Anselmo Zago edited this page Jun 2, 2018 · 10 revisions

Laissez-faire, let's be fair, let's play fair


White Paper on the Marketplace of Shared Economy Apps

Anselmo Zago, Mateus P. Dias, Tiago O. M. Santos, Guilherme Pagotto

April 2017


The free market allows the adoption of technologies that cause disruption even in its own kind of economic relations. An example of this is cryptocurrencies, which use Blockchain technology, and promises to be the future of money and means of payment. Another example is shared-economy applications, also called "Uberization", which are profoundly altering the trust-building model among unknown people. Trust is the currency in the economy of reputation. In this scenario emerges Letsfair, a platform that will use Blockchain to deliver a user-friendly technology infrastructure, aimed at businesses of all natures, for both classical and shared-economy commerce. It will address critical privacy issues for digital identities. Create an interoperable multidimensional reputation system between different businesses. Facilitate consensus among people in the division of labor and income. Besides allowing the issue of certificates of digital ownership and make negotiations for trade transactions easy.

1. Introduction

Free market capitalism is not only economically, but also morally superior to any other way of organizing the economic behavior of individuals. When voluntary interactions are carried out, there is no coercion and no one is obliged to support third parties or intermediaries.

In the free market, the consumers, through their decisions, are the ones who decide which enterprise should go ahead and become significant and which should disappear. This behavior is related to a kind of economic Darwinism, based on the principle of natural selection.

Today's technology has revealed some more free-market facets that were previously difficult to predict. The search for solutions to problems such as the scarcity of resources, in response to the increasing consumption around the planet, has resulted in the practice of sharing the use of products and services, facilitated mainly by applications that allow greater interaction between people. This is the so-called shared economy, also known as "Uberization", when applications such as Uber and Airbnb have provided a new model for doing business.

It is already possible to observe that technology is profoundly altering the trust-building model among unknown people, in ways and on a scale, that could not be easily noticed or predicted in the past. We are at a time in history where the reputational capital, generated in these new trust networks, is reinventing the concepts of wealth, markets, power and personal identity.

Reputation has already been intangible and is now continuously more concrete. Reputation is to the digital world what money is to the physical world, that is, it represents value. In this economy, an individual's online history becomes as or more important than their financial credit history. Reputable capital can create a major disruption in the market.

At the same time, the free market economy has allowed the emergence and rise of several digital currencies, also called cryptocurrencies, which are already generating movement in the global financial market without the direct intervention of central authorities. The technology behind these coins is called Blockchain, which can also have different practical utilities. Reputation systems are a natural application for this technology.

In this scenario emerges Letsfair, a platform developed under a Blockchain, that intends to facilitate the performance of various types of enterprises, providing digital infrastructure for the simplified management of identity, reputation, consensus, digital ownership, finance and negotiation.

2. Definition

Letsfair will be a collaborative, general-purpose platform for the free market and shared economy. It will stimulate the generation of reputation capital among individuals, for the establishment of consensus in the production and negotiation of goods and services of all natures.

Letsfair has at its core its most valuable asset, which is the reputation of an individual. It will allow strangers to establish secure business with each other. The building of online reputation will conceive positive changes in behavior, aiming at improving the standard of products and services, as buyers will tend to look for people or companies with good classification. The need to preserve and protect one’s reputation simply means that people will adopt better behavior, creating a healthy and prosperous environment for negotiations.

The platform is based on six pillars, which are its main functionalities:

  1. Identity - A digital identity system with a high level of privacy will be created, unlike current systems that have business models based on the exploitation of personal and behavioral data for commercial purposes, such as the sale of targeted ads. The goal is to put identity control in the hands of the user, who will be able to use it in other applications on the Internet. Each user will have a unique identity in the web, which solves the problem of lack of interoperability of information between organizations. Letsfair will promote the creation of more meaningful data, enriching attributes and allowing users to improve the quality of their digital identity over time, as they see fit, knowing the benefits of having a complete identity for their purposes.

  2. Reputation - Although successful reputation systems have been implemented across multiple web services, they are all centralized and fragmented in different areas. This prevents reputation data from being interchangeable and accompanying the user from one activity to another. Because reputation data is directly associated with identities, Letsfair will also place that information in the hands of the user himself, who can benefit from it in different services. Various measures must be considered to form many layers of reputation, which are built in several places, and reflect who the user really is. Letsfair will use a multidimensional reputation data approach, which will measure different dimensions of human characteristics.

  3. Consensus - Allows identities to use good reputations to create income and labor sharing arrangements. Letsfair will be a business facilitator among people from many different areas of the economy, promoting the decentralized organization of work, thanks to the agility and security in establishing consensus. It is an evolution in the historical concept of labor.

  4. Digital Ownership – Issue of financial assets and certificates through cryptographic tokens, which may have specific ownerships in the services of several economic activities.

  5. Finance - Digital wallets for users to store cryptocurrencies, necessary to make payments on trading transactions, and ensure the flow of information necessary for the generation of reputation data.

  6. Negotiation - Represented by applications that interact with the Letsfair platform, promoting the professional relationship between users. Establishes consensus and carry out the transactions of financial exchanges. May be native or developed by third parties within the community, and must act in the most diverse economic activities. The main application of Letsfair will be available at, and also in mobile versions, such as a large marketplace.

In short, the differentials of Letsfair, compared to other applications or business networks of the shared economy, are its own functionalities:

  • Allows the decentralization of numerous services. The use of a Blockchain will reduce the cost of financial transactions and ensure the integrity of operations;

  • Identity, reputation and financial data shall belong to the user himself, ensuring a high level of privacy;

  • Allows various economic activities to use identity and reputation data interchangeably;

  • Facilitates the consensus for the division of income and labor;

  • Enables the issue of digital ownership certificates;

  • In addition, it will also be the payment system itself, transcending current methods, such as card processors.

3. Trust is the currency in Reputation Economy

In a rational definition, trust is a type of risk assessment of the likelihood that something will work out, for example a business. It is the act of relying on analysis based on past experiences. However, it cannot be defined as something rational and predictable. Trust empowers people to connect securely with the unknown. Human beings are exceptional in giving votes of confidence. An example of this is the fact that we share our credit card information on websites. We rely on trust for our lives to work.

According to Rachel Botsman, a world-renowned authority on the power of collaboration and trust, technology is profoundly altering the trust-building model between people. In her presentation at TED in 2016 [1], entitled "We've stopped trusting institutions and started trusting strangers", she stated:

"Technology is creating new mechanisms that are enabling us to trust unknown people, companies and ideas. And yet at the same time, trust in institutions, banks, governments and even churches, is collapsing."

Airbnb, Uber, BlaBlaCar and Tinder are just a few examples of mechanisms that allow its members to rely on strangers. A change is underway, the transition from "institutional trust" to "distributed trust."

3.1 The Trust Stack

A research conducted by Rachel Botsman [1] demonstrates how technology is transforming social bonds and trust between people. By studying hundreds of networks and markets, she discovered a pattern that people follow, which she called "The Trust Stack”.

Trust stack; 1st trust the idea, 2nd trust the platform, 3th trust the other user

In the first layer of the Trust Stack, people have to trust that a new idea is safe and worth taking risks. The next layer is to rely on the platform, system or company that is making it easy to trade. The third and last layer refers to using little information to decide whether the other party in the business is trustworthy.

For example, to use BlaBlaCar, the consumer first needs to be convinced that sharing a car trip, reducing expenses, and getting into a car with a stranger is a possible idea. Of course, there are risks. Next it is necessary to trust that the service will not only remove the rotten apples, but also help solve problems, should they occur. And finally, trust that the driver and passengers present are honest. This is the process of someone going through the Trust Stack.

Over time, people will be open to changes in behavior. The more they experience this type of trust structure, the greater the disruption in existing systems, and eventually, even changes in the regulation of society and politics may occur. Laws that reduce risk, which add security and perhaps increase the penalty for fraudulent companies, are a few examples.

3.2 Trust is evolving

Through human history, trust has evolved in only three important chapters: local, institutional, and distributed. Society is now in the latter.

1st local, 2nd institutional, 3rd distributed

When people focus on being trustworthy, they demonstrate integrity in the way they behave. They do what they say they will, and with ethics. That is why trust was born locally. When you know a new person within a community, you already have some information about her beforehand, if she is who she says she is. In this way, there is consistently more social pressure for her to behave as expected.

Nowadays, the world has become accustomed to institutionalized trust. By not knowing other people in person, consumers end up putting trust in organizations with authority systems in the black box model, things like legal contracts, legislation and insurance, and less trust directly in individuals. Trust has become institutional and brokered.

On the other hand, it cannot be denied that, thanks to technology, society's way of trusting is changing. Trust is no longer hierarchical, it is being disaggregated and inverted. A new revenue of trust is emerging and is distributed among people, more inclusive and responsible.

Today, many people are comfortable getting into cars driven by strangers. With just a finger slide on the screen, they make appointments with other people they find compatible. They share their homes with people they do not know. These are powerful examples of how technology is stimulating trust among people in ways and on a scale, that was previously not possible. This is a break in paradigm.

This change will be further accelerated due to the appearance of blockchain technology, which obviates the need for reliable intermediaries. The implications are immense and will be seen by the Letsfair platform.

3.3 Reputation capital

People understand, in practice, what reputation means, as well as other valuation concepts that are part of the same semantic field, such as character, ethics, or even honesty. These are terms that indicate both the positivity and the negativity of a subject, institution or social group. Reputation comes from trust and develops over time as people interact with others in a consistent and repeated way.

Since ancient times, personal reputation has been a decision-making factor in various social situations. Nowadays, the difference is that network technologies allow you to digitally experience trust between strangers, through interactions and exchanges. Society is at a time in history where the reputation capital generated in these new networks of trust is reinventing the way people think about wealth, markets, power and personal identity.

The concept of online reputation is not new, and one can cite the example of eBay sellers, Amazon star ratings or reputation levels in online games like World of Warcraft. The point is that nowadays, with every transaction carried out, comment posted, contact established, distinctive assigned, an open book is built on who or what that person or organization is, before the members of that context.

But this wealth of data raises an important question - who owns reputation information? Shouldn’t online prestige, so difficult to obtain, be portable? If an Airbnb network member is an experienced host, shouldn’t he be able to use that reputation to get a loan at a future online bank? For what, in addition to the service itself, does that reputation serve in practice? Today, reputation data is not transferred between verticals and is not useful for other purposes outside the specific context. This is a misconception about information security aspects.

People should own their own reputation data, which also should have some real, not just subjective value, since they were the ones who have personally invested in its construction in various aspects. These data should accompany the individual from one activity to another. If a person starts selling books, why should they start from scratch? And when they move to a different country, it should be easy to redeem what has already been built.

Reputation is a reference within a context, but could it be transported to other situations? Would a flawless host on Airbnb be trusted to drive someone's vehicle on BlaBlaCar? Or when an Uber driver does a good deed by rescuing an animal run over by another vehicle that left him in the middle of the road, shouldn’t people be able to give him social honors?

"Real actions demonstrate the character of a person. When you build reputation in a particular system, it must be seen in the light of social dynamics, population and the unique characteristics of this system", explains Coye Cheshire, associate professor at UC Berkeley School of Information (ISchool), whose work focuses on the dynamics of trust in online interactions.

Many of the businesses that have entered the reputation economy are measuring different dimensions. On Stack Overflow website, for example, reputation is a measure of knowledge; on Airbnb, it’s a measure of trust; in Wonga, a measure of propensity to pay; in Klout and PeerIndex, it is a measure of influence.

Reputable capital does not consist of combining a selection of different measures into a single number. After all, people are complex and have their particularities; it is not fair to define them in single digits or binary classifications. The culmination of many layers of reputation, which the individual constructs in different places, actually defines who they are, under various factors, uses and behaviors.

3.4 The cost of lack of trust

An economic environment in which there is a lack of trust among its agents, relationships become more complex and, consequently, transaction costs increase, as well as risks and uncertainties. Sometimes such costs become so high that there is no incentive even for economic exchanges to take place. In other words, transactions are not completed.

In his main work, “The Society of Trust” [2], Alain Peyrefitte, sociologist, political scientist, diplomat, eleven times minister and member of the French Academy, explains:

"A society of distrust is a fearful, win-lose society: a society in which life in common is a game whose result is null, or even negative (if you win, I lose). [...] A society of trust is a society in expansion, win-win (if you win, I win), society of solidarity, common project, openness, exchange, communication."

For Alain Peyrefitte, the driving force behind development is trust, and in his book [2] he explains how the development of capitalism came from this theory. Mutual trust among people, trust of a society in its institutions, compliance with unwritten rules of mutual respect, and behavior that can minimize uncertainty lead individuals to know what to expect from others and institutions, and are vital and decisive elements to promote business, entrepreneurship, contracts, material progress and social development.

Distrust is something intangible and difficult to measure. Stephen Covey, in his book "The Speed of Trust" [3], exemplifies various effects of the low chain of trust. One is the cost of bureaucracy, presented as one of the seven low-trust taxes, estimated at $ 1.1 trillion in the US alone, accounting for more than 10% of its GDP.

The less confidence in the economic environment, the more it requires lawyers, judges, contracts and even non-monetary values. In addition to burdening businesses with a very high indirect cost, the climate of general mistrust causes the loss of countless opportunities and investments. The more trust there is, the more and better businesses prevail.

3.5 Blockchain in the chain of trust

In short, Blockchain is a decentralized, scalable, high potential peer-to-peer (P2P) network. Its integrity is based on a consensus mechanism. In practice, it solves a dilemma known among distributed-computing scientists called the "Byzantine Generals Problem." Its solution uses the concept of work proof to enable people who do not trust each other to agree on a common goal even when there are people with opposing intentions.

Blockchain has the potential to transform the way people and businesses cooperate. By creating a system of distributed consensus, the true disruptive potential of the blockchain is in the chain of trust, bringing significant benefits to society's day-to-day lives. It enabled the emergence of a new generation of highly scalable, robust and intelligent applications for the registration, exchange of information and ownership registry on: physical and virtual assets, tangible and intangible goods. Thanks to the key concepts of cryptographic security and decentralized consensus through a public ledger book, this technology can profoundly change the way a company's economic activities are organized and have social, political, and scientific influence.

Blockchain allows a second generation of economic platforms in a truly decentralized ecosystem, making people the owners of their own identity and digital reputation. In addition to allowing the reliable registration of products, intellectual or artistic ownership, as well as all other transactions.

Just as the Internet has made the information age available to everyone, blockchain tends to revolutionize trust on a global scale. Bringing benefits to financial inclusion and economic acceleration, the democratization of opportunities, and the insertion of an element that the world lacks desperately: trust in each other.

4. Ethereum

Ethereum is a digital platform which its main mission is the implementation of decentralized applications (Dapps) and Smart Contracts. Dapps are computer programs that remove the need for intermediaries in virtually any existing centralized service by allowing anyone to rely on an unknown counterpart to perform the most varied types of business. To this end, it provides a technological infrastructure for solving crises of trust among counterparts, which are so ubiquitous in the world, as well as offering a new range of solutions to the daily problems that are currently solved at high costs, such as: voting machines, name domain registration, registration of legal documents, transfer of goods, services, smart property and contracts between individuals, reputation systems and financial derivatives.

Letsfair will run under the Ethereum network in its public blockchain because a decentralized economic platform needs an encrypted network to exist, in which there is no central authority. The user's personal information, as well as their identity and funds will be under the control of the user. More information is available in the Project’s White Paper [4].

It may be that the Ethereum applications that will have the greatest impact in the world have not yet been invented, in the same way that it took several years for social networks to appear on the web. In the future, perhaps people will use applications without even knowing that they are using the Ethereum platform as a foundation. Letsfair aims to be one of these applications of great relevance even for the global economy.

4.1 Privacy

The market for security and privacy in the construction of robust and safe cryptocurrencies is constantly evolving. A feature that drew much attention to some cryptocurrency was the issue of privacy and anonymity in transactions, which allowed, for example, the cryptocoin Monero, to receive a rapid increase in value in a short period of time.

In addition to Monero, Zcash offers privacy using advanced encryption. Payments are published in a public blockchain, but the information on the sender, recipient, and amount of a particular transaction remains private. Only those with a “View Key” can know the content involved in the transaction. In addition, Zcash uses a zero-knowledge proof called zn-SNARK. The acronym stands for "zero-knowledge Succinct Non-Interactive Argument of Knowledge". It is used to prove that in the transaction no one is cheating or stealing.

Therefore, instead of just providing anonymity, Zcash offers a mechanism to solve a very serious problem that is allowing everyone to use the public blockchain without revealing anything, but at the same time revealing everything to those who need to know what they should, in a way that makes it impossible to corrupt the data.

In combination with Ethereum's intelligent contracts, absolute privacy technology can become very powerful, and it's one of the next big frontiers in the development of secure applications. It is with this feeling that there has been an approximation between the development teams of Zcash and Ethereum [5], which are currently working on mutual integration [6].

Originally, public blockchains have limited privacy requirements and are not readily compatible with the traditional standards of the information security industry. For this reason, companies in several areas are implementing the Ethereum in private networks and perfecting these implementations in bifurcated code, that is, derived from the original project.

With the same interest, major financial institutions, technology giants and natural resource companies such as JP Morgan, CME Group, BNY Mellon, Santander Bank, Microsoft, Intel, UBS, Wipro and British Petroleum have formed an alliance for the development of an enterprise version of the Ethereum (Enterprise Ethereum Alliance) [7], which promises to maintain compatibility and interoperability with the public version. This shows a significant evolution of the platform Ethereum, from the perspective of privacy.

4.2 Technical architecture of access to Letsfair

The core of Letsfair will be developed as smart contracts in the Ethereum network. As the users themselves will be keeping track of their personal data and funds, they will be able to have access to them directly through command-line consoles or through the decentralized browser. But since the goal is to reach a large audience, which is generally unaware of blockchain concepts, access to Letsfair will also be possible by simple web or mobile interface.

technical architecture

Communication with the smart contracts will be done by a middleware that will connect to the Ethereum network, which in turn will be accessible by an API, which will eventually allow access via the HTTPS protocol for the presentation layer.

Centralization through an API seems controversial to the purpose of a decentralized platform. However, what should be clear is that there will be no centralized data, but only a communication interface. All relevant information of the user will be accessed and persisted directly in the blockchain Ethereum, through the permission of the respective person in charge. This feature deserves attention and represents a differential from the point of view of digital business.

In addition to Letsfair's native application, others developed by third parties will also be available, which in the case of web or mobile access, will be able to interact with the centralized communication API. And in the case of decentralized applications (Dapps), they can directly access the core smart contracts.

The API will also be available for any websites or mobile applications that wish to integrate with specific Letsfair services, especially those related to reputation management.

5. Letsfair

The Letsfair platform is built on six pillars, which represent the main functionalities that will be available to its users.

Letsfair essentials

5.1 Identity

A critical aspect of any digital identity system, especially one that is intended to have wide application, is privacy. The question to ask is how to protect the data to ensure the privacy of the individual. However, with the use of Blockchain technology it will be possible to solve known identity problems better than what has been done by current models.

The key issues in question are: Identities with a low level of privacy; Identification and authentication activities within the same scope; Lack of interoperability of identity information between organizations; Lack of significant identity data.

Many of these problems are already addressed by Ethereum, since identity data is owned by the user, and protected by encryption, and must be unique for each individual throughout the network. Applications should only ask the identities for an authorization to run their services, and not take ownership of them and all their information.

The purpose of Letsfair in this context is to promote the creation of more meaningful identity data. Enrich the attributes and allow the user to improve the quality of their identity over time, especially in the aspects related to reputation.

A digital identity system with a high level of privacy is crucial for today's markets. This change will enhance the way people and organizations behave within the economy.

5.1.1 Digital identity and privacy

Large-scale business models that support many private sector identity systems, contribute to the increase of privacy threats. This is because their business models are often based on the exploitation of personal and behavioral data for commercial purposes, such as the sale of targeted ads.

Although monolithic identity providers, such as Facebook and Google, offer easy-to-use digital identity credentials, they follow this model that is contrary to consumer privacy. The business model should be designed to prioritize the protection of privacy.

One of the principles of privacy is individual autonomy, very important for the development of an identity system. The individual should be placed at the center of privacy protection. An identity system should be built primarily to ensure high levels of trust, not to serve the interests of the database owner.

Consult Hyperion [8], an independent consulting firm, globally recognized in the technical and strategic leadership in areas of electronic payment and digital identity, conducted a review of the privacy and data protection aspects of digital identity services in its report "Digital Identity: Issue Analysis "[9]. It identifies seven different models of digital identity architecture, which correlates distinctive degrees of centralization and decentralization with different levels of privacy.

positioning architectural models

Facebook and Google in particular, are the examples of a Monolithic Identity Provider model, considered highly centralized and with low privacy. This model is designed to be inclusive, that is, to maximize the potential customer base, but does not provide high-quality identification capability, as it tends to rely on self-confirmed data, since it is sufficient to achieve its end, that can be summed up in advertising in most cases. In this model, there are privacy risks associated with sharing large amounts of personal data with third party websites, often without the consent of the individual. This architecture provides an easy-to-use service, but is based on a business model that is not focused on privacy rights.

At the other end of the chart is the model with No Identity Provider, exemplified by the Blockchain technology. Due to its highly-decentralized feature, this model was defined by the study as the most prone to greater privacy. According to the report, the Blockchain community is in the process of exploring many issues surrounding identity, through cryptographic techniques, such as Zero Knowledge Proof, in which it is possible to carry out secure transactions without revealing the real data of those involved.

5.1.2 Identification separated from authentication

Three concepts are essential for a system of digital identities:

Identification: The user tells who he is. Authentication: It is verified if the individual really has control over the presented identity, usually validating a credential. Authorization: Determines which actions can be performed by the identity.

Currently, several identity systems combine identification and authentication activities within the same scope. These systems include a standard dataset that is always shared, even when it is not required for the service being accessed. Separating the authentication ID allows to minimize the disclosure of data.

While it is possible to use all relevant data to a particular operation, this approach raises unnecessary data or information and privacy issues. An appropriate approach is one that minimizes the amount of information needed. In many cases, the answer to a proof-of-identity question should simply be Yes or No, rather than "full name, date and place of birth, sex, address, phone, etc.".

Systems should be designed to respect privacy and release only the smallest amount of information possible. This means that to authorize many types of transactions, what is really needed is to know whether the identity is valid, and not the data of that identity. An example is proof of housing, where the person is required to provide documents proving the address of their residence. The answer that satisfies the criteria should be only Yes or No. The same logic is applied to other claims, such as the age of the individual, locality, if they have graduated, if they are registered to vote, among several other possibilities.

5.1.3 Interoperability

Isolated identity systems create identification information silos. It prevents the holistic view and delivers a bad user experience, who is submitted to repetitive sign ups and numerous logins with usernames and passwords. These isolated systems and the fragmentation of the identification markets limit the ways in which useful identity information can be generated, derived, and enriched. Although some applications today use Single Sign-On (SSO) logins, such as OAuth, OpenID and Facebook Connect, these systems keep data on centralized servers and do not prioritize the privacy of users.

Blockchain technology offers a promising solution to these problems. It puts control over the ownership of identity in the hands of the individual himself, and away from centralized services, after all a person's data is invariably owned by them. This decentralized approach of data pushes computing to a frontier where it is economically less valuable to hackers as it requires immense effort to individually break many identities one by one. Using this approach, the veracity of the information can be confirmed within the community to reduce fraud and allow users to only share the desired data with whom they really want.

On the other hand, organizations do not need to continually repeat identification procedures, reducing costs substantially. In addition to providing remarkable convenience to users.

5.1.4 Enriched identities

Interoperability between identity systems raises an issue associated with poor quality of identities and a lack of meaningful data. An example of enriched identity data is reputation. It is derived from the attributes of an identity and can be dynamically changed and used simultaneously within the blockchain environment.

The contribution of the Letsfair project to the decentralized identities will be the enrichment with verified attributes. Any user can define values in identity attributes, such as full name, country, state, city, address, zip code, date of birth, gender, SSN or government issued ID, parentage, picture, level of education, passport, email, social media profiles, blogs, websites, among many other groups of attributes that can be defined by the community. Companies and other organizations can also create their own identities, with attributes related to their type.

The user can individually define the visibility of each attribute as public or private, where, respectively, any individual on the network can view it or a permission to view the attribute will be requested when necessary, to execute a transaction with another identity.

It will be the users themselves who will verify the attributes of third-party identities. The verification will be performed in four different levels of credibility, being Bronze, Silver, Gold and Diamond. All four levels receive scores, which are incremented by votes from other identities.

At the level of credibility Bronze any user on the network can cast a vote for the identity attribute, certifying its truthfulness. However, this is the lowest level of credibility. At the Silver level, the identity can cast credibility votes only if a transaction with the individual has been finalized.

Gold is the level of verification in which there is manual validation, that is, when the counterpart is obliged to send an artifact that proves the veracity of the data. An example of this is the premium user accounts created in Exchanges of cryptocurrencies. In this case, the user is asked to send the scanned document and also a photo or video with the document in hand, with focus on the face. Only highly reputable entities will be able to offer this type of verification.

When receiving some verification of Gold level, the tendency is for other users to trust the verification already done and not feel the need to re-validate. In addition, in this case it will not be necessary for the counterpart to perform any verification, as well as receive attribute data, but only believe that the identity has already been seriously verified, and waive the breach of privacy.

At the Diamond level of verification only the entity generating the attribute in question will have the right to verify. An example would be a university to certify the graduation of its students. Another example would be the social network itself to certify if the profile informed in the identity in question is true and if it really belongs to the individual. Still another example would be the possibility of the entity that issues the passport in the country to certify this attribute. After all, for certain operations in the Letsfair network, such as the acquisition of goods, this level may be required by the other counterpart.

Inevitably, while the platform does not reach common and widespread use, the vast majority of the verifications will be performed at Silver and Bronze credibility levels. In addition, some attribute types will not even have a Diamond level, because there is no generating entity.

The verification of identity attributes is also important to rescue the prestige that the individual or organization already possesses outside the network, and to certify its legitimacy before the public. Consequently, the uniqueness of the identities will be guaranteed and credibility will tend to grow, creating a prosperous and reliable environment for negotiations.

5.1.5 Calculation of the identity quality index

The verified attributes enrich identities with many values, and together with the different levels of credibility associated with each attribute, it is possible to compare them to mathematic elements as matrices. To simplify all the information in a single numerical value, we can find a scalar magnitude, similar to the determinant of the matrix. In this way, we will have an index that defines the degree of relevance of the identity.

The quality index of an identity can be set on a scale of 0 to 10, with 0 being those identities with no quality and 10 those with the highest quality. The quality of identity is inherent in how it is associated with the real individual.

The different types of attributes of an identity make references to the individual in several degrees. This means that an attribute like zip code makes little or no reference to the real person, but on the other hand, an attribute like passport does a lot. Due to this variation, it is necessary to use different weights for each type of attribute, which is called the coefficient of identifiability, also defined on a scale of 0 to 10. In the example above, for the zip code could be used the value 1 and the for the passport, the value 10.

Quality of identify

Other important unknown aspects are the number of verifications done for each level of credibility of an attribute, being Bronze, Silver, Gold and Diamond, and respectively the total proportion to which each of these levels can total in the index: 10%, 20%, 30% and 40%. That is, the higher the level of credibility and its checks, the greater the weight of its values when calculating the quality index of each attribute.

AQI = Attribute Quality Index (0 to 10)

IC = Identification Coefficient (0 to 10)

Vb = Total checks for the Bronze credibility level

Vs = Total checks for the Silver credibility level

Vg = Total checks for the Gold credibility level

Vd = Total checks for the Diamond credibility level (0 or 1)

IQId = IQI of the entity generating the attribute for the Diamond credibility level (0 to 10)

The Identity Quality Index (IQI) is simply the highest value among all the attribute quality indexes (AQI) of the user. This means that the better the level of checks of an attribute with high coefficient of identifiability, the higher the identity quality index.

IQI = greatest AQI

IQI = Identity Quality Index (0 to 10)

AQI = Attribute Quality Index (0 to 10)

5.2 Reputation

An entity's reputation is usually estimated based on a history of relationships with other entities. Typically, a reputable entity influences others, and induces high trust in interactions with it. An individual's trust is derived from a combination of references received and personal experience, while reputation is a measure of collective trust, calculated based on references or qualifications of the members of a community.

According to Resnick [10], reputation systems should collect, distribute, and add several evaluations made by participants about the use of some service. Based on these assessments, a reputation value can be calculated for a certain entity. This type of system helps people decide who to trust, encourages interaction behavior, and suggests that they should not interact with someone who is considered dishonest.

According to Jøsang and Golbeck [11], the purpose of systems of trust and reputation is to strengthen the quality in communities by providing incentives for good behavior and punishing poor quality services.

5.2.1 Relationship of trust

Trust is the subjective and dynamic belief that is based on the honesty and ability that one entity has in another, in a given context, when considering past experiences and/or recommendations from third parties.

A relationship of trust is determined by the association between two entities and is characterized by the level of trust, the context, and the time. Context is one of the most important characteristics in determining the trust value of a relation. This is because two entities may have different trust values between them, depending on the context in which the interactions occurred. For example, there is a big difference between trusting a person to drive a car and relying on that same person to fly an airplane. It is the same person, who must be seen, however, from the standpoint of different capacities. Different tasks, in general, require different skills.

Previous experiences of an individual, obtained in a specific context, can be used to assist in determining the behavior of entities when similar needs are identified in the future. In addition, the relationship of trust evolves according to the interactions that occur in the contexts of the entities.

5.2.2 Reputation systems

According to the W2OS report [12], eBay currently has the most widely used reputation system and processes more than one billion transactions per day. Each transaction can result in two reputation points, one being left by the buyer and another by the seller. It is essential that reputation systems deal with large numbers of transactions and significant data.

E-commerce reputation systems often implement multidimensional reputations, which allows the user to evaluate the seller on a number of factors such as cost of shipping and quality of communication. All major e-commerce sites use the traditional client-server model, where reputation data is stored centrally, computed, and distributed on a centralized server, and all customers can request to see this data.

In the eBay system, the percentage of positive feedback is calculated based on the total number of positive ratings, and negative feedback is calculated based on the operations over the last 12 months, excluding repeated feedback from the same member for purchases made within the same week [13].

Reputation scoring, when centrally calculated by the e-commerce site, has the negative effect of the company being able to change the reputation calculation algorithm and force its implementation to all users without their consent. For example, eBay recently prevented sellers from leaving negative feedback to buyers. This appears to be a policy that forces the seller to be made responsible for frauds, weakening the seller's authority in his business space and the right to sell or not to whomsoever he wishes.

Although successful reputation systems have been implemented in various web services, they are all based on the centralized server model, which makes them unsuitable for implementation in a P2P network, where the main purpose is the decentralization of control, distancing from a single authority. So far, effective communication and the sharing of information regarding trust and reputation remains an unresolved issue [14].

All reputation systems, no matter how they are implemented or the type of network they constitute, face the same fundamental problems: the ability to bind an identity to a single user and perpetuate it beyond the lifetime of that service, site, or network. Preventing the user from obtaining more than one identity is essential to avoid exploitation of the system with transactions between false identities.

Another limitation, which remains an open issue, is how to quantify reputation. Also, how is it possible to ensure the accuracy of the reputation score left by a user and whether it is based on a real transaction or not.

5.2.3 Vulnerabilities

Reputation systems have many types of vulnerabilities that make them relatively easy targets for attacks and manipulations cited in [11] and [15]. In the literature, some examples of vulnerabilities are reported, such as:

Playbooks: consists of a sequence of actions that maximizes the gain of a participant according to certain criteria. A simple example of this type of vulnerability is a user who over a period of time performs services honestly and receives a high reputation value, but after a while, with this reputation value, offers low quality services, using his reputation as a weapon to cause fraud.

False evaluations: the ratings provided do not reflect the true opinion of the appraiser. This behavior is considered unethical and represents a type of attack. However, it is difficult to determine when this type of attack occurs, because agents in a community do not have access to genuine user opinions, and can only see what other users express.

Discrimination: it means that an entity provides high quality services to one group and low quality services to another group.

Collusion: this attack is based on a group of entities that conspire with the purpose of diminishing the reputation of another entity. For example, they provide unfair or false recommendations, or practice discrimination.

Proliferation: it means that one agent offers the same service over several different channels, thus increasing the likelihood of being chosen by a third party.

Late Reputation: it means that the attacker uses the time interval between an instance of a service provision and the evaluation effect by the corresponding service, for example, to provide several low-quality services in a short period of time before their reputation undergoes significant degradation.

Reentry/change in identity: within a community there may be different sub-communities. Reentry/change in identity means that a low score entity leaves one sub-community and subsequently joins another with a different identity, thus avoiding the consequences of the low score associated with the previous identity within the community.

Value Imbalance: this type of attack is possible when the value of an evaluation is not proportional to the actual value of the evaluated object. An example of this type of attack is the designation of low values, in great quantity, to the services made available by other users and the assignment of high values to services of only a few users, but in a small amount. This type of attack is considered misleading and unethical.

Sybil Attack: is when a single entity establishes multiple false identities within a domain of trust and reputation systems to provide multiple evaluations about the same service.

Little incentive to provide reviews: entities have little incentive to provide reviews. If everyone did the evaluations, then everyone would win, but no one wins directly by performing the peer evaluation.

Difficulty in getting negative feedback: in some cases, it can be challenging to get feedback on negative experiences because of people's reluctance to offend the people evaluated. One study [16] showed that users may retaliate in the event of negative feedback, and thus, evaluators are less likely to make such an assessment.

Ballot stuffing: also known as shilling, it means to artificially raise the reputation of an entity. This is done by adding opinions about transactions that did not occur in the system. This attack takes advantage of the facility of falsifying the occurrence of an interaction to inflate reputation. Similarly, it is possible to destroy the reputation of an entity by sending several negative opinions.

Notorious attackers: the existence of participants whose only purpose is to disturb order in a community, for whom incentives for good behavior or sanction for bad behavior have no effect.

5.2.4 Immunity mechanisms

Just as the literature reports several types of vulnerabilities in reputation systems, it also suggests some immunity mechanisms to mitigate them. Perhaps the most difficult attack to avoid is that of false evaluations. Jøsang and Golbeck describe a possible defense against this type of attack, being that of comparing the evaluations of some users in the network with others left by more reliable users [11]. This means that reviews provided by low reputation users have less weight in the final evaluation. Another approach found for this type of attack is to calculate reputation using only the latest iterations, a technique known as aging.

Collusion is another very common attack on reputation systems. One solution to a collusion attack is to perform the reputation score calculation based on the average of all reputations received from a pair. This type of attack is often used in conjunction with the Sybil attack.

The Sybil attack represents one of the most elemental attacks of authenticity. John Douceur describes how the success of a Sybil attack depends on the cost of obtaining an identity [17] and clearly shows how the effectiveness of an attack is reduced when the cost of creating a new identity increases. The most effective countermeasure is to link identities to real humans, as described by Haifeng Yu [18]. However, such a countermeasure penalizes the network itself, due to the resources needed to validate each identity.

The reentry attack also exploits the low cost of entering a network. Baptiste Prêtre evaluates this attack as very efficiently, not only because of the low cost of entry but also because network participants (people) consider a user with zero reputation points of higher value when compared to a user with negative punctuation, which provides the user an incentive to abandon the identity [19].

For the attack by reputation inflation, one way to deal with the problem is through proof of transaction [20]. An evaluation will only be valid if a proof of transaction, certified by the assessed entity, is presented.

Another vulnerability of reputation systems is the little incentive given to users to provide reviews. Xiong in [21] encourages evaluations by incorporating in the metric a component related to the percentage of evaluated items, so that a higher rate of evaluations generates an increase in reputation score.

To avoid the problem of retaliation for cases of negative feedback, several solutions have already been proposed for the preservation of privacy. Some of them try to hide their identity in the evaluation [22], while others try to hide their own assessment by consolidating the overall reputation score [23]. However, in some cases the identity of the customer is revealed, for example in e-commerce, in which most situations the client's address is required to send the purchased items.

5.2.5 Reputation in the Blockchain

Reputation systems tend to become natural applications of blockchain technology and this project makes this application clear. There have already been some attempts to build these systems [24]. But such solutions haven’t stand out effectively yet [25].

As with all reputation systems, classification data is assigned to the user's identity. In the blockchain, this information receives all the advantages of an identity in this environment, such as: user-assigned custody, high level of privacy, interoperability of information between organizations, among others. In addition to being able to be changed dynamically and used simultaneously throughout the environment, it encourages the individual to truly work to improve their reputation. For it is the single point, which reflects to all the entities within the network, their level of reliability.

Reputation data further enriches the decentralized identities in the blockchain and, in addition to the proven attributes provided by Letsfair, significantly reduces the cost of obtaining multiple identities. This contributes to the reduction of several problems in reputation systems, including those of the security systems mentioned above.

5.2.6 Multidimensional reputation

In relationships of trust, one of the most important characteristics is the context. Two entities may have different trust values between them depending on the context in which the interactions occur. Reputation reflects these values in their various situations, and the characteristics of social interactions must be observed through different and in specific situations.

The human being is complex and it is not fair to define their reputation by a single number. Various measures should be considered to form many layers of reputation, which are built in different places, to really reflect who they are.

Multidimensional reputation data must measure different dimensions of human characteristics. This approach is applicable to the decentralized identities in the blockchain. Similarly, as the Letsfair adds value to attributes that are verified in identities, it also enriches them with multidimensional reputation data.

The purpose is to establish, along with network participants, a taxonomy of human characteristics, represented by values and skills. In information systems, the taxonomy is used to classify and organize data in a hierarchical way, to facilitate its identification.

taxonomy of human characteristics

Characteristics of values can be exemplified as: cordiality, efficiency, enthusiasm, generosity, honesty, humility, integrity, loyalty, maturity, optimism, patience, parsimony, sincerity, among several others. But the scoring for these types of values depends very much on the context in which economic transactions are made. In a more pragmatic approach it may be necessary to adopt measures of values such as: propensity to pay, communication, quality of products/services, fulfillment of deadlines, among others.

Competence, according to the best authors of management, is the intercession between three elements: knowledge, skill and attitude. It concerns the individual's ability to perform tasks or certain functions, and their level of performance. Competency attributes should be listed and ranked in all its possible variations, for all skills of all professions.

Most importantly, the attributes of values and competencies can be created according to arket demand, and the score of each characteristic for each user will be unified in the network and can be shared among all the activities in the economic ecosystem.

5.2.7 Calculation of reputation

Whenever there is an economic transaction between two entities, that is, there is an exchange of product or service for financial assets, both entities will have the right to evaluate one another on certain reputation attributes. These attributes refer to economic activity, and are chosen by the developer of the trading application used.

The evaluation is performed by means of the star-ranking metaphor (as in hotels), using a scale of 1 to 5, for each of the related scoring attributes, when 1 means poor (negative); 5, excellent (positive); and 3, fair (zero point). This methodology is powerful because not only it provides a simple, empirical, and easy-to-understand metric, but also because people are generally willing to vote without anguish by just star-clicking.

The reputation of an identity will be visualized through a card, with the summary the scores for the most valued attributes. The platform will not perform a calculation to obtain a unique reputation value for the identity, but the reputation score for each of the reputation attributes will be calculated individually.

The observer user, the one who checks the reputation of another user to assess the risk of a future transaction, is the one who will decide which attributes will be most relevant to their judgment. In the reputation card the summary of the most relevant attributes will be displayed and it will also be possible to consult the list with all the attributes that have been punctuated so far for the identity.

Importance of IQI in calculation of reputation

The Quality of Identity Index (IQI), discussed earlier in this document, is critical to reputation score. This means that if the quality of the evaluating identity is low, their score will have a reduced weight in the calculation of the final reputation of the attribute. Likewise, if the quality of the identity is high, the weight of its evaluation will be greater.

The influence of the quality of the identity in the calculation of the reputation refers to the PageRank algorithm [26], originally introduced by Brin and Page in the Google search engine, to position websites among the results of their searches. PageRank measures the importance of a page by counting the quantity and quality of links pointing to it. In the same way, the number of evaluations and the quality of the evaluating identities influence the reputation attribute score.

The score for each star-rating is calculated by summing the IQI (s) of all individuals who casted a vote for the respective score.

Average rating

Many e-commerce sites erroneously perform the calculation of customer ratings. For example, imagine the subjective comparison of two products:

Product A - Average rating: 5.0 (1 rating)

Product B - Average rating: 4.5 (50 ratings)

Which product has the best rating? It seems clear that Product B has more appraisers and many of them are probably satisfied, even with the average rating below Product A.

This means we need something smarter to perform the rating, than just calculating an average. After all, the average applied without standard deviation and analysis of samples, does not present real meaning, because above is presented only a quantitative criterion of calculation. It is also necessary to have some empirical metric for the average classification that incorporates in the score the number of evaluations. Most of the methodologies available today use Bayesian inference, which considers when there is not have enough data to make an estimate using an average.

Bayesian inference

When it comes to probabilistic models to evaluate reputation, implementations of reputation calculations involving Bayesian inference are used, as in works [27], [28] and [29]. Bayesian inference is a statistical method based on the Bayes' theorem, which considers information on a quantity of unknown interest in statistics to be fundamental. The idea is to try to reduce this misunderstanding.

Paul Masurel, in his article [30], explains the use of Bayesian estimation for the calculation of star classification. Instead of trying to calculate the estimate directly, it is first calculated a probability distribution that describes the observable space, that is, what is already known of the value to be estimated, and then, and only then, will it be possible to extract an estimate of this value.

The Bayes' theorem allows the calculation of the following probability:

P(O) is the probability of prior observation. P(X) is what is believed to be the distribution of the probabilities X before checking any data. When nothing is known about the probabilities, one can start with the assumption that there is an underlying uniform distribution where all possible values of the random variable X are equally likely. However, we are interested only in the proportionality relationship between P(X|O) and P(O|X). The probability of P(O|X) is given by multiplying the probability of each observation, assuming they are independent:

For each observation, the positive probability is X, and the negative probability is 1-X, so if we observe K positive for N observations, we can derive the posterior probability as follows:

This is called a binomial distribution. As we collect more observations, the distribution becomes more and more refined, and with that refinement we can compute a range in which the value is likely to be.

As an example, consider a first positive observation, the posterior distribution is given by K=1 and N=1, which is plotted in the first chart of the figure below. After 5 observations, we have 4 positives and 1 negative; after 10 observations, we have 7 positives; after 25, we have 18; after 50, we have 37; and after 100, we have 72. The distributions at each point of the observations are plotted in the charts below:

As we can see, the more observations we get, the more refined the posterior distribution becomes, and the more confident we are with the interval that surrounds the value with the maximum probability of distribution.

Joint distribution

To use the Bayesian estimation to calculate the posterior probability of star-classifications, we must use a joint distribution. This is because we are not estimating the distribution of some scalar value X, but rather the joint distributions of the estimates that the evaluator will give in ranking of 1, 2, 3, 4 or 5 stars, and not just simple positive or negative. In this case, the random variable is a categorical distribution because it can take some value within {1,2,3,4,5} with probabilities as follows:

The reasoning for the binomial distribution still applies, the probability is still the product of the probabilities of each observation, and each individual observation is given by an associated probability.

This is not a binomial distribution, but a multinomial distribution, with a parameter that can be expressed as follows:

If we include our prior as a distribution in exactly the same way in the proportionality equation, as for example a Dirichlet Distribution with parameter 0, we can factor the distribution as follows:

This is another Dirichlet distribution with another parameter, which can be characterized as follows:

Expected averages

Since we are primarily interested in estimating the average star-rating, we can rephrase our problem as: "What is the expected value of the average rating, given the posterior, in the form of the Dirichlet distribution?".

The average value of a categorical random variable is the weighted average of the values of the random variable weighted by the probability values respected. In other words, the sum of the probability of obtaining a star, given the observations, multiplied by the value of that star for each star value from 1 to 5. Thus, for the categorical variable of star classifications, the average value would be:

The expected value of the average rating based on the posterior is then calculated for the star ratings as follows:

It is possible to calculate the probability of a star value given to the observations, as the ratio of the Dirichlet parameter to that star with the sum of all other parameters:

This probability can be connected to the expected value, with a small simplification, as follows:

In Paul Masurel’s article [30], this formula is digested to make it usable in real life. The Bayesian average would then be:

The parameters T and N are known, but in the Bayesian average for star classification it is necessary to define values for the parameters C and M. These amounts are better defined when there are already many relevant data, to adjust to a Dirichlet distribution. However, it is common to choose only a couple of parameters that mimic the behavior sought.

Let's go back to the example of Products A and B. We can adopt two possible values, such as C = 5 and M = 3.

As expected, Product B has better Bayesian average in comparison to A.

Conclusion on the calculation

The formula presented in the previous topic requires some adjustments to adapt to the needs of Letsfair to consider the weights of the ratings according to the IQI of the respective identity. Therefore, the final formula is given by:

An interesting fact in calculating the score in Letsfair is that the IQIs are dynamic. That is, they are incremented according to the progress of identity attribute verifications. In this way, the reputation score is also positively affected as the evaluating identities are enriched. This means that if one identity is evaluated by another in the past, and at that time the IQI of the evaluator was low, its assessment did not weigh heavily on the calculation. However, as that identity improves its quality, it also improves the reputation score it has evaluated, without necessarily occurring new iterations. This is a practical example of how the quality of identities is of utmost importance for the strengthening of the economic ecosystem in general.

This calculation demonstrated cannot predict abrupt behavioral changes in the network, such as, for example, a service that was good and has recently become bad. This will be resolved later by adding specific weights so that the most recent evaluations have a greater significance in the reputation of those individuals or entities. Preventing that the reliability of that transaction be doubtful.

Reputation calculation results in the most important information for decision making, and it is at the heart of Letsfair. Reputation is the most valuable asset of the individual; the process of its construction generates positive change of behavior. The need to preserve and protect reputation simply means that people behave better by creating a trustworthy and prosperous environment for everyone.

5.3 Consensus

In the realm of virtual currencies, much is said about the decentralized consensus model, which allows the blockchain to function as a distributed registry, without the need for a central authority to validate transactions. The nodes within the blockchain network execute the same algorithm, which generates consensus according to the established rules, and thus validates and registers each transaction in the unchanging ledger.

But in Letsfair, the idea of consensus has a more humanized approach. It is not about the consensus between algorithms, but the establishment of agreements between members of a group or community, where everyone participates in the decision making. It is, in short, evolution rather than competition.

By creating a trustworthy economic environment where the reputation of each individual is common knowledge, the relationships between them are facilitated and the natural consensus process becomes a reality. In a practical way, in the Letsfair platform, the consensus will be established by agreements of division of labor and income.

In the negotiation applications developed by the community itself, for the numerous existing economic activities, rules will be implemented for the respective work processes as well as agreements for the division of the income obtained. The products or services offered to customers, most often are the results of the joint work of several people. These applications, aided by the capabilities available on the platform, will be responsible for organizing the work process and will also be the providers of compliance mechanisms. It is they that actually facilitate consensus among individuals.

As an example, many software developers can partner to offer the custom application development service. The responsibilities and percentage of income must be pre-established initially, and at the time of payment by the client, the money will be distributed automatically as agreed between them. In the end, all parties involved should evaluate themselves in the reputation system.

The assemble of physical products can also be divided into a sort of distributed production line, where unknown professionals come to consensus based on their reputation scores. The assembly process begins only after the sale, and each professional must deliver the pre-made product to the next entity in the production chain, until it is delivered to the buyer. All parties involved should also evaluate themselves later.

Distributed production can be used, for example, for the licensing of products developed for the sake of a cause. The consensus among all parties involved would not come about naturally without a decentralized intermediary, but with Letsfair it is possible for everyone to enter into an agreement and each party can only do its job without worrying about the process as a whole.

Another example that can be mention is that of imagining an application that offers educational services in the form of a decentralized university. Several groups of teaching activities can be approached, such as undergraduate, postgraduate, professional education, support activities, sports education, art and culture, languages, in short, any knowledge that can be aggregated and transmitted.

The teachers alone, assisted by the application, can organize themselves to define the curricula of the courses, and many teachers are needed to cover all topics of interest. It also takes many other people to produce these courses. This whole range of professionals must be organized in a workflow that meets the pre-established demand. In addition, as the courses are marketed, the income earned must be distributed in the manner previously agreed upon.

Just the same, many other applications would serve as examples, with decentralized business models such as maintenance and repair services, real estate activities, miscellaneous manufacturing, printing and reproduction, e-commerce, passenger and cargo transportation, couriers and delivery, electronic publishing activities, video production, music production, software development, news agencies, financial services activities, pension plans, insurance, health plans, legal activities, accounting, auditing, architectural and engineering services, scientific development, advertising and market research, photographic activities, veterinary activities, tour operators, reservation services, security and private investigation, social and health services, artistic, sports and recreational activities, among others.

The Letsfair platform, along with custom trading applications, will be a business facilitator among people from many different areas of the economy, and not an intermediary like in the centralized economy. The decentralized organization of labor, thanks to the agility and security in the establishment of consensus, is an evolution in the historical concept of labor, changing positively even the organization of society.

5.4 Digital ownership

On Letsfair, any entity can issue digital certificates, called tokens. Such certificates may be transacted between entities and represent an asset, such as goods, values, credits or rights. Cryptocurrencies in their essence are digital tokens, which represent a fungible transactional good.

The tokens on Letsfair will have specific properties and can be used in several different situations. As an example, an artist can issue a token to certify the authenticity of his work of art, and transfer it to the buyer. A university can issue tokens for its students as a course completion certificate. A singer can offer certificates as concert tickets. Just as a movie theater can sell tokens as tickets to watch movies.

In addition to being used as tangible property titles, tokens can also be used for digital ownership. That is, when buying a digital product, such as music, movie, book, photo, game, software, etc., the buyer can receive a token representing the ownership of this digital artifact, and this will give him the right to access it for consumption. Because the digital certificate is only associated with the identity that acquired the digital product, and it is the key to opening that product, preventing other identities from being able to enjoy a copy, we realize that this functionality can minimize even digital piracy.

Another utility is an industry being able to issue a certificate for each good produced, and transfer it to each buyer. Specific features can even be defined to be run only by the owner. As a hypothetical example, a car manufacturer could offer tokens as property rights, and this be used to unlock the doors and start the vehicle from a smartphone. That way, only the owner of the token will have the right to use it. Still using this example, at the time of the resale of this automobile, the simple transfer of the token unifies the process of financial transfer and ownership.

Another possibility allowed by this technology, but still far from becoming a reality, is a hospital issuing birth and death certificates. The latter would guarantee the execution of the inventory contract of the individual, making the automatic sharing of their assets. Just the same, marriage certificates may also exist, with the stipulation of the self-executable property regime. Such rules differ among countries and cultures. But they could be defined in situations where there is a community of stakeholders.

5.5 Finance

Letsfair will provide wallets for its users to store digital currencies, that is, cryptocurrencies that have interoperability with the Ethereum network. Such currencies are required to make payments in exchange transactions, which must be performed within the platform to ensure the flow of information necessary for the generation of reputation data.

In addition to cryptocurrencies, it is also necessary to use cryptographic assets with a stable value. That is because one of the main problems in using cryptocurrencies to make payments is the issue regarding the volatility of their values, which experience large fluctuations in a short period of time. One approach to solving this problem is to use cryptographic assets that accompany stable asset rates, such as fiats money issued by governments, like US dollar.

At first, it is not part of the development scope of Letsfair the creation of its own stable assets. The proposal is to use the best assets already available in the market, since the asset must have liquidity, which means that it should allow the user to buy or sell quickly, even in large quantities. However, we cannot exclude the possibility of creating our own stable assets, if it is feasible and necessary.

Entry and exit of money from the platform will take place through the exchange of cryptocurrencies, and a mechanism will be provided to facilitate and encourage this exchange through them. Usability should be prioritized for the user, allowing people with no familiarity with cryptocurrencies to easily enter or withdraw money from their digital wallets.

From the medium to long term, the goal is that users do not consider it necessary to withdraw money out of the platform. As the ecosystem evolves, with the creation of new applications for the most diverse economic activities, the user will be able to receive their income and enjoy it fully within the limits of the network.

The goal is to bring the use of cryptocurrencies to people's daily lives, just as social networks are used. Once many individuals and businesses receive their payments through Letsfair, the use of trading applications will be common, and the method of payment through smartphones will compete directly with current methods, such as card processors.

The advantages will be that, in addition to automating some or many processes within the business, the cost of the payment operation will be very low due to the use of cryptocurrencies. These advantages are irresistible in any economic activity, and will put Letsfair at the center of the current financial market's disruptive trend.

5.6 Negotiation

The mechanisms of negotiation are those that promotes the professional relationship among the users, establishes consensus and carries out the transactions of financial exchanges. They are represented by all the applications that interact with the platform. They can be native or developed by third parties within the community, and must act in the most diverse economic activities.

These applications are responsible for bringing Letsfair to life. They connect all the points, or better yet, they interconnect the main functionalities that will be available, which are the six pillars: identity, reputation, consensus, digital ownership, finance and finally, negotiation.

Applications can be of the decentralized type, those that directly access the core of Letsfair, or can be of the web or mobile type, that provides access to the Letsfair through the API that will be available.

Letsfair's main application will work on the web, at, as well as on mobile versions for iPhone and Android. All key functionalities will be implemented in a generic way in a large marketplace. User identities will engage in behaviors such as social networking profiles, and they may create product and service stores of any nature. They will also be able to create and subscribe to channels of shared economy, to offer their goods and skills as they wish. Today, shared-economy systems are fragmented into niche markets, and in this application, we want to address a significant number of them, while respecting their peculiarities.

The differentials compared to other marketplaces on the Internet are Letsfair's own features:

Identity, reputation and financial data shall belong to the user himself, ensuring a high level of privacy; Users may issue certificates of ownerships related to their products and services; Several identities may become associated to offer unified products and services, as described in this document, in the topic on Consensus; The most importantly is that Letsfair will be the payment system itself, and will not have to charge software-as-a-service (SaaS) sales commissions or fees. The infrastructure will be maintained only with the small fee charged under the transactions, replacing the high fees already charged by payment processors today.

Those who are interested will be able to develop their own trading applications, specific to economic activities that require a more elaborate workflow, or that need additional technological resources. But for this they will be able to count on the basic economic infrastructure functionalities arranged in Letsfair, as well as benefit from the entire network of users who are already engaged in the platform.

6. Conclusion

As presented in this document, Letsfair is a platform that aims to provide economic and technological infrastructure for businesses of any nature. It will enable simplified management of identity, reputation, consensus, digital ownership, finance and negotiation.

It will place identity control in the user's own hands, solving privacy problems and lack of interoperability of information between organizations, providing the enrichment of identities with more significant attributes, among them, reputation data.

As the creation of reputational capital lies at the heart of Letsfair, a multidimensional reputation data approach will be used to measure different dimensions of human characteristics, thus contributing to shared economics, profoundly altering the trust-building model between people.

Quality identities and relevant reputation data will tend to promote the decentralized organization of labor, allowing agility and security in the establishment of consensus among people, bringing with it a historical milestone in the evolution of work relations.

It will also allow the issue of financial assets and digital ownership certificates through cryptographic tokens. In addition to using cryptocurrencies for payments in exchange transactions, ensuring the flow of information necessary for the generation of reputation data.

Trading applications from a wide range of market segments will promote the professional relationship between users and become responsible for bringing Letsfair to life, by interconnecting the main features that will be available. They may be native to the platform or developed by third parties within the community.

With all these features, the use of Letsfair will promote the use of cryptocurrencies, bringing them closer to the public. The issue of cryptocurrency price volatility will be solved by using cryptographic assets with a stable value, which follow the fiats money index. It will allow users to index the prices of their products and services in the currency of their country, reducing a barrier of acceptance.

Moreover, it is not Letsfair's intention to create an alternative antigovernment market. The fiscal jurisprudence must be respected in every locality. In addition, platform capabilities will also be available to reduce bureaucratic barriers and automate government services.

Lastly, the Letsfair platform aims to be a link between many services on the Internet today, interconnecting them to the benefits provided by an intelligent Blockchain. And with this union, it is hoped to provide the emergence of types of businesses not yet conceived.

7. References

[1] TED Talks - Rachel Botsman: We've stopped trusting institutions and started trusting strangers. Available at: ( Last access: May, 17, 2017

[2] Alain Peyrefitte (1995) La société de confiance. Essai sur les origines et la nature du développement Broché.

[3] Covey, Stephen R. and Merrill, Rebecca R. (2006) The Speed of Trust: The One Thing That Changes Everything.

[4] Ethereum White Paper - A Next-Generation Smart Contract and Decentralized Application Platform. Available at: ( Last access: May, 17, 2017

[5] Bowe, Sean. zkSNARKs in Ethereum, 2016. Available at: ( Last access: May, 17, 2017

[6] An Update on Integrating Zcash on Ethereum (ZoE). Available at: ( Last access: May, 17, 2017

[7] Enterprise Ethereum Alliance. Available at: ( Last access: May, 17, 2017

[8] Consult Hyperion. Available at: ( Last access: May, 17, 2017

[9] Digital Identity: Issue Analysis Report (PRJ.1578). Consult Hyperion. Available at: ( Last access: May, 17, 2017

[10] Resnick et al. (2000) P. Resnick, R. Zeckhauser, E. Friedman e K. Kuwabara. Reputation Systems: Facilitating Trust in Internet Interactions. Communications of the ACM, 43(12).

[11] A. Jøsang and J. Golbeck, “Challenge for Robust Trust and Reputation Systems,” in Proc. of 5th International Workshop on Security and Trust Management, [2009] © Elsevier Science B.V, Sept. 2009.

[12] WSO2. (2011). EBay uses 100% Open Source WSO2 Enterprise Service Bus to Process more than 1 Billion Transactions per Day [online]. Available at: ( Last access: May, 17, 2017

[13] Ebay. (2015). Changes to Feedback – FAQ [online]. Available at: ( Last access: May, 17, 2017

[14] P. Dewan and P. Dasgupta, “Securing reputation data in peer-to-peer networks”, in Proc. of Parallel and Distributed Computing and Systems (PDCS 2004), Cambridge, MA, 2004.

[15] Jøsang(2010) A. Jøsang. Trust management in online communities. Workshop on New Forms of Collaborative Production and Innovation:Economic, Social, Legal and Technical Characteristics and Conditions-SOFI.

[16] Paul Resnick and Richard Zeckhauser. Trust among strangers in internet transactions: Empirical analysis of eBay' s reputation system, chapter 6, pp. 127-157.

[17] J. Douceur. “The Sybil Attack,” In IPTPS ’01 Revised Papers from the First International Workshop on Peer-to-Peer Systems, Cambridge, MA, 2002, pp. 251-260.

[18] H. Yu et al, “SybilLimit: A Near-Optimal Social Network Defense against Sybil Attacks,” in 2008 IEEE Symposium on Security and Privacy, [2008], © IEEE. doi: 10.1109/SP.2008.13

[19] B. Prêtre, “Attack on Peer-to-Peer Networks,” Semester Thesis, Dept. of Computer Science, Swiss Federal Institute of Technology (ETH) Zurich, 2005.

[20] Chu Yee Liau, Xuan Zhou, Stephane Bressan, and Kian-Lee Tan. Efficient distributed reputation scheme for peer-to-peer systems. In Web and Communication Technologies and Internet-Related Social Issues - HSI, pp. 54-63. Springer Berlin / Heidelberg, 2003.

[21] Li Xiong and Ling Liu. Peertrust: Supporting reputation-based trust for peer-to-peer electronic communities. IEEE Transactions on Knowledge and Data Engineering, pp. 843-857, 2004.

[22] Elli Androulaki, Seung Geol Choi, Steven M. Bellovin, and Tal Malkin. Reputation systems for anonymous networks. In Proceedings of the 8th International Symposium on Privacy Enhancing Technologies, PETS '08, pp. 202-218, Berlin, Heidelberg, 2008. Springer-Verlag.

[23] Elan Pavlov, JereyS. Rosenschein, and Zvi Topol. Supporting privacy in decentralized additive reputation systems. In Christian Jensen, Stefan Poslad, and Theo Dimitrakos, editors, Trust Management, volume 2995 of Lecture Notes in Computer Science, pp. 108-119. Springer Berlin Heidelberg, 2004.

[24] Davide Carboni. Feedback based reputation on top of the bitcoin blockchain. arXiv:1502.01504, 2015. Available at: ( Last access: May, 17, 2017

[25] Ethereum Wiki: Reputation systems. Available at: ( Last access: May, 17, 2017

[26] Brin e Page (1998) S. Brin e L. Page. The anatomy of a large-scale hypertextual web search engine. Computer Networks and ISDN Systems, pp. 107-117.

[27] L. Mui, M. Mohtashemi, C. Ang, P. Szolovits, and A. Halberstadt. Ratings in Distributed Systems: A Bayesian Approach. In Proceedings of the Workshop on Information Technologies and Systems (WITS), 2001.

[28] A. Withby, A. Jøsang, and J. Indulska. Filtering Out Unfair Ratings in Bayesian Reputation Systems. The Icfain Journal of Management Research, pp. 48-64, 2005.

[29] Quercia, Daniele; Hailes, Stephen; Capra, Licia. B-Trust: Bayesian Trust Framework for Pervasive Computing. Trust Management, 4th International Conference, iTrust 2006, Pisa, Italy, 2006.

[30] Paul Masurel, Of Bayesian average and star ratings. Available at: ( Last access: May, 17, 2017

Clone this wiki locally
You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.
Press h to open a hovercard with more details.