“Who I am” is any self-conscious person’s self-consciousness about himself, and in this self-consciousness, he becomes the object of his own thinking. Identity is a comprehensive and complex concept that psychologically constitutes the traits, beliefs, personality, appearance and expression of a person or group. How to express the concepts of on-chain identity, digital identity, and online identity, and what kind of identity management system do we need in a decentralized narrative? How does decentralized identity (DID) outperform existing web2 identity systems?
We have survived in the silos of the web2 world for too long, with privacy leaks, information abuse, and algorithm exploitation. This article attempts to explore how we can build a new “I own my data” identity system.
The more one writes about “identity,” the more the word becomes an unfathomable term because it’s so ubiquitous. —Erik Erikson
As the psychologist Erik Erikson, who coined the term “identity crisis,” I often feel it’s true: identity is a vague concept with many connotations whose meanings are highly dependent For context, this is no exception in Web3.
In this post, I will try to solve this problem: build a framework – a framework that treats identities in the web as a tool primarily used to store, manage and retrieve information.
It won’t clarify all uses and misuses of the term, but I hope it helps to think clearly about how identity shapes Web3, how applications are built for it, and what those choices mean for our web experience.
What’s in the name? Three Meanings of “Identity”
When people talk about identity, they usually mean one of three related but disparate scopes: a) a unique identifier, b) a holistic view of an entity, or c) a specific context about an entity.
Unique identifiers are critical in any social setting. Among friends, family members or small tribes (below Dunbar’s threshold of 150 people) that can be assumed to be familiar, a name is a sufficient “identifier”. Among other things, stricter identifiers help make participants “clearly visible” in the wider system. States implement ID cards to manage taxes, conscription, and social programs. The web application has user IDs in the user table for tracking, managing, and serving its customers.
The holistic view refers to all possible information about the user or other participants. Attempting to attach large amounts of data to unique identifiers can create a rich set of information about a person or entity. The pursuit of this can be seen in Facebook and Google’s user databases, India’s Aadhaar and China’s social credit system, and customer data platforms like Segment and LiveRamp.
A particular context can be represented as any of many subsets of the overall view. KYC or identity verification – a multi-billion dollar industry – is about verifying that someone is the only verifiable identity they claim to be within a national system. Likewise, authentication, anti-fraud, anti-spam, and credit algorithms are specific services that focus on subsets of information in the overall view.
Contradictory? Well, I am contradicting myself, then I am strong and I include many people. –Walt Whitman
Identity: appending information to an identifier
Unique identifiers are necessary, but useless by themselves. They are almost always used to jump to some information. This could be names and addresses in state records, documents in filesystems, passwords in application databases, or token balances or transaction history on the blockchain. In any case, an identifier is useful because it conveys relevant information.
Many situations require the retrieval or validation of a specific context linked to an identifier. For example, Gitcoin needs an “identity system” to prevent outside attacks on its Grants platform. In practice, they need to map proof of personality (KYC verification, twitter account) to unique identifiers. The more information they have about the possibility of that individual being unique or fraudulent, the better their platform can work.
The holistic view of identity is always incomplete – just as we can never perfectly describe our “true self” in space, our digital selves will never be fully consistent or comprehensive. But the more data we collect around a single (or set of linked) identifiers, the more information we can use for any given context.
The common denominator is: Identity systems create the ability to reliably associate information with unique identifiers.The more reliable the identity system, the more useful it is and:
- More Reliable: Available, Fault Tolerant, Tamper Resistant
- More flexible: can handle more types of information
- Easier to use: can be used in more contexts, unifying rather than dispersing information
Different environments require different privacy and security considerations; for example, whether to gain trust through third-party auditability or decentralization; more emphasis on consistency or availability. But at the most primitive level, an identity system becomes more powerful if it has the potential to make more information clearer and more consistent.
Digital identity as the key to the web
In the simplest terms, the network runs on hardware, code, and data. Every website you visit has logic and rules written in code, and almost all websites are populated with information encoded in data. This data—whether it’s today’s news, your friend’s tweet, or your most recent email draft—must be retrieved accurately and reliably when you arrive at the site. This is done via identifiers.
Just as a unique identifier is useless without additional data, it’s not very useful if the data on the network cannot be retrieved at the right time. Unique identifiers, and the routing tables and logic built around them, are used to organize the data that populates it on the network. Who is creating these identifiers? Who is organizing data around them?
Today, it’s almost every site you visit, every product you use, or company you encounter. The identifiers are listed in the database they create, which is mostly private and isolated from every other company’s database. The data is placed there and linked there as well. This is usually organized around a user table: each row represents a user, each column represents a type of data, and the table stores or points to each user’s record for that type of data.
Traditional user table on application database
Does this identity system meet our criteria above?
- Reliable: very reliable, but not auditable, highly vulnerable to hacking and bugs
- Flexible: database types can be chained to handle a variety of information, although it can be a bit confusing
- Usable: each application needs its own identifier, information (and its management) is very fragmented, redundant and inefficient
From a macro perspective, this is a really bad identity system for the web – because it’s not one identity system, it’s many different identity systems. It fragmented information, limiting its value and use to every participant. (It also creates a dire incentive for hoarding and misusing user data beyond the scope of this article).
On a more microscopic level, in the user’s experience with any given application, it is the application that is responsible for the user’s identity – their unique identifier, the data associated with it, and the reliable link between them. This happens only because we don’t have any other options at the moment. This is intuitively wrong.
Decentralized Identity: How Web3 Surpasses Web2
Blockchain is a form of distributed ledger technology (DLT), which is basically a shared database. A shared database seems like a good place to put a unified user table and get rid of the antiquated need for each application to create its own identity system.
This is the vision of the future of decentralized identity and a core pillar of the Web3 vision: every user and builder can control their own data, values, relationships, and information. In this vision, each user becomes a unified point of discovery for their own data, creating reuse and composability between applications. This creates shared network effects, interoperability, and composite experiences that isolated, centralized applications cannot compete with.
The original vision envisioned a unified user registry (on one DLT) and a standard way for all applications to add information to that registry. Users can control their own encrypted sovereign address (or identifier) with which they sign all data to create the trust needed for data in an open environment. We let every application use the same registry (blockchain) and publish data using a standard format ( NFT ), and in theory we are in identity nirvana – a network that brings social graphs to applications, independent of audiences and communities Sew interactive platforms and easily move between new products and services as soon as they become available because they are all interoperable.
However, this vision of decentralized identity, which relies on addresses and NFTs, quickly collapsed in practice. It is too rigid to manage and route to large-scale data well as an identity system. On our standards:
- Reliable: Today’s blockchains, designed for consensus on scarce financial assets, cannot scale to meet the scale of large amounts of data; nor can they handle off-chain (or partial) updates
- Flexible: Most on-chain ledgers support new data structures and standards, but within the constraints of the consensus system. This limits the use cases and applications of this system
- Accessibility: A single registry restricts users and applications to a single DLT or blockchain, and we inevitably use different chains and networks
We can learn from the flaws of the original cryptographic identity system to understand what is needed for a more reliable decentralized identity system. Obviously, a single registry (index), identifier standard or data structure standard is always too rigid.
It must be used with various identifiers. It must be open to a flexible, extensible set of data models and structures. It has to work across network environments and networks. It should be designed following the principle that identity is about managing and discovering information, so it should put data first.
How Web3 will handle the user table
In order to manage data, we need a protocol to store, discover, and route information about identifiers. For Web3 to live up to its promise, this routing table should a) be uniform, not isolated by application or any other boundary, and b) be sovereign, granting control of data directly to each identifier.
This suggests a simple design: each identifier maintains a table with its own data. Unified, these identity-centric user tables form the Internet’s distributed user table. This distributed user table is not an actual table, but a virtual table, which is produced by several components corresponding to the parts of the traditional user table:
- Identifiers: Decentralized identifiers should not be entries in the application database, but should be provably unique and cryptographically controlled. Accessibility requires acceptance of multiple forms of identifiers across various networks – similar to the DID standard for decentralized identifiers.
- Data Structures: Similar to how application developers define their own data structures, a decentralized data layer needs to enable developers to define custom data models while ensuring that these models are reusable and publicly stored.
- Index: When the application defines the data model, the user brings their identifier. A standard index can combine these elements into a user table (or application table) so that when a user interacts with an application (creates data), that information is appropriately cataloged for future routing. This creates an easily discoverable user data record – mapped to the data model and cryptographically linked to the identifier.
A distributed virtual user table with various DIDs from different networks, a developer-defined data model, and a record associated with each
For this distributed user table, from @Ceramic blog: “Each user has complete autonomy over their rows and can bring this data to every application they access. If an application wants to know which data Available and how to use it, they can reference data models that contain names, descriptions, and other metadata.”
Build with distributed user table
How does this identity system based on DID and data model and distributed user table meet our criteria?
Reliable: Operates on a collection of public networks that anyone can participate in, including partitioned or local networks
Flexible: Works with any data structure a developer can define
Availability: Works with any open network and unique identifiers
The system also has many additional properties that make up a highly flexible and reliable identity system. include:
Pseudonym First: No account creation or verification required to get started, the user (or other entity) simply carries an encryption key pair and can start accumulating information around it
Generable: Information accumulates over time, creating an emerging holistic identity
Composable: Discover and share information across contexts without pre-defined integration or portability criteria
Separable and selective : Information sets can be encrypted or obfuscated, or separated across multiple identifiers, or otherwise split according to the controller’s preference
If “identity systems can reliably associate information with unique identifiers”, as mentioned earlier, we need an internet identity system that establishes a minimal protocol for managing and routing to trusted data, leaving everything else to ingenuity and diversity of app developers.
We want to avoid siloed systems – including specific applications, registries or blockchains – and maximize data type flexibility. We need an easy-to-use system that allows us to build applications with rich forms of data, associate that data with appropriate identifiers, and get the most out of our identities and collective information.
This technology and model is relatively new, but it is developing rapidly. Thousands of Web3 developers are already building this identity system using tools like Spruce and infrastructure like Ceramic. Not only does this identity and data model help developers build more robust and scalable Web3 applications faster than ever before, but Web3 together can build a composable data universe that the Web2 platform cannot compete with.
The future of web3 and the Metaverse blurs the boundaries between the off-chain world and on-chain space. Therefore, for the construction of an identity management system, how to establish a mapping relationship between off-chain identities and on-chain data is a digital living space for developing a new social system. must do homework and stand on the ground.
The isolated identity system is an obstacle to the seamless Dapp experience in the future. Building a distributed user table is beneficial to open up existing application silos. How to find the balance of the DID system between flexibility, openness and reliability, the author recommends For infrastructures such as Ceramic, we will make project analysis in subsequent related articles.
H.Forest Ventures will do its best to fully understand the relevant information for each shared content. If you have any ideas about the content of this article, you can contact the H.Forest Ventures team.
Effective communication is everything.
Posted by:CoinYuppie，Reprinted with attribution to:https://coinyuppie.com/how-does-decentralized-identity-outperform-the-existing-web2-identity-system/
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.