Snowflake is an Identity Protocol. We believe that by creating a modular, open framework in the decentralized identity space, we will help individuals craft their digital identities and enable projects of all flavors and varieties to easily and securely access this data in a decentralized manner. Snowflake is an ERC-1484 Provider that provides on-/off-chain identity management. For more details, see our whitepaper.
Hydro’s mission is to become the public ledger for financial applications, empowering a new decentralized global economy. This ecosystem consists of standard smart contracts that can be used by financial applications, connected through APIs. These standardized smart contracts and API connections reduce the need for firms to hire blockchain developers and allow them to add decentralization to existing systems.
A core component of the financial ecosystem is the notion of identity and identity management. It is vital to creating security for account opening, transactions, and interactions with end-users of digital applications. With the growth of digital commerce, identity management becomes even more crucial to the future economy.
In this paper, we propose a new paradigm for identity in financial services that leverages a public and decentralized blockchain, called Snowflake. Users own their Snowflake identities, they are immutable, and they can be validated. Each Snowflake identity is tied to cryptographic keys embedded in the user’s local devices, which are together connected to a unique ID on the blockchain. This identity can be shared with third-parties at the discretion of the user, and creates the groundwork for complex digital applications to recognize a user by a persistent identity.
This proposed identity platform can help to solve multiple problems:
- Paper-based identity management is currently expensive and prone to theft and hacks
- The dark web has increased the profitability for stolen identities
- There are many competing implementations of identity projects, none of which are natively interoperable
- E-commerce and digital banking has made it increasingly difficult for companies to confirm their users’ identities
- Emerging economies fail to utilize the growing use of mobile phones to their advantage in identity creation and management
Hydro’s Snowflake empowers construction of complex digital identities created under ERC 1484, a global standard for digital identity aggregation. This enables applications with identity implementations that are able to comply with private and sovereign KYC standards, while empowering billions to own and secure their personal identities. This has implications for instant account onboarding, global ID passports, and when combined with the Hydro Raindrop, secure authentication of end-users.
The Origins Of Identity
In our first Whitepaper, we examined the the need for proper authentication of users within a technology ecosystem. Yet, authentication is not sufficient if the authenticated entity does not share additional identification information. In this whitepaper, we examine the urgent need for a better identity management framework that can be applied across financial services. Before exploring how a public blockchain can provide the missing piece to this puzzle by making identity immutable and standardized, let us first examine a more fundamental question: what is identity and where does it come from?
It is believed by scientists that names, the most fundamental forms of identity, pre-date the written word, originating perhaps as soon as humans gained the ability to speak. There is even compelling evidence that other mammals, such as dolphins, establish “identities” using signature whistles as naming conventions for one creature versus another. As with humans, these identities help to forge tight communal relationships among mammals.
The idea of governments formally tracking names, via birth certificates, is a relatively modern one. For years, births and naming were a church function in most western societies. During the immigration wave in the U.S. of the 1800s, which brought 30 million immigrants, reformers pressured the U.S. government to formally track births and deaths in the census, but it wasn’t until World War II that standardized birth certificates establishing where you were born and what was your name became mandatory in the U.S.
Yet, formally establishing these basic pieces of information is not enough to establish a unique identity. From the earliest time, surnames were used to establish what was your occupation, or who your parents were. The surname Smith was applied to blacksmiths, Robertson was the “son of Robert,” and Bin Ali was the “son of Ali.” The commonality of surnames makes identity management incredibly challenging. According to Ancestry.com:
Most of the approximately 100,000 Japanese surnames in use today only date from 1868, when surnames were mandated for the first time, There are just a few hundred common Chinese surnames, and 20 of them are shared by half the population. There are about 250 Korean surnames, three of them comprising almost half the Korean population, and just about 100 Vietnamese ones, with three making up 60 percent of all names in that country.
In the United States, Smith is the most common surname in 40 of the 50 states. Because of the commonality of names worldwide, proper identity management necessitated governments to start national identification numbers.
National Identification Numbers
In the U.S. the Social Security number (SSN) was created in 1936 for the sole purpose of tracking the earnings histories of U.S. workers, for use in determining Social Security benefit entitlement (after the adoption of the Social Security Act in August 1935) and computing benefit levels. Names and addresses were considered, along with fingerprints, but names encountered many of the same concerns with commonality raised previously, and in the 1930’s fingerprinting was most associated in the United States with criminal activity.
There are nine digits in the U.S. SSN. The first three digits are assigned by geographic region or zip code, and the middle two numbers are group numbers that further identify the geography, while the last four numbers are random serial numbers. Thus, it is common for businesses to confirm only the last 4-digits of the number for identification purposes, making these numbers extremely prone to theft.
Likewise, many other countries have similar national identification numbers, sometimes called Social Insurance Number, Tax Identification Numbers, or National Identification Numbers.
Black Market For Identity
Because identities are not immutable, and many governments and private actors in the ecosystem cannot be trusted to create or confirm identity records, there is a thriving black market for fake and forged documents. Take for example, a Cuban birth certificate. To an illegal immigrant in the U.S. having a Cuban birth certificate could mean a path to citizenship in only 1 year, so those forged birth certificates have been sold for $10,000 or more.
It has also been estimated that up to 40% of the passport fraud in the United States involves counterfeit or stolen birth certificates from Puerto Rico, and Security Alliance reports that in 2008, 45,622 children were born in Puerto Rico, but in that same year 860,000 certified copies of birth certificates were issued! The current government identity system is severely broken, it incentivizes bad actors both in the private and public sector, but also creates a black market that incentivizes impoverished populations to sell their identities for extra income.
With this in mind, how can we expect to have a fair and honest financial ecosystem? Banks, credit card companies, and even startups struggle to properly identify users, and even when they do, is there ever a way to truly know if the person is who they say they are?
Identity Theft & E-Commerce
As examined in our Raindrop Whitepaper, identity theft is a growing problem in the U.S. and worldwide. In April 2017, Symantec published its Internet Security Threat Report, which estimates 1.1 billion pieces of PII (personally identifiable information) were compromised in various capacities over the course of 2016. Commerce is increasingly done remotely, and this lack of personal contact has made financial fraud easier to perpetrate, and harder for authorities to stop. There is no identification required for the vast majority of online transactions. It is based on a system of “trust” and that there will be a certain percentage of chargebacks and fraud cases that are written off. The average cost of chargebacks alone is approximately 1.47% of a merchant’s yearly revenue, and $118 billion in lost sales occur during false positives – when a merchant stops a legitimate customer from making a purchases because they falsely identified them as a fraudster. These are staggering figures that can easily be solved with immutable IDs tied to the public blockchain that are used during credit card and e-commerce transactions.
Current Forms of Identity
The overall concept of identity is changing as we approach the Web 3.0 and will continue to evolve over time with the advent of more artificial and sentient intelligence in global society.
Identities currently take two common forms – private and public.
- Private identities are things like telephone numbers, and email addresses.
- Public identities are things like tax identification numbers, passport numbers, and other forms of identification that are formed within civil societies.
Increasingly, research has shown that private forms of identities have begun to tell us more about ourselves, and produce more emotional attachment than public forms. Many have argued that a cell phone number has now replaced a Social Security Number in the United States in importance. According to Statista, global cell phone users are expected to reach 5 Billion in 2019, with over 50% using smartphones.
An example of the longevity and utility of cell phone numbers can be found in the U.S. mobile to mobile porting stats. In 2003, the first year data was collected, only 795,000 numbers were in the U.S. mobile porting database, by 2009 that number ballooned to over 40 Million we would now estimate the number to be in the hundreds of millions but no public data exists. There are now governmental initiatives to make the practice commonplace because lawmakers view phone numbers as “important identifiers” and have said “consumers overwhelmingly prefer to keep their numbers. Psychological studies have shown high levels of distress and anxiety when people are without their cell phones, and they now possess similar attachment qualities as teddy bears for youth.
The idea of what is your identity is changing, with technology more and more intertwined into one’s sense of self. When creating an identity protocol, it is therefore important to not only include public identity, but also private identity in the equation.
There are increasing problems with document portability across county, state, provincial, or country lines. Not all identity documents are created equal, and the lack of a global standard is one of the chief concerns of those fighting terrorism, money laundering, drug trafficking, and other illegal activities. Take the U.S. as an example, the problem has gotten so bad that the U.S. government has made it mandatory that residents of Kentucky, Maine, Minnesota, Missouri, Montana, Oklahoma, Pennsylvania, South Carolina, and Washington use a U.S. passport for all domestic travel, because these states do not currently issue a state ID that meets the standards of the REAL ID Act of 2005.
This is not just a problem with centralized identification protocols. Because of a desire of entrepreneurs to create proprietary solutions, rather than open architecture ones, many identity protocols that have or will be proposed on top of distributed databases fall victim to the same problems. It is the view of the Hydrogen developers that tokens or platforms must have the flexibility to read, write, extend, connect, and interact with an identity protocol for it to be most effective. In other words, just adding a public blockchain into the equation does not change fundamental flaws in the approach. Putting a proprietary solution at the protocol level for identity management is similar to the 9 U.S. states that do not meet U.S. identity requirements. There are painstaking changes that will need to be made to create compatibility. This is why Hydrogen is fully committed to a simple and standardized protocol.
Government Blockchain Initiatives
Some of the issues listed above have had the obvious consequences in the public sector, greater interest in blockchain technology. For example, a task force in Illinois presented a plan to create tax records, voting and health histories on a state-run blockchain. The government would become the verifier, rather than the custodian, of people’s public service identity, moving from providing data storage to verifying identity. Estonian citizens and e-residents are issued a cryptographically secure digital ID card powered by blockchain infrastructure on the backend, allowing access to various public services, and other countries exploring blockchain ID initiatives include Singapore, Georgia, the UK, and Dubai.
The problem with this is twofold: 1. Governments acting as sole verifiers creates centralization and corruptibility; 2. If every local, state, and national government creates unique private or semi-private blockchains, they will just compound the problems of their non-distributed predecessor databases. For identity management to be truly decentralized and fungible, there must be multiple verifiers and standardized protocols.
Governments may falsely see blockchain “the technology” as a panacea, ignoring the fact that migrating current database infrastructure to a more distributed service is at best only marginally better for the end citizen. At worse, it may make their lives more difficult and complex. Putting false information in an immutable database might be worse than not having any information in there in the first place. Creating systems that are closed architecture and not standardized may compound the current issues seen with global passports and border security.
To solve the problems mentioned thus far, the team introduces the Hydro Snowflake – unique identity powered by the public blockchain. Why call it Snowflake? Snowflakes are one of the most beautiful, random, and individualistic phenomenons in nature. Not only do scientists believe they come in 35 different types, there are differences in the atomic structure of the atoms making up the water molecules, making each Snowflake unique.
Snowflake builds an extensible identity following the ERC 1484 digital identity standard. Each end-user of a snowflake-powered application is minted a globally unique Ethereum Identification Number which connects to all of their owned Ethereum addresses and ties the user to arbitrarily formatted data. Snowflake handles the complexities of identity management – trustlessly connecting to new user devices, allowing applications to pay gas costs on behalf of the user, and allowing a user to interact with multiple applications and instances of identity with a common user experience. Snowflake identities must be “minted.” Each minted Snowflake acts as a base identity layer that can be interacted with and extended.
ERC 1484 Global Identity Registry in Snowflake
Every application, be it web or mobile, consumes some form of identity information. As a result, users are stuck either repeatedly providing the same data to multiple applications, or relying on central parties to relay said information. If those central parties mishandle user data, users are inevitably vulnerable to identity theft. Further, apps must deal with verifying that the provided user information is valid in order to build meaningful user profiles.
To address these shortcomings, each Snowflake user has a uniquely owned space in a global identity registry that is mapped to a unique HydroID and associated with the seed created through Hydro Client-Side Raindrop on a mobile device. The HydroID provides a human-readable interface for a user to connect to applications built within the Hydro ecosystem while remaining ERC 1484-compliant through an EIN on the global registry.
This structure supports storage of any arbitrary form of data, stored in external smart contracts called “Resolvers.” We call contracts structured to encode data that can be associated with a Snowflake a Resolver as a nod to the analogous concept in services such as DNS and ENS, where external reference points are able to eliminate abstraction by encoding arbitrary data structures that point to, or “resolve to” an atomic entity – in this case, the Snowflake. The architecture of storing information in external contracts that tie back to a singular Snowflake enables unlimited scope for how a user defines their identity, rather than restricting an identity to a pre-defined set of standards. An intuitive comparison would be a device that is able to encode passports, driver’s licence, social media accounts, medical records, and other identity-defining data all in a singular source that can be queried in a standard way by any system. In this way, any dApp with which a user interacts can easily become a resolver, maintaining its core functionality while streamlining the mechanisms by which a user can interact with it.
Because Snowflake only acts as a singular point for a user to manage their data and places no restrictions on the encoding of the data itself, it offers complete flexibility for the implementation of keeping the data within a resolver public or privatizing it, per the needs of a given application.
Multiple Address Ownership
Native to Snowflake is the capability for users to irrefutably claim multiple Ethereum addresses through cryptographic signatures. In the absence of Snowflake, when users want to interact with dApps on Ethereum, they must either transfer their private key to a wallet on the new device or create a new, entirely unassociated address. Fundamentally, any transfer of a private key away from a local device exposes a user’s Ether, tokens, and data tied to that address to phishing and malware. Even careful users are likely to fall victim to systems that structurally require the transfer of private keys across devices. Alternatively, creating new, unassociated keypairs doesn’t offer much value to users, as all of their data is stratified among addresses on a separate device. Snowflake allows users to provably tie multiple addresses back to their core Snowflake, extending any Resolver data across their devices without requiring any transfer of private keys. Procedurally, when linking a new address to their Snowflake, the owner address initiates a signed claim
sign(EIN, addressToClaim, timestamp). Subsequently, the address to be claimed sends a transaction with the
approvingAddress. The initial claim is timestamped in order to prevent a malicious observer from recognizing which address an owner is trying to claim prior to the completion of the claim.
The concept of identity is relatively meaningless without validation. In general, validators are third parties who can vouch for the validity of data tied to an identity. Since Snowflake empowers users to self-associate with on-chain identity information, it seems apparent that Snowflake only enables for self-attested data, which is relatively limited in scope. However, because the nature of a user’s self-attested data is arbitrary, Snowflake allows for norms to drive data attestation in fundamentally decentralized fashion. The logic goes as follows: Because entities observing a user’s self-attested data will not derive value from this data without third-party attestations, users will find value in connecting with data structures that enable third-party attestation. Users are able to choose which third-party attestations to pursue, which is closely tied to the ability for entities to choose which third-party attestations they accept. In this way, Snowflake enables a free-flowing ecosystem of third-party attestations under any set of standards adopted by corresponding parties, rather than a pre-defined attestation structure native to the identity protocol.
In practice, validation structures will likely develop whereby off-chain validators can act as parties trusted by individual actors in the decentralized ecosystem. A hypothetical privacy-preserving implementation is as follows: Once the validation happens off-chain, an on-chain merkle hash would be tied to a user’s Snowflake. Because nobody can impersonate a validator without access to their private keys, business entities can incorporate on-chain validations into any relevant business logic by querying a user’s Snowflake. If a user changes any of the data, it would lose its prior validations along with any corresponding significance due to the deviance from the on-chain record. Various implementations of similar validation structures can develop, each catered to their particular use-cases (KYC, reputation, identity implementations, etc) while all resolving to a user’s Snowflake.
In itself, Snowflake is a compliant
provider within the ERC 1484 framework. As a provider, Snowflake is unique for a few reasons:
- It natively supports metatransactions. This means applications built using Snowflake are able to act as service providers, abstracting gas costs away from end-users. This means a user’s cryptographic keys can be generated within an application, stored in the user’s device memory, and tied to a persistent user-identity without the end-user needing to go through the complex onboarding common with many dApps today.
- It establishes an optional human-readable HydroID which is tied one-to-one with a user’s EIN. This allows applications to recognize users by a persistent username, allowing implementations like the Hydro Mobile App by Hydrogen to empower users to interact with applications in a consistent approach within the Hydro ecosystem.
- It provides complex plug-and-play payments functionality for dApps out-of-the-box. This will be covered in further detail in (HYDRO Tokens In Snowflake)[Hydro-Tokens-In-Snowflake].
The Role of Snowflake Metadata Within an Ecosystem of Businesses and dApps
While immutable and standard data native to an identity built on top of the Snowflake protocol help a user establish a secure and global standard for core identity information, Snowflake itself allows users to tie an unlimited range of metadata to their identities by setting third-party dApps as resolvers. As users’ interactions with dApps helps build their on-chain identity, they may set a dApp as a resolver for their Snowflakes, allowing businesses querying the associated metadata to incorporate it into the logic of their own applications.
Fundamentally, a third-party or dApp acts as a certain attribute on a user’s Snowflake. An application prescribes meaning to a user based on associated data with their snowflake. Let us more closely examine these meanings in the context of a few examples.
What does a Resolver mean from the perspective of the user? Resolvers establish a user’s core data. Acquiring validations from reputable sources enhances the integrity of a user’s Resolver data which can be relied on by businesses, governments, or decentralized applications querying any relevant identity information.
What does a Resolver mean from the perspective of an app? Since resolvers encode data about a user, a business can programmatically incorporate this data to drive business logic on its platform to improve user experience on its platform or offer users certain permissions. Requirements for third-party validations of this data could be encoded as part of the business logic. If an app were to recognize a given aggregated identity standard processed through Snowflake, it could prescribe whatever meaning to Resolver data best suits its needs.
- Example 1, KYC validation: a government could (off-chain) allow users to register their Snowflakes through a Resolver for a voting validation. The user could then cast their vote through a third-party voting dApp, which would only tally votes from registered Snowflakes. This process would ensure that each person can only vote once, eliminating voter fraud or Sybil attacks, while providing complete transparency and much greater efficiency to elections. This can apply to any arbitrary set of digital voting structures and does not enforce a specific implementation.
- Example 2, Social Media: to prevent the creation of fake accounts, an app could implement a system that links a user’s Snowflake to an account on the app. Any accounts that have their name validated by at least X existing users of the app could display a ‘verified’ checkmark on the UI of the app. Unlike the first example, this example factors validations from individuals instead of from large institutions. Also, this example observes validation for the HydroID associated with the Snowflake itself rather than a resolver. The openness of the Snowflake identity framework allows for a wide range of use-cases
Fundamentally, both applications identify users taking an entirely different approach, but they remain globally interoperable from a user’s perspective through a standard user-owned Ethereum Identification Number, created in the ERC 1484 identity registry, with interactions between the user and the application handled through the Snowflake Smart Contract.
What does being a Resolver mean from the perspective of the Resolver? Resolvers are able to leverage the native functionality of users’ Snowflakes. This offers two key points of value:
- Rather than existing as an independent dApp with which the user interacts, they exist as dApps within the ecosystem of Resolvers associated with a given Snowflake. This means a user can seamlessly manage their associated data from a dashboard, and they are not burdened with the struggle of aggregating data from multiple Ethereum addresses and various dApps whenever they want to leverage it.
- Resolvers are able to easily leverage token transfer functionality native to users’ Snowflakes. This streamlines the development process for dApps aiming to monetize without wanting to create sophisticated payment structures within their apps. Users are able to manage Hydro withdrawal balances for all of their resolvers from a single dashboard.
HYDRO Tokens In Snowflake
As Snowflake provides out-of-the-box solutions for complex dApp development, one component of this framework is payments.
The HYDRO token is not only intended to be a gateway into the ecosystem of Resolvers built on top of Snowflake, but also a centerpiece for convenient programmatic token transfers between users and resolvers. Through Snowflake, users can set specific allowances for resolvers to withdraw HYDRO. The user sets the limits on allowed withdrawals on a per-resolver basis, streamlining replication of user-facing models that are currently managed through third-party financial institutions. Snowflake’s flexibility with programmatic token transfers allows businesses to encode an arbitrary set of criteria into their business logic for recurrent processes, such as verifying membership within a particular group in order to offer discounted subscriptions for a product.
More specifically, Snowflake provides a powerful set of tools for dApp developers to easily monetize their products. Currently, in the Snowflake phase, we have created a structure for direct payment gateways between users and dApps. This is structured to provide an intuitive and secure model for user-dApp interaction through the opening and closing of gateways called
allowances. The allowance-based structure has a few advantages over direct token-transfer functionality.
- Allowances allow dApps to programmatically withdraw user funds as certain events are triggered instead of requiring user-initiated transactions every time. This significantly simplifies the user experience.
- Allowances let dApps escrow user funds on a user-driven basis. This prevents dApp/user-flow breakage whenever a user doesn’t have the necessary balance to call a function on a dApp through a front-end.
- Allowances are handled on a per-identity basis rather than a per-address basis.
- Allowances are globally useful within a dApp. This means that rather than each function independently handling balance throw errors, global logic checks for allowance-handling can be put in place by a dApp.
What does all this mean for a user? A cleaner user-experience, more consistent with traditional payments rather than cryptocurrency-based payments.
Beyond opening simple user-dApp payments gateways, Snowflake allows dApps to open user-contract-dApp payments gateways through what we call Via Contracts. Via Contracts allow token transfers to undergo arbitrary logic before arriving at their end destination. This gives dApps an easy and consistent way to encode transactions that would otherwise be incredibly complicated and nonstandard to implement for simple applications. Examples include:
- Hot-swapping; dApps are able to allow their users to interface cleanly with the HYDRO token while accepting or driving logic within their application within their own native token or a token of their choice. This lets them have complex tokens as-needed within their applications without disrupting the user-experience.
- Routing transfers through a side-chain. This allows developers to leverage complex scaling solutions within their application without having to build out custom integrations.
- Applying payments logic to a transaction. This empowers transactions that can do things like accept subscription payments, accrue interest for a loan, check for completion of an event before transferring funds to an end-destination, escrowing funds temporarily for reversible transactions, or more – alll implemented at the discretion of the developer. These functionalities are taken for granted in most financial applications today.
The Tide phase of Project Hydro will focus on building globally useful Via Contracts; however, the infrastructure exists in Snowflake for this to be done today.
Apps, dApps, products, or platforms built on top of Snowflake can also incorporate HYDRO tokens into their processes. For example, certain kinds of validations or actions may require on-chain HYDRO token transactions, where users would be required to maintain or transfer HYDRO balances, such as an implementation of our Server-side Raindrop Authentication Protocol.
The Snowflake Dashboard and DApp Store
As Resolvers built on top of Snowflake need user permission to tie into a user’s Snowflake Identity Token, the Snowflake dApp store provides an easy means for users to manage their associated resolvers and interact with them. The dApp store allows developers to build arbitrarily complex front-ends for their Resolvers using React, and upload these front-ends much like app developers upload their apps to the Google play or iOS store. The dApp store will ultimately be hosted on IPFS using an ens domain. While the react front-end requires pull requests for new dApps to be added to the store, any forked implementation can exist and remain accessible to users in tandem with the implementation created by the Hydro team. We envision the dApp store as an easy gateway for users to interact with dApps, and, accordingly, for dApp developers to monetize their dApps using Hydro.
The dApp store is broken into four sections: address management, resolver management, balance management, and the dApp store.
- Address Management allows users to link multiple addresses to their Snowflake Identity. This allows them to access a common dashboard experience regardless of which device or dApp browser they are using.
- The dApp store consists of all possible dApps built on Snowflake that have been added to the store by developers.
- Resolver Management allows users to interact with dApps they have linked from the dApp store and manage their allowances for each dApp.
- Balance Management allows users to acquire HYDRO tokens directly deposited into their Snowflake Identities, and withdraw as needed.
It is important to note that the proposed framework is an open protocol for identity management. Unlike other blockchain products, there will be no centralized decision on the strength or authenticity of data associations, attestations validations, or those who provide them. It will be up to the global community to identify and punish bad actors. Later in this paper, we examine potential apps, dApps, and platforms that can be built on top of or integrated into Snowflake to increase the effectiveness of the ecosystem.
Snowflake: Technical Details
There are four main entities generally important to Snowflake implementations: users, validators, resolvers, and business entities.
Users mint ERC 1484 EINs through Snowflake to represent their Snowflake Identities. They attach data to their Snowflakes and set resolvers in order to tie any form of metadata to their base Snowflake identities. Users can also maintain balances of HYDRO within Snowflakes, creating an easy and intuitive mechanism by which any dApp can interact with a user’s Snowflake. User data can take any arbitrary format and is associated with Smart Contracts called Resolvers on the Ethereum blockchain.
While validators attach validations to a Snowflake, resolvers are set by users themselves. Resolvers are dApps that contain identifying data about a user. A simple, but intuitive example is CryptoKitties – setting CryptoKitties as a resolver can tie the ownership of certain kitties back to the user’s Snowflake identity, even if the Kitties are owned by various Ethereum addresses. dApps that then want to integrate a user’s owned Kitties into their experience are able to reliably draw on and regenerate the Kitties.
While validators are not included in the core Snowflake protocol, they are worth mentioning, as they may add significant value when introduced at the dApp level. Resolvers built on Snowflake can encode any range of validation criteria. The simplest level would be a validation dApp in which a trusted KYC provider affirms whether a Snowflake is owned by a real person or not. Any Snowflake owner could complete KYC through the trusted party, set the validation dApp as a Resolver for their Snowflake, and prove that they exist to anybody else who also trusts the KYC party without ever meeting them. In practice, dApps can build nonbinary validation structures with a much more far-reaching range of implications in a variety of industries, with meaning derived from off-chain reputation, or their own native on-chain reputation protocols.
Snowflake Smart Contracts
The ERC 1484 Identity Registry stores, addresses, resolvers, and providers for a given identity. The Snowflake smart contract acts as a complex provider built on this registry. It allows users to mint identities, deposit HYDRO, set resolvers, and prove ownership of multiple Ethereum addresses. Users, resolvers, or any interested party are able to deposit HYDRO tokens to the Snowflake smart contract and interact with them. This facilitates payments, and other token functionality that will be integrated seamlessly into the Hydro ecosystem as it develops.
The Base snowflake identity consists of the summation of:
- Multiple Ethereum addresses claimed by the user
- Resolvers – set by Snowflake owners through their Resolver dashboard
- Token allowances and transfer permissions set for each Resolver
Snowflake API and Hydro Mobile App
Hydrogen’s Hydro API acts as a reference implementation for what can be built on top of Snowflake. It will facilitate user and business entity interactions with Snowflake smart contracts. The API provides end users the ability to register their HydroIDs, a prerequisite to minting their Snowflakes, through the Raindrop API integration into the Hydro mobile app. In the next iteration, users will be able to mint their own Snowflake associated with their HydroID through the Hydro mobile app while the Hydro API handles the direct interaction with the blockchain. They will also be able to claim ownership of external addresses through their mobile app. This streamlines their ability to transact with Snowflake dApps without compromising the security of the wallet stored locally in the memory of the Hydro app on their phones. They will be able to use these owned addresses to attach a broader range of metadata to their snowflake and interact with Resolvers who have implemented Hydro-based payment structures.
Implications for Financial Services
The open framework of Snowflake will make products, platforms, apps, and dApps built on top of the protocol integral to its long-term success. Below we examine some initial applications that the Hydro community will champion:
Universal Account Onboarding
Through integration with global KYC providers as validators, Hydro can create a standard one-touch digital account onboarding standard that can adapt to the wide range of KYC standards existent in today’s world.
Validator Rating System
This simple dApp would create an upvoting/downvoting system for validators across multiple identity areas. A reputation-based scoring system would create incentive-driven honesty in votes. This decentralized reputation-building can prescribe greater meaning to some validators than a binary process.
Document & Contract Verification Standard
The next phase of Project Hydro is called Ice. This is a document management protocol. The Snowflake, combined with Raindrop, will be integral for the scaling of Ice. A large problem with e-signing technologies is there is never any authentication or verification done, other than single factor usernames and passwords. Validated snowflakes and multi-factor authentications for any document signings or contract sealing can be built into the Hydro mobile app. Some simple dApps built on top of the Ice protocol can also integrate popular e-signing softwares such as DocuSign and Adobe Sign.
Payment Verification Standard
An upcoming phase of Project Hydro is called Tide. This is a payment and data privacy protocol. There is great opportunity to create a dApp that can reference validations and create “Levels” or strength of the validation. An implication of this might be a plug-in for financial services websites that checks the level of validation. A highest level validation may require no extra verification or authentication, while the lowest validation may require a secondary form of ID.
Anti-Fraud POS dApps
Most point of sales systems require simple swiping or chip verifications on debit and credit cards. There are no additional identity checks done during the sale, only post-sale, as examined in this paper. As firms like Square, Paypal, Braintree, and Stripe increase their presences globally, billions will have access to instant point of sale transaction protocols. There are very interesting dApps that can be built that integrate the Snowflake, Raindrop, Tide, and Mist phases of Hydro to produce a decentralized point-of-sale network.
Global Credit Scoring
Access to credit is a huge global problem, and one of the major issues affecting the 2 Billion unbanked. To counter this problem, there has been a FICO scoring system setup in the U.S. to attempt to use financial data to validate credit worthiness. But this system is fundamentally flawed in that it takes in no “social proof” and only accounts for those already in the financial system with bank accounts and credit cards. The Hydro API can provide business entities with the information they need to build more robust credit scoring models without relying on central parties to store sensitive user data.
A third-party business entity can rely on information tied to a user’s snowflake through a resolver for that user’s credit scoring data in order to drive its business decision in delivering a loan.
The Snowflake ecosystem is dependent on off-chain validators as best actors, and a global decentralized community that would punish bad behavior. For example, we do not want to encourage the same “black markets” for verification and documentation that exist off-chain and bring them on-chain, so it is imperative that the ecosystem function as designed. This is why the HYDRO token and staking are so vital to keeping the ecosystem honest.
There is additional risk of “fragmentation” within the ecosystem. It will be up to a business or website to decide how much validation or attestation is acceptable to approve a user. Because Snowflake is a protocol, and not a product, it will be imperative that best use cases are standardized quickly to prevent this fragmentation of acceptance. For example, a validated state driver’s license may be accepted on Amazon for checkout, but on Google there might be an additional social attestation required. It is the Hydro team’s goal to provide best practices, sample apps, and widgets for sites to use for various use cases to combat this issue.
There is an “educational” risk to the ecosystem as well. Many incumbent blockchain identity ecosystems are built in a closed architecture or private way. There has been a huge disinformation campaign from private blockchain providers on the perceived weaknesses in public blockchain. The Snowflake ecosystem is dependent on users trusting the encryption on the public blockchain and their smartphone.
Identity management globally is broken and it is only getting worse as we approach the dawn of the Web 3.0. Identity remains centralized, prone to corruption and theft, and billions are still shut out from financial services solutions in part because of it.
The Hydro Snowflake framework is being implemented to solve key parts of the identity management problem:
- Creation of a standard decentralized framework to store personal information that can be encrypted on-chain
- Creation of an off-chain validation and attestation network that links to on-chain data
- Creation of standardized apps, QR codes, UI widgets, and other products on top of the Snowflake protocol that can be easily integrated
The Hydro team believes the framework set forth can be the standard identity management protocol of the Web 3.0. Due to the ease of use of the Hydro API both apps and dApps can quickly implement the Snowflake architecture into their own.