a twenty years, fiat style currencies will hundred percent replaced by crypto style financial services.    

Notes on purchasing power of a cryptocurrency

As cryptocurrencies are getting into the mainstream, we might as well start to think over if there is a meaningfully definition of purchasing power of a cryptocurrency.

"Purchasing power is the number and quality or value of goods and services that can be purchased with a unit of currency." - Wikipedia.

However, we need some differentiation here. For the first run, the question is what goods and services can you buy directly with the given cryptocurrency. Certainly there is always the possibility to exchange the given crypto into some other crypto or some fiat currency, however it is pretty much the question if we want to consider purchasing power this indirect way as well. On the other hand general price volatility is somehow feel to be somehow pretty high.

Considering these facts, I think we should distinguish two different measures: 

1. Cryptocurrency coverage: A measure that show what, how many goods or services, can you buy for a given cryptocurrency. It is certainly not so easy to formulate it exactly, however we can argue that Bitcoin has certainly a bigger coverage than Litecoin, which has surely a bigger price coverage than the other noname cryptos. 

2. Cryptocurrency purchasing power: it should look similar to the classical purchasing power, like we might as well interpret as the purchasing power of the crypto after changing it to US dollar for example. Certainly the US dollar choice is pretty arbitrary, the reason for that is that most services and goods that are sold by crypto at the moment are usually denominated in US dollar. However on a long run, perhaps better measures can be found as well.