모듈러 블록체인의 개념은 핵심 블록체인 기능이 별개의 구성 요소로 세분화되면서 생겨났다. 2023년에는 각 세분화된 영역 내에서 혁신을 목표로 하는 프로젝트가 등장하였다. 반면 2024년에는 이러한 프로젝트 간의 협업의 중요성이 부각될 것으로 예상된다. 모듈러 블록체인 프로젝트들은 필연적으로 상호의존되기에, 협업은 업계에서 중요한 역할을 할 것이다.

본 호스트들은 업계 종사자들과의 교류를 하는 과정에서 개발자, 리서처, 프로젝트간 의견과 목표를 공유할 필요성을 인식했다. 이에 따라 셀레스티아, 어베일, 니어, 이니티아, 헤로도토스, 스타크넷 등 업계 리더를 포함한 다양한 팀을 초청했다. 모듈러 블록체인의 미래에 대한 인사이트를 얻고 이러한 프로젝트들이 어떻게 서로 시너지를 낼 수 있는지 이해하는 것이 본 행사 기획의 목표이다.

1. RSVP

Co-hosted by: Radius, Four Pillars, ChainLight, Decipher

날짜 : 2024년 3월 29일 (금요일) 오전 11시 30분 — 오후 6시 30분(한국 시간)

Location : 서울시 강남구 강남대로 374, 20층 (구글 지도, 네이버 지도)

Modular Harmony · Luma

2. 타임라인

• 11:30–12:00 | 네트워킹
• 12:00–12:30 | ChainLight
• 12:30–13:00 | Celestia & Four Pillars
• 13:00–13:30 | Fhenix
• 13:30–14:00 | Initia
• 14:00–14:30 | Avail
• 14:30–15:00 | 네트워킹
• 15:00–15:30 | Scroll
• 15:30–16:00 | Radius
• 16:00–16:30 | Herodotus
• 16:30–17:00 | Axelar
• 17:00–17:30 | Near
• 17:00–17:30 | 네트워킹

3. 스피커와 주제

Chainlight: “보안 관점에서 보는 모듈러 블록체인 아키텍처”

(Twitter)

ChainLight는 스마트 컨트랙트를 보호하고 블록체인에서 성공할 수 있도록 맞춤형 보안 솔루션을 제공한다. 그리고 이번 이벤트에서 모듈형 블록체인 아키텍처의 잠재적 위험에 대해 이야기할 예정이다.

“모듈러 블록체인의 보안 리스크”에 대한 Chainlight의 최근 리서치도 확인해보자.

Celestia and Four Pillars: “모듈러 블록체인의 역사”

(Twitter)

셀레스티아는 누구나 쉽게 자신만의 블록체인을 안전하게 운영할 수 있는 DA 인프라를 제공하는 프로젝트이다. 이번 행사에서는 셀레스티아의 닉과 포필러스의 희창이 모듈형러 블록체인의 역사에 대해 이야기할 예정이다.

Fhenix: “프라이버시 — FHE vs ZKP”

(Twitter)

피닉스는 이더리움에 암호화된 데이터에 대한 연산을 제공하는 최초의 완전동형암호화(FHE) 기반 L2이다. FHE(완전 동형 암호화)와 그들의 제품에 대해 발표할 계획이다.

Initia: “롤업의 미래”

(Twitter)

Initia는 연결되고 통합된 (Interwoven) 모듈 네트워크 시스템을 도입하여 L1 오케스트레이션 계층, L2 네트워크, 통신 프로토콜을 아우르는 독특한 아키텍처를 제공한다. 롤업 프레임워크를 통해 Initia는 자체 L2 롤업을 만들 수 있게 하면서, 생태계 내에서 연결성을 보장 가능하다. 이들은 “롤업의 미래”에 대해 이야기할 예정이다.

Avail: “롤업의 파편화 문제의 해결책”

(Twitter)

Avail은 모듈러 실행 레이어가 신뢰를 최소화하는 방식으로 확장하고 상호 운용할 수 있는 웹3 인프라 레이어이다. 현재 블록체인 시스템의 문제인 유동성 파편화는 DA를 통해 해결할 수 있다는 것에 대한 발표를 진행할 예정이다.

Scroll: “영지식 증명의 현재 개발자 경험”

(Twitter)

스크롤은 바이트코드 수준의 호환성을 갖춘 네이티브 L2 zkEVM을 구축하고 있다. 덕분에 기존 이더리움 앱을 그대로 스크롤에서 운영할 수 있다. 스크롤은 영지식 증명의 개발자 경험에 대한 발표를 진행할 예정이다.

Radius: “롤업을 위한 컴포저빌리티 솔루션”

(Twitter)

래디우스 공유 시퀀싱 레이어는 MEV 보호, 검열 저항, 재정적 지속 가능성으로 롤업 탈중앙화를 제공합니다. 이 팀은 현재 구축 중인 공유 시퀀싱 레이어에 대해 이야기할 예정이다.

Herodotus: “스토리지 증명과 Herodotus 데이터 프로세서”

(Twitter)

헤로도토스는 스마트 콘트랙트를 위해 특별히 설계된 안전한 온체인 데이터 스토리지와 증명 기반 데이터 액세스 솔루션을 제공한다. 이 플랫폼은 스토리지 증명과 같은 기능을 통해 이더리움 레이어 전반에서 현재, 과거 및 다른 체인 데이터에 액세스할 수 있도록 지원한다. API는 스토리지 증명 생성에 드는 비용을 상호화하여 개발자의 프로세스를 간소화하므로 상당한 시간을 절약할 수 있다.

이 플랫폼은 다양한 블록체인을 지원하며 거버넌스, 계정 복구, 보험, 브리징, 랜다오 액세스, 크로스체인 대출, 게임, 보안 강화를 위한 회로 차단기 구현 등 다양한 사용 사례를 제공하기에 이에 대한 발표를 진행할 예정이다.

Axelar: “블록체인간 상호운용성”

(Twitter)

Axelar는 프로그래밍 가능한 Web3 상호운용성 플랫폼으로, 차세대 인터넷 애플리케이션을 수십억 명의 사용자로 확장하는 것을 목표로 한다. 다른 크로스체인 커넥터와 달리 Axelar는 지분 증명 블록체인이며 이번 이벤트에서 상호운용성에 대해 이야기할 예정이다.₩

Near: “Near DA & 체인 추상화”

(Twitter)

NEAR는 현재 체인 추상화 스택을 구축하여 빌더들이 수십억 명의 사용자와 모든 블록체인에서 확장 가능한 앱을 만들 수 있도록 지원하고 있다. 이번 이벤트에서 체인 추상화의 개념을 소개할 예정이다.

아래 트윗을 통해 행사의 테마에 대해 더 알아보세요:

Theme1: “과거”

Theme2. “현재”

모듈러 하모니에 초대합니다 was originally published in Four Pillars on Medium, where people are continuing the conversation by highlighting and responding to this story.

Four Pillars is pleased to announce the successful completion of its funding round, raising $0.5M from Kakao Ventures, Hashed, and Bass Investment.

Four Pillars, a research-based blockchain startup, announced on January 16th that it has secured an investment of $0.5M (700 million won) from Kakao Ventures, Hashed, and Bass Investment. Four Pillars assists businesses in effectively implementing blockchain technology, offering customized solutions based on their collaboration experience and technical understanding of blockchain projects. We provide timely insights into blockchain technology, fundamentals, legal aspects, and market analysis, saving time and costs for companies considering entering the blockchain market.

The global blockchain market trends are rapidly evolving, and the importance of research is much greater than in Web2 businesses. Companies face practical difficulties in finding and applying the vast amount of information scattered worldwide. This is also the reason why global protocols struggle to enter the Korean market and Korean companies face challenges in expanding to the global market. Four Pillars aims to grow as a global research-based blockchain company by connecting Korean and global companies and projects, thereby resolving information asymmetry.

Four Pillars is led by protocol experts from Penn State, KAIST, and Seoul National University. The three co-founders established the company in 2023 after accumulating research experience in blockchain companies and conglomerates. The team, known for its outstanding blockchain industry understanding, has been recognized for its capabilities through collaborations with various global blockchain projects.

With the new investment, Four Pillars plans to acquire talent and accelerate product development. It will continue to publish research content and aims to contribute to the blockchain ecosystem by facilitating the participation of various stakeholders in the blockchain scene.

Brian Jang, Director of Kakao Ventures, acknowledges the key team members of Four Pillars for their unique insights in the blockchain industry. He expects the company to grow rapidly by converting the demands of protocols and companies into business outcomes based on its unique research capabilities.

CEO Steve Kim of Four Pillars stated, “Our priority and company goal is to create a developer-friendly environment with high-quality research and products based on it. We aim to lower the entry barriers to blockchain and expand the overall market for Web3 by addressing unfriendly situations for users and developers.”

About Four Pillars

Four Pillars empowers businesses to launch successful blockchain ventures through customized solutions based on deep industry insights and technical expertise. We deliver cutting-edge knowledge in blockchain technology and market trends, streamlining the path for companies to effectively enter the blockchain domain.

Four Pillars Links: https://link3.to/fourpillars

Four Pillars secures $0.5M from Kakao Ventures, Hashed, and Base Investment was originally published in Four Pillars on Medium, where people are continuing the conversation by highlighting and responding to this story.

Author: Jay(Jay Jeong)

* 본 글을 포함한 다양한 리서치 아티클을 Four Pillars 공식 웹사이트에서 확인해보세요!

세 줄 요약:

1) 블록체인의 표준은 우리가 블록체인으로부터 꿈꾸는 가치들이 어떤 방식으로 구현되는 지와 관련이 있는데, 특히 이더리움의 표준 제안인 EIP를 살펴보는 것은 전체 블록체인 생태계의 트렌드를 톺아보고 향후 구현 방식을 논의하는 데에 상대적으로 효과적일 수 있다.

2) 더욱이, 점점 더 많은 플레이어들이 EIP에 참여하고 있는데 새로이 제안되는 EIP들 중 대부분은 이 신규 참여자들에 의해 제안되고 있다.

3) 이외에도 EIP는 좋은 프로토콜 거버넌스 사례로써, 혹은 특정 아이디어를 탐험하고 구체화해볼 수 있다는 점에서 의의가 있다.

본 아티클은 리서처 관점에서 관찰하였던 EIP의 역사 및 참여 현황을 간략히 톺아보며 다양한 참여자들에게 EIP가 블록체인 업계의 과거와 현재를 이해하는 데에 얼마나 효과적인 리소스인 지 그 중요성을 알리는 데에 1차적인 목적이 있지만, 나아가 현 EIP들의 개선을 위한 직접적인 참여 계기, 궁극적으로는 이더리움 내외에 새로운 트렌드를 전파하는 오피니언 리더로써 전향하는 계기를 만들어주는 데에 그 목표가 있다.

1. 서론

블록체인은 우리에게 분산화된 네트워크 구조를 제안하였으며, 스마트 컨트랙트의 적용은 다양한 자산들이 P2P 네트워크 상에서 프로그래밍되어 활용될 수 있도록 그 가능성을 제시하였다. 이를 통해 우리는 몇년이 채 안되는 짧은 기간동안 데이터 주권, 검열 저항성, 보안성, 투명성 등 블록체인을 통해 달성할 수 있는 일련의 가치를 설정하고 새로운 디지털 상호작용을 맞이할 준비를 한다.

하지만 실질적으로 위 가치들의 구현을 통해 기존의 시스템을 대체하거나 보완하기에는 아직까지 블록체인 업계가 갈 길은 먼 듯 하다. 사용자들이 위 가치에 대한 중요성을 인식하는 것은 둘째치고, 사용자 혹은 어플리케이션들이 당장에 블록체인 상의 다양한 서비스들과 상호작용하기 위해서는 프로토콜 인프라의 세부사항을 이해해야하거나 관련된 부가적인 과정을 거치는 불편함을 겪어야하기 때문이다 — 그런데 우리는 일반적으로 (블록체인을 통해서가 아닌) 디지털 상호작용을 함에 있어서, 그것들이 기저하는 백엔드 혹은 네트워크의 이슈 따위를 전혀 신경쓰지 않는다.

혹자는 이를 ‘블록체인의 UX 이슈’라고 포괄하여 표현하기도 하는데, 결국 사용자의 이러한 불편함을 해소할 수 있는 방법은 사용자들에게 요구되는 스텝들을 인프라 단에서 떠안는 것이다. 즉, 어플리케이션들이 기본 인프라에 직접 접근할 필요없이 서비스 인터페이스에만 집중할 수 있도록 인프라 단에서 관련한 통신 표준을 잘 구성해두어야 이러한 불편함은 해소될 수 있다*.

*통신 표준에도 다양한 계층이 존재하는데, 특히 미들웨어의 성격을 띄는 표준들이 이와 관련이 깊은 경우가 많다.

따라서 현존하는 블록체인 메인넷들의 표준들을 살펴본다면 아직까지 추상적이기만한 블록체인의 가치들이 왜 아직 이렇게까지 밖에 구현될 수 밖에 없는 지 이해를 도울 수 있고, 나아가서 다양한 어플리케이션들이 향후 ‘어떠한 방식’ 으로 서로 상호작용될 수 있을 지에 대한 구체적인 논의를 함께 촉진해볼 수도 있을 것이다.

이러한 맥락에서 특히 이더리움의 표준들을 살펴보는 것은 블록체인 전체 생태계의 현 상태를 진단하고 미래를 대비하는 데에 상대적으로 효과적일 수 있다 — 이더리움의 표준 역사 속에는, 오늘날의 다양한 블록체인이 만들어지기까지 참고할 수 있을만한 고민의 흔적들이 모두 담겨있기 때문이다. 더욱이, 메인넷들 중에서 가장 강력한 커뮤니티를 가지고 있는 이더리움 상에는 네트워크를 한층 발전시켜줄 새로운 표준들이 끊임없이 제안되고 있을 뿐만 아니라, 성격이 다른 타 메인넷에서 참고할 수 있을만한 표준들까지 실험적으로 활발히 제안되고 있다. 요컨대, 이더리움의 표준 기록은 블록체인의 거시적인 발전 트렌드를 좇을 수 있는 거대한 아카이브인 셈이다.

2. EIP 개요

2.1 EIP의 범위

이더리움의 표준들은 EIP(Ethereum Improvement Proposal)라는 이름 하에 제안된다. EIP-1에 따르면 EIP의 범위는 크게는 3가지, 혹은 세분화해서 6가지로 분류된다.

많은 EIP들이 제안되어왔지만(i.e., 10월 17일 기준으로 총 689개), 이중에서 사실상 논의가 지속적으로 일어나는 범주는 Standard Track이다. Meta 혹은 Informational 에 해당하는 EIP들은 EIP 프로세스 자체를 설명하거나, 이더리움 네트워크의 하드포크와 같은 프로토콜의 주요변경사항들을 설명하는 메타 문서에 가까운 안건들로 구성되어 있다 — 따라서, 실제로 Meta나 Informational의 EIP들은 대부분 2020년도 이전에 제안되었다.

Standard Track 내에서도 특히 Core와 ERC 표준에 대한 논의가 활발히 일어나는데, Core EIP의 경우에는 주로 합의 레이어 혹은 실행 레이어와 관련하여 이더리움 네트워크의 변경사항을 적용하는 것과 직접적으로 연관된 표준들로 구성돼있다. 따라서 다른 범주의 EIP들에 비해, Core EIP들은 더욱 신중히 검토되고 채택에 앞서 AllCoreDevs Call 등의 부가적인 과정을 통해 클라이언트들과의 긴밀한 토의가 요구되는 경우가 많다.

반면에 ERC의 경우, 어플리케이션 단에서 적용할 수 있는 표준 혹은 컨벤션 등을 포함하기 때문에 조금 더 어플리케이션 혹은 최종 사용자와 가까운 표준이라고 할 수 있다 — ERC-20과 같은 토큰 표준, ENS와 같은 Naming Resolution Sevice, 라이브러리 패키지, 계정추상화 등 우리에게 친숙한 표준들이 ERC에 포함되어 있다. Core EIP와 달리 ERC는 배포되는 대상이 어플리케이션 단인 경우가 많다.

Networking과 Interface 범주에 속하는 EIP는 이더리움 네트워크 내 데이터의 통신이 더 원활하게 동작하거나 Core EIP 및 ERC 들을 지원할 수 있는 표준들로 구성돼있다.

2.2 EIP 프로세스

위 EIP들은 모두 사전 논의를 시작으로, 어느 누구나에 의해서 제안될 수 있다. 하지만 제안되는 안건들이 모두 채택(Finalized)이 되는 것은 아니다 — EIP는 그 이름처럼 Proposal 이기 때문에 개별 안건들이 정식 표준으로 채택되기 위해서는 위와 같은 일련의 프로세스를 따르게 된다.

• Idea : 본격적인 채택 프로세스에 들어가기에 앞서, 전체 커뮤니티 구성원을 대상으로 해당 안건이 EIP로써 포함되는 것이 적절한 지 논의되는 사전 단계이다. Ethereum Research 혹은 Ethereum Magicians에서 해당 논의가 이뤄진다.
• Draft : 제안된 안건이 EIP의 폼을 갖추면 EIP 에디터들에 의해 EIP 저장소(Repository)에 병합된다.
• Review : 해당 안건이 동료 평가(Peer Review)를 받을 수 있는 단계이다.
• Last Call : 안건이 Final 단계로 가기 이전의 마지막 검토에 해당하는 단계이며, EIP 에디터를 통해 이 단계가 결정된다 — 검토를 위한 최종 데드라인이 정해지는데, 일반적으로 14일 이후로 설정된다.
• Final : 해당 안건의 최종 확정이 되었음을 의미한다.
• Stagnant : Draft, Review, Last Call 상태에 있는 안건들 중에서 6개월 이상 변경 사항이 없는 경우, 이 단계로 변경된다. 단, 작성을 재개하거나 업데이트 시 다른 상태로 변경된다.
• Withdrawn : 해당 안건이 완전히 회생될 수 없는 상태를 뜻한다. 이 단계의 EIP는 다시 복구될 수 없으며, 해당 안건의 논의를 이어가는 유일한 방법은 새로운 EIP를 제안하는 것 뿐이다.
• Living : 이 단계는 해당 안건이 지속적으로 업데이트 되는 특별한 상태이며, 해당하는 EIP는 현재 2개 뿐이다 (i.e., EIP-1 & EIP-5069)

비트코인의 BIP(Bitcoin Improvement Proposal)와 파이썬의 PEP(Python Enhancement Proposal)에서 영감을 받은 이 프로세스의 세부 사항, EIP의 형식, 그리고 에디터들에 대한 책임 및 역할은 EIP-1과 EIP-5069를 각각 참조하길 바란다.

다만 몇 가지 짚고 넘어갈 것은 이 EIP 프로세스가 각 단계로 나아감에 있어서 EIP 작성자, EIP 에디터, 이더리움 코어 개발자, 그 외 이해관계자 및 커뮤니티 구성원 등 광범위한 주체들의 참여를 필요로 한다는 점, 그리고 해당 논의들은 모두 오프라인에서 이뤄지는 오프체인 거버넌스를 채택한다는 점이 있다. 따라서 프로토콜의 현 상황에 따라 특정 EIP가 매우 빨리 채택될수도, 혹은 무기한으로 채택이 연기될 수도 있다. 이에, 보다 효율적인 프로세스를 위하여 Ethereum Magicians 등에서의 사전 논의가 매우 권장된다.

또한, 지금까지 제안된 약 689 여개의 안건들 중 약 34%에 육박하는 안건들이 리비전 단계 혹은 최종적으로 채택되었다는 사실은 ‘이더리움 네트워크에 그간 많은 발전이 있었다는 것’을 방증하는 동시에, ‘표준 채택이 사실상 쉽지 않다’는 사실 또한 동시에 방증하기도 할 것이다 — 그도 그럴것이, 표준이 많으면 많을수록 네트워크의 복잡도는 상승하고 기존의 표준들과의 충돌을 피하기 위해 검수 작업에 더욱 많은 비용이 들 수 있다.

3. EIP의 역사와 블록체인 트렌드

지금까지 기록된 EIP들은 네트워크가 구성되기 위한 매우 기본적인 요소부터 당시의 업계 트렌드까지 직간접적으로 상당부분 반영해오고 있다. 물론 초창기에 비해 다양한 특성을 가진 메인넷들이 점점 더 많이 등장함에 따라, 모든 영역에 있어 지금 시점의 EIP가 업계를 선두하는 표준 프레임워크라고는 말할 수는 없겠지만, 여전히 표준들의 큰 흐름들은 이더리움을 중심으로 이뤄지고 있는 듯하다.

3.1 ‘~ 2017’

2017년도 이전까지는 블록체인이라는 개념 혹은 정체성이 주로 확립되고 다양한 블록체인들이 등장하는 시기였다. 따라서, 이 기간동안에는 네트워크가 동작하기 위한 기초적인 기능의 구현 혹은 안정적인 운영을 위한 논의가 주로 이루어졌다.

당시에 제안되었던 EIP들은 업계 내에서 표준 트렌드를 선두하였는데 네트워크 통신 로직과 관련된 스펙, 그리고 오늘날 존재하는 다양한 토큰 및 컨트랙트들의 기본이 되는 표준들이 이 시기에 제안되었다.

대표적으로는 가장 대중적인 토큰 표준인 ERC-20, ENS의 뼈대를 이루는 ERC들, 그리고 서명된 데이터를 다루는 표준인 ERC-191 등을 꼽을 수 있겠다. 이외에도 다양한 토큰, 컨트랙트, 라이브러리 및 OPCODE(Operation Code), 그리고 코드 최적화와 관련된 표준들이 제안되었다.

3.2 ‘~ 2019’

이어지는 2년 동안에는, 기초적으로 구성된 메인넷들 상에 어플리케이션들이 조금씩 배포되기 시작한 시기였다 — 특히 이 시기는 NFT와 DeFi가 본격적으로 등장하여 대중들에게 블록체인이 처음으로 많이 알려지는 시기이기도 하다. 가장 널리 알려진 NFT 표준인 ERC-721도 이 시기에 제안되었는데 이 외에도 다양한 토큰 표준 및 컨트랙트들이 생겨나기 시작하다보니 자연스레 네트워크 단에서는 스마트 컨트랙트를 관리할 수 있는 체계적인 프레임워크의 필요성이 대두되었다.

이에, 각종 컨트랙트들을 탐지하고 그것의 소유권 관리와 관련된 표준들이 다수 제안되었는데 이외에도 주목할만한 표준들은 업그레이드 가능한 컨트랙트의 포문을 열어준 EIP-1167, 그리고 컨트랙트 코드를 기반으로 컨트랙트 주소를 만드는 새로운 OPCODE(i.e., CREATE2)를 제안하는 ERC-1014 등이 있다.

또한 네트워크 사용량이 점차 늘다보니 네트워크 가스비에 대한 정책도 본격적으로 이슈화되기 시작한 때가 이 때이다 — 과거, 가스비 경쟁에 지나치게 의존적인 구조 때문에 사용자들은 가스비 예측이 매우 힘들었다. 이에, 기본 수수료(Base Fee) 와 선택적 수수료(Priority Fee)를 분리하여 트랜잭션 당 가스비 예측을 더욱 쉽게 할 수 있도록하는 EIP-1559도 이 시기에 제안되었다.

3.3 ‘~ 2021’

2년에 걸쳐 이더리움 메인넷을 통해 NFT와 DeFi를 비롯한 어플리케이션들의 잠재력이 어느 정도 확인된 이후, 다양한 메인넷에서는 저마다의 특징들을 내세운 다양한 토큰 프로젝트와 DApp들이 등장하였다. 이에, 이전보다 훨씬 더 많은 L1들이 등장하여 주목을 받게 되었고, NFT와 DeFi 생태계의 내러티브 역시 대중들로부터 더욱 더 주목을 받게 되었다.

생태계가 풍부해짐으로 인해 자연히 더욱 많은 트랜잭션을 감당하게 된 이더리움 진영에서도 확장성 솔루션 혹은 가스비 정책에 대한 논의가 더욱 심화되었으며, NFT와 DeFi 단에서는 자신들의 내러티브를 더욱 강화할 수 있는 고도화된 기능들을 구현케하는 표준들이 제안된다 — 크리에이터들의 로열티 정책을 구현하기 위한 ERC-2981, 그리고 한 트랜잭션 내에 무담보 대출과 상환을 모두 가능케하는 플래시론(i.e., ERC-3156) 등이 이 시기에 제안되었다.

3.4 ‘2022 ~’

2022, 23년은 블록체인에 대한 사람들의 관심이 극에 달했던 시기인 동시에 테라 및 FTX 사태, 그리고 매크로 경제의 악영향이 뒤이어짐에 따라 그 관심이 급속도로 식기도 한 해이다. 이에 따라, 각 프로토콜들은 자신들의 정체성 및 성장전략을 더욱 확고히하게 되었는데 이 과정에서 상호간의 이념 충돌(e.g., Monolithic vs. Modular), 더욱 유연한 환경을 위한 프로토콜 마이그레이션, 혹은 자체 생태계 구축(e.g., AVAX — Subnet / Polygon — Supernet / Cosmos — ICS, etc.)과같은 이벤트가 발생하였다.

특히 해당 논의들은 이더리움을 중심으로 많이 이루어지기도 하였는데 이더리움이 PoS로의 전환을 위한 로드맵의 주요 전환점을 돌기도 하였고, 또한 상대적으로 타 메인넷들에 비해 보안성과 탈중앙성이 우수한 이더리움 및 롤업 생태계가 일련의 사태들로 인해 큰 주목을 받았기 때문이다. 이에, PoS를 위한 기초적인 네트워크 환경 조성과 관련된 EIP, 그리고 PoS 전환 로드맵의 세부사항에 따라 롤업과 관련된 일련의 EIP들 역시 많이 제안되어오고 있다.

한편, 업계 내외에서는 블록체인에 대한 유스케이스 혹은 UX에 대한 논의가 대두되기 시작하기도 하였다. 이에, 실질적으로 사용자들의 어플리케이션 경험을 개선하고 웹2 — 웹3 간의 간극을 이어줄 수 있는 미들웨어 성격의 표준들이 상당히 많이 제안되어오고 있다 — 오랫동안 논의가 멈춰져있었던 ERC-4337 기반의 계정 추상화가 다시금 주목을 받고, 이외에도 트랜잭션, 신원, 토큰 바운드 계정 등과 같은 미들웨어 성격의 표준들이 많이 제안되어오고 있으며, ZK 및 AI를 접목하여 새로운 경험을 보여줄 수 있는 확장 표준들 역시 많이 제안되어오고 있다.

4. 새로운 참여자들로부터의 지속적인 기여

이전 그래프에서 발견할 수 있는 다른 재미있는 포인트들은 1) 블록체인이 대중으로부터 많은 주목을 받기시작한 2018년, 그리고 2022년에 EIP 제안의 수가 최적값을 기록했다는 점, 2) 그리고 이 최적값들은 제안된 ERC의 수에 의해 견인되었다는 점이다 — 다른 연도에는 Core EIP가 가장 지배적이었는데 특히 2018, 2022, 그리고 2023년도에는 ERC가 가장 지배적인 것을 관찰할 수 있다.

그런데 만일 EIP를 제안하는 주체들의 구성이 변함이 없다면(즉, EIP에 초창기부터 기여하고 있던 주체들에 의해서만 계속해서 EIP가 제안이 된다면), 제안되는 EIP의 수가 늘어나는 이러한 트렌드는 잘 설명되지 않는다. 더욱이, 표준이 네트워크에 많이 누적될수록 그 복잡도는 증가하기때문에 새로운 표준이 제안되고 채택되는 과정은 더욱 느려질 수 밖에 없을 것이다.

살펴본 결과, 이더리움 네트워크에 대해 가장 잘 알고있고 초기부터 네트워크를 만들어온 주체들(에디터들을 포함하여)이 많은 EIP를 제안한 것은 사실이다. 하지만 더욱 놀라운 점은, 이들을 제외하고도 상당 부분의 EIP 제안들이 EIP 제안에 경험이 적은 사람들로부터 제안되었다는 것이다 — 지금까지 EIP 제안자의 수는 800명 남짓을 기록하고 있으며, EIP를 제안한 경험이 5번 미만인 제안자들의 비율이 90%가 넘는다.

연도별로 새로운 제안자(i.e., 이전에 EIP를 제안한 경험이 없는 주체)의 수를 프로젝션해보면 이는 더욱 명확해진다 — 아웃라이어(i.e., 2018’s & 2022’s)를 제외하고 새로운 제안자의 수는 꾸준히 상승하고 있다. 하지만 더욱 놀라운 것은 이렇게 새로 들어온 제안자들이 당해 제안된 전체 EIP들 중 평균적으로 7–80%가 넘는 비율로 기여를 하고 있다는 것이다.

즉, 적지않은 사람들이 매년 새로이 EIP에 참여를 하고 있으며, 이들은 기존의 이더리움 커뮤니티와 함께 어울리며 새로운 표준을 만들어나가는데 상당한 기여를 하고 있다.

5. 마무리하며

참조할만한 레퍼런스가 많이 없는 백지와 같은 상황에서 네트워크를 개선하기 위해 거쳐야했던 수많은 논의의 흔적들이 성문화된 형태로 존재하는 메인넷은 이더리움이 유일하다. 이에, 앞서 간략히 톺아보았듯 EIP를 살펴보는 것은 블록체인의 역사와 현상태를 진단하는 데에 굉장히 도움이 되는 자료라고 할 수 있다. 더욱이, 점점 더 많은 플레이어들이 새로이 참가함으로 인해 제안되는 EIP들은 우리가 블록체인을 통해 꿈꾸는 미래를 구체화시키는 데에 크나큰 기여를 하고 있다.

이외, 현 EIP 프레임워크가 가지는 부가적인 의의 및 한계점을 간략히 생각해보며 본 아티클을 마무리해본다.

5.1 좋은 거버넌스 사례로써의 의의

거버넌스란 본디 한 조직이 그들의 비전을 추구하기 위한 주요 의사결정으로, 어디까지나 제1 목적은 조직의 비전 달성이다. 따라서 거버넌스에 의해 결정되는 프로토콜의 표준들은 프로토콜의 비전과 일치하도록 제안되어야 하며 비전을 달성하는데에 효과적이어야 한다. 이러한 측면에서, EIP의 범위는 ‘보안성과 탈중앙성을 해치지 않는 선에서 확장성을 개선하고(i.e., 트릴레마의 개선), 다양한 사람들이 네트워크를 통해 이점을 누릴 수 있도록하는 월드 컴퓨터(World Computer)가 되는 것’을 목표로 하는 이더리움의 비전과 상당히 정렬되어 운영되고 있다.

또한, 이더리움 생태계는 로컬라이즈드 컨퍼런스, 포럼, 블로그, 정기적인 개발자 콜, 번역톤(Translathon) 주최, 유튜브 채널, 이외 다양한 컨텐츠들을 통해 다양한 참여자들로하여금 EIP의 온보딩이 용이하도록 진입장벽을 낮추려는 노력을 하고 있다 — 더욱이, 결과적으로 매년 새로이 EIP에 참여하는 주체는 꾸준히 증가하고 있으며 이들이 실질적으로 새로이 제안하는 EIP들은 이더리움 네트워크가 새로운 가치를 구현하는데에 상당한 기여가 되고 있다.

요컨대, 함께 네트워크를 만들어나간다는 것으로부터 기원한 탈중앙화의 개념을 ‘다양한 참여자들이 온보딩하기 쉬운 환경’으로써 정의한다면, 이더리움의 거버넌스는 그것의 미션과 일치하면서도 결과적으로 점점 더 참여자들의 분포가 탈중앙화되고 있으므로 좋은 프로토콜 거버넌스 사례라 할 수 있을 것이다.

5.2 포크가능성을 통한 특정 아이디어의 실험 기회로써의 의의

한편, EIP는 다양한 표준들이 집약되는 공간이긴 하지만, 이더리움과 같이 광범위한 상호작용을 지원하기 위한 목적으로 설계되는 General-Purpose 프로토콜들은 표준들 간 다양한 충돌가능성을 고려해야하기 때문에 기본적으로 개별 표준의 채택이 빠르게 이루어지지도 않을 뿐만 아니라, 특정 문제에 집중한 표준이 최종적으로 완결(Finalized)되기 또한 쉽지 않다.

따라서 특정한 유스케이스를 관찰하기 위해서는 이더리움과 같은 General-Purpose 프로토콜을 통해 그것을 기대하기보다 그 의도에 최적화된 프로토콜을 선택하거나 구축하여 민첩하게 실험해보는 것이 더욱 효과/효율적일 수 있다. 하지만 별도의 프로토콜을 탐색하고 구축하는 것 역시 많은 시간과 비용을 요한다.

이에, 특정 EIP를 베이스로 이더리움을 포크하여 이 과정을 개시하는 것은 좋은 시도일 수 있다 — 예를 들어, ERC-20의 저자인 Fabian Vogelsteller는 기존 토큰 및 계정 표준들의 비유연함을 지적하며 아직 Draft 단계인 ERC-725를 기반으로 크리에이티브 경제만을 위한 LUKSO라는 프로토콜을 구축중이다. 이러한 접근법은 EIP를 극대화하여 활용하는 동시에 이더리움과의 호환성으로부터 오는 시너지까지 누리며 특정 유스케이스를 실험해볼 수 있다.

즉, 다양한 EIP를 제안해보거나 논의에 참여해보는 것은 어쩌면 특정 프로젝트 탄생의 기회이자 이더리움 네트워크에 간접적으로 기여할 수 있는 또 다른 방법일 수 있는 것이다.

5.3 하지만 비개발자의 온보딩은 여전히 어렵다

우리가 이더리움 네트워크를 통해 잠재적으로 꿈꾸고 있는 다양한 유스케이스 및 가치들은 소수의 개발자들에게 국한된 것이 아닌, 네트워크에 온보딩할 수 있는 모든 종류의 참여자에 의해 만들어진다. 따라서 궁극적으로는 이더리움의 표준을 함께 만들어나가는 주체 역시 크리에이터 혹은 비즈니스 관계자 등 다양한 시각을 가진오피니언 리더들로 구성돼야 할 것이다.

하지만 비개발자들의 EIP 온보딩은 여전히 어렵다. 앞서 언급한 바와 같이 이더리움 생태계는 다양한 소통 채널을 통해 EIP에 대한 진입장벽을 낮추려는 노력을 하고 있지만, 결정적으로 비개발자들의 쉬운 이해를 돕기위한 하이레벨의 자료들은 거의 존재하지 않거나 파편화되어 존재한다. 더욱이 EIP들을 한눈에 톺아볼 수 있는 랜드스케이프 등 역시 부재하다 — 어떤 EIP가 주류인지 추적하기는 여전히 어렵고, EIP의 넘버링 규칙은 혼란스럽다.

여전히 서구권, 그리고 개발자 중심으로 맞춰진 문서 및 주요 이벤트의 타임라인 등을 포함하여 비개발자들을 위한 온보딩 과제들이 개선된다면 더욱 다채로운 가치 구현 및 실질적으로 채택에 필요한 표준들이 빠르게 논의되는 데에 큰 도움이 될 지도 모르겠다.

이 글의 비주얼을 제공해주신 Kate에게 감사의 말씀을 전합니다.

고지 사항: 본 글은 단순 정보 제공을 위해 작성되었고 투자, 법률, 자문 등 어떤 부분에 대해서도 책임을 지지 않습니다. 특정 자산에 대한 투자를 추천하는 것이 아님을 밝히며, 본문의 내용만을 바탕으로 투자에 대한 의사결정을 내리지 마십시오.

EIP에 시간을 투자하라 was originally published in Four Pillars on Medium, where people are continuing the conversation by highlighting and responding to this story.

Author: Jay(Jay Jeong)

* Visit the Four Pillars official website to explore this and other research articles!

Key Takeaways:

1) Blockchain standards have a lot to do with how the values we envision from blockchains are realized, and looking at Ethereum’s standards proposals can be a relatively effective way to see trends in the entire blockchain ecosystem and discuss future implementations.

2) More and more players are participating in EIPs, and many of the new EIPs are being proposed by these new players.

3) In addition, EIPs have some implications as good protocol governance practices and as a way to explore and refine specific ideas.

This article provides a brief overview of the history and the current state of participation in EIPs from a researcher’s perspective, with the primary goal of educating a wide range of participants on the importance of EIPs as resources for understanding the past and present of the blockchain industry, but it also aims to inspire direct participation in improving current EIPs and ultimately becoming opinion leaders in spreading new trends within and outside of Ethereum.

1. Introduction

Blockchain gave us the idea of a decentralized network structure, and the introduction of smart contracts gave us the possibility of various assets being programmed and utilized on a peer-to-peer network. With these, we establish a set of values that can be achieved through blockchain, such as data sovereignty, censorship resistance, security, and transparency, and prepare to face new digital interactions.

However, the blockchain industry is still a long way from actually implementing these values to replace or complement existing systems. This is because, apart from users recognizing the importance of the above values, users or applications need to understand the details of the protocol infrastructure or suffer the inconvenience of going through the extra steps involved to interact with various services on the blockchain — however in general, we do not care about the back-end or network issues that underlie them when we engage in digital interactions.

Some people refer to these as ‘Blockchain’s UX issues,’ but in the end, the only way to resolve these inconveniences for users is for the infrastructure to handle the processes required by users. In other words, the protocol standards at the infrastructure should be well organized so that applications can focus only on service interfaces without having to access the underlying infrastructure directly*.

*There are many layers to protocol standards, but especially middleware-like standards are often associated with this.

Therefore, looking at the standards of existing blockchain mainnets can help us understand why the value of blockchain is still so far from being realized, and it can also facilitate concrete discussions about how different applications can interact with each other in the future.

In this context, looking at Ethereum’s standards in particular can be relatively effective in diagnosing the current state of the entire blockchain ecosystem and preparing for the future — as the standard history of Ethereum contains all the traces of thought that led to the creation of the various blockchains we see today. Furthemore, Ethereum has one of the strongest communities of any mainnet, so not only are new standards constantly being proposed to further develop the network, but there is also an active experimental effort to propose standards that can be learned from mainnets of a different nature. In short, Ethereum’s standards are a giant archive that allows you to follow the macro evolution of the blockchain history.

2. Brief Overview of EIP Framework

2.1 Scope of EIP

Ethereum standards are proposed under the name of Ethereum Improvement Proposals (EIPs) process. According to EIP-1, the scope of EIPs is categorized into three broad categories, or broken down into six categories.

While many EIPs have been proposed (i.e., a total of 689 as of October 17th), the category that actually generates the most discussion is the Standard Track. EIPs categorized as Meta or Informational consist of proposals that are more of a meta-document describing the EIP process itself, or major changes to the protocol such as a hard fork of the Ethereum network — in fact, most EIPs in the Meta or Informational category were proposed before 2020.

Within the Standards Track, Core and ERC standards are particularly active, with Core EIPs consisting of standards that are directly related to implementing changes to the Ethereum network, primarily related to the consensus layer or the execution layer. Therefore, compared to other categories of EIPs, Core EIPs are more carefully reviewed and often require closer discussions with clients through additional processes such as AllCoreDevs Call prior to their adoption.

ERCs, on the other hand, are more application- or end-user-specific as they include standards and conventions that can be applied at the application level — they include token standards such as ERC-20, Naming Resolution Services such as ENS, library packages, account abstractions, and other familiar standards. Unlike Core EIPs, ERCs’ are often deployed at the application level.

EIPs, which falls under the Networking and Interface categories, consist of standards that make the communication of data within the Ethereum network work more smoothly or support Core EIP and ERC.

2.2 Process of EIP

All of the above EIPs can be proposed by anyone, starting with preliminary discussions. However, not all proposals become finalized — as the name suggests, EIP is a sort of proposal, so individual item should follow the process as above to be finalized.

• Idea : This is a preliminary step before entering the full adoption process, where the appropriateness of the proposal’s inclusion as an EIP is discussed in a forum.
• Draft : When a proposed proposal has been formatted, it is merged into the EIP Repository by the EIP editors.
• Review : This is the step where the proposal is eligible for peer review.
• Last Call : This is the step for the last review before the proposal moves to Final, which is determined by the EIP editors — a final deadline for review is set, typically 14 days or later.
• Final : This step means that the proposal has been finalized.
• Stagnant : Proposalsn the Draft, Review, and Last Call statuses that have not been changed in six months or more will be moved to this step. However, it can be changed to another status when it is resumed or updated.
• Withdrawn : This step means that the proposal cannot be fully revived where the only way to continue discussing the proposal is to propose a new EIP.
• Living : This step is a special status where the proposal can be continuously updated. There are only two EIPs in this step at the moment (i.e., EIP-1 & EIP-5069).

Please refer to EIP-1 and EIP-5069 for the details of this process, the format of the EIP, and the responsibilities and roles for EIP editors.

However, it’s important to note that the EIP process requires the participation of a wide range of actors at each stage, including EIP authors, EIP editors, Ethereum Core developers, and other stakeholders and community members, and that these discussions take place offline. As a result, depending on the current state of the protocol, a particular EIP may be adopted very quickly, or it may be delayed indefinitely, and prior discussion on forums like Ethereum Magicians is highly encouraged to make the process more efficient.

The fact that nearly 34% of the 689 EIPs proposed so far are in the revision or final stage is a testament to how far the Ethereum network has come, but it’s also a testament to how difficult it is to adopt standards — it is no wonder that the more standards there are, the more complex the network, and the more expensive it can be to vet them to avoid conflicts with existing standards.

3. History of EIPs and Blockchain Trends

The EIPs that have been documented so far have directly or indirectly reflected the industry trends of the time, as well as the very basic elements of how networks are organized. Of course, as more and more mainnets emerge with different characteristics than in the early days, it’s hard to say that the EIP is the leading standards framework in all areas at this point in time, but the stream of standards still seems to be centered around Ethereum.

3.1 ‘~ 2017’

Prior to 2017, the concept or identity of blockchain was largely established and various blockchains were emerging, so discussions were focused on implementing basic functionality for the network to work or to ensure its stable operation.

The EIPs proposed at the time were trendsetters for the industry, as many of the standards underlying network communication specifications and those related to tokens and contracts that exist today were proposed around this time.

These include ERC-20, the most popular token standard, ERCs that form the backbone of ENS, and ERC-191, the standard for signed data. Other EIPs have been proposed for various tokens & contracts standards, libraries, and OPCODE (Operation Code), as well as standards related to code optimization.

3.2 ‘~ 2019’

Over the next two years, applications began to be deployed on these rudimentary mainnets — notably, this is when NFTs and DeFi really took off, and blockchain became widely known to the public for the first time. The most widely recognized NFT standard, ERC-721, was also proposed at this time, and with so many other token standards and contracts emerging, the need for an organized framework for managing such smart contracts on the network level naturally arose.

In response, a number of standards have been proposed to detect various contracts and manage their ownership, including EIP-1167, which opens the door to upgradable contracts, and ERC-1014, which proposes a new OPCODE (i.e., CREATE2) to create contract addresses based on contract codes.

This is also the time when the policy on network gas fees started to become an issue as network usage gradually increased — in the past, users had a very difficult time predicting gas fees due to the structure that relied too much on gas fee competition. As such, EIP-1559 was also proposed at this time, which separated the Base Fee from the optional Priority Fee to make it easier to predict the gas fee per transaction.

3.3 ‘~ 2021’

After two years of seeing the potential of applications including NFTs and DeFi on the Ethereum mainnet, various token projects and DApps have emerged on different mainnets with their own unique features. As a result, many more L1s have emerged and gained traction, and the narrative of the NFT and DeFi ecosystem has been intensified by the public.

On Ethereum, where the richness of the ecosystem has naturally led to more transactions, the discussion of scalability solutions or gas fees has intensified, and standards have been proposed to enable more advanced features to further their narrative in the NFT and DeFi space — ERC-2981 to enable royalty policies for creators, and flashloans (i.e., ERC-3156) to enable both unsecured lending and repayment within a single transaction.

3.4 ‘2022 ~’

2022, and 2023 were the years when people’s interest in blockchain was at its peak, but also when it cooled down rapidly due to the Terra and FTX debacles, followed by the adverse effects of the macroeconomy. As a result, each protocol further solidified its identity and growth strategy, which may include clashing ideologies (e.g., Monolithic vs. Modular), migration for more flexible environments, or building their own ecosystems (e.g., AVAX — Subnet / Polygon — Supernet / Cosmos — ICS, etc.).

In particular, much of this discussion has focused on Ethereum, as Ethereum is approaching a major turning point in its Proof of Stake(PoS) transition roadmap, and a series of events have brought the relatively more secure and decentralized ecosystem of Ethereum and Rollup to the forefront of attention compared to other mainstream mainnets. As a result, a number of EIPs have been proposed related to creating the underlying network environment for PoS, and a series of EIPs related to Rollup have also been proposed, depending on the details of the PoS transition roadmap.

Meanwhile, discussions about use cases or UX for blockchain have begun to emerge both within and outside the industry. As such, many middleware-like standards that can actually improve users’ application experience and bridge the gap between Web2 and Web3 have been proposed — ERC-4337-based account abstraction, which has been on hold for a long time, has received renewed attention, and many other middleware-like standards related to transaction, identity, and token-bound accounts have been proposed, and many extension standards that can show new experiences by combining ZK and AI have also been proposed.

4. Steady Engagement by New Proposers

Other interesting points that can be found in the previous graph are 1) that the number of EIP proposals peaked in 2018 and 2022, when blockchain started to receive a lot of public attention, and 2) that these peaks were driven by the number of ERCs proposed — in other years, Core EIPs were dominant, but we can observe that ERCs were dominant in 2018, 2022, and 2023.

However, if the composition of the actors proposing EIPs remains unchanged (i.e., if EIPs continue to be proposed only by actors that have been contributing to the EIP since its inception), this trend of increasing the number of proposed EIPs is not well explained. Moreover, the more standards accumulate in the network, the more complex they become, which should slow down the process of proposing and adopting new standards.

It’s true that many EIPs have been proposed by those most familiar with the Ethereum network and who have been shaping it since its inception (including EIP editors). But what’s surprising is that, aside from them, a significant portion of the EIPs are made up of people with less experience in proposing EIPs — to date, there are approximately 790 EIP proposers, with over 90% of proposers having experience with fewer than five EIP proposals.

This becomes even clearer when we project the number of new proposers (i.e., entities that have never proposed an EIP before) by year — with the exception of the outliers (i.e., 2018’s & 2022's), the number of new proposers is steadily rising. What’s even more striking, however, is that these new proposers contribute, on average, over 7–80% of all EIPs proposed in a given year.

In other words, this means that a significant number of new people join the EIP every year, and they make significant contributions to the creation of new standards as they mingle with the existing Ethereum community.

5. Final Thoughts with Some Implications

Ethereum is the only mainnet that started from a blank slate with few references and codified the many discussions that took place to improve the network. So, as we’ve briefly touched on, looking at EIPs can be a very helpful for diagnosing the history and current state of blockchain. Furthermore, more and more new players are joining and the EIPs they propose are helping to shape the future we envision for blockchain.

Below are some musings on additional implications and limitations of the EIP framework.

5.1 Good Protocol Governance Practice

Governance is the decision-making process of an organization, and its primary purpose should be to achieve the organization’s vision. Therefore, a protocol’s standards, as determined by its governance, must be consistent with the protocol’s vision and proposed to be effective in achieving that vision. In this respect, the scope of EIPs is quite consistent with Ethereum’s vision, which aims to be “a world computer that allows a wide range of people to benefit from the network without compromising security and decentralization (i.e., solving the trilemma)”.

Ethereum ecosystem is also working to lower the barriers to entry to facilitate the onboarding of EIPs through localized conferences, forums, blogs, regular developer calls, Translathon hosting, YouTube channels, and a variety of other content — and as a result, the number of new EIPs each year is steadily increasing, and the new EIPs they propose are making significant contributions to the value of the Ethereum network.

In short, if we define decentralization as “an environment that makes it easy for a wide range of participants to onboard,” then Ethereum’s governance is a good example of protocol governance as it aligns with its mission and results in an increasingly decentralized distribution of participants.

5.2 Another Opportunity from Blockchian’s Forkability

On the other hand, EIP is a space where different standards converge, but general-purpose protocols designed to support a wide range of interactions, such as Ethereum, have to consider various possible conflicts between existing standards, so by default, adoption of individual standards is not fast, and it is not easy for standards focused on specific problems to be finalized.

Therefore, to observe a specific use case, it may be more effective/efficient to choose or build a protocol optimized for that intent and experiment agilely with it, rather than expecting it through a general-purpose protocol. However, exploring and building a separate protocol is also time-consuming and expensive.

As such, it may be a good idea to start this process by forking Ethereum based on a specific EIP — for example, Fabian Vogelsteller, author of ERC-20, has pointed out the inflexibility of existing token and account standards within Ethereum and is creating a protocol called LUKSO to allow a diverse creative economy to be built on top of ERC-725, which is still in draft. This kind of approach allows you to experiment with specific use cases while maximizing the use of EIP and taking advantage of the synergies that come from compatibility with Ethereum.

This means that proposing various EIPs or participating in their discussions could be another opportunity for certain project to be born and another way to indirectly contribute to the Ethereum network.

5.3 However, Still Hard to Onboard for Non-Devs

The various use cases and values we potentially envision for the Ethereum network are not limited to a handful of developers, but are created by all sorts of participants who could potentially onboard the network, so ultimately, the people who co-create Ethereum’s standards should also be made up of opinion leaders with diverse perspectives including general users, creators, business people, etc.

However, non-devs’ EIP onboarding is still challenging. As mentioned above, the Ethereum ecosystem is trying to lower the barriers to entry for EIPs through various communication channels, but crucially, high-level resources to help non-developers understand them are either nonexistent or fragmented. Furthermore, there is no landscape things for EIPs — it’s still difficult to keep track of which EIPs are mainstream, and the numbering conventions for EIPs are very confusing.

If the onboarding challenges for non-developers, such as documentation and timelines of key events, which are still largely Western and developer-centric, could be improved, it might go a long way toward accelerating the discussion of standards that are needed to bring more value and actual adoption.

Thanks to Kate for designing the graphics for this article.

Disclaimer: This article is intended for general information purposes only and does not constitute legal, business, investment, or tax advice. It should not be used as a basis for making any investment decisions or relied upon for accounting, legal, or tax guidance. References to specific assets or securities are for illustrative purposes only and do not represent recommendations or endorsements. The opinions expressed in this article are those of the author and do not necessarily reflect the views of any affiliated institutions, organizations, or individuals. The opinions reflected herein are subject to change without being updated.

Invest Your Time in EIPs was originally published in Four Pillars on Medium, where people are continuing the conversation by highlighting and responding to this story.

Disclaimer: This article is intended for general information purposes only and does not constitute legal, business, investment, or tax advice. It should not be used as a basis for making any investment decisions or relied upon for accounting, legal, or tax guidance. References to specific assets or securities are for illustrative purposes only and do not represent recommendations or endorsements. The opinions expressed in this article are those of the author and do not necessarily reflect the views of any affiliated institutions, organizations, or individuals. The opinions reflected herein are subject to change without being updated.

Author: Jay : : FP (@JayLovesPotato)

Key Takeaways

• Web3 has been attracting great attention, as it has the potential to create a participant-centered symmetric market where participants can independently control and share their data and content based on blockchain technology.
• Identity is a key factor that enables interaction between entities. However, current Web3 stack lacks a holistic identity that can interact with a variety of applications and different types of interactions.
• CyberConnect is creating a social graph that defines the holistic identity through account abstraction and builds Web2-like UX & participant-centric creative economy.
• As the most agile to the latest technologies in the Web3 space, CyberConnect has onboarded the most ERC4337-based smart contract wallets to see use cases of account abstraction, and integration with existing technologies such as DID & VC and ZKP has the great potential to bring further innovation.
• In addition, it is considered that CyberConnect is one of the protocols that is closest to Web3 adoption, with a broad ecosystem and the largest user base to actually transform technology into new experiences and share them.

Account Abstraction for Web3 Adoption

The Rise of Web3

The development of IT technology and the increasing time people spend in the online space have led to a variety of digital interactions. In particular, social media platforms first appeared only for social networking, but they have gradually penetrated into people’s lifestyles and now become powerful channels for marketing, information search, and other purposes. As a result, the top social media platforms have become even larger, creating new types of participants and markets called “Content Creators”, “Influencers”, and the “Creative Economy”. In this market, where it is important to ‘stand out’ from the crowd, identity has become a way to increase one’s own value and social impact, and Individual content has also become more than just a one-time-consumption item for interaction.

However, as the creative economy grew, concerns about its sustainability have increased at the same time. Centralized media platforms have used user data to maximize advertising revenue, and have also collected a significant portion of the revenue of content creators. Despite such a dismal incentive structure, content creators have been forced to continue to work with the huge media platforms, as they have no alternatives. Media platforms increasingly have prioritized one-way communication from brands to people, which has led to the spread of biased content and a decline in user interaction experience.

In order to restore the smooth interaction between people on social media and the creative economy to its original purpose, a new approach to the way participants interact with each other is needed, along with a new market structure that can be sustainable. And currently, Web3 is being mentioned as a promising alternative to this — By using blockchain technology, users can take full control of their data and content and decide how it is shared, without relying on any central platform. As the distribution path of the creative work can be transparently tracked, content creators can design a reasonable revenue structure for themselves. In addition, creators can inject properties into the creative work itself or issue their own tokens to redefine the way they interact or align the incentive structure with the community. In short, the existing creator economy was driven by a centralized and inefficient market, but Web3’s creator economy can finally be a symmetrical market centered on the entities that actually participate (i.e., creators and consumers).

Lack of Identity-Middleware In Web3

Nevertheless, it seems that adoption of Web3 still has a long way to go. There are several reasons for this, but the most intuitives are the inconvenience and lack of interactivity. If we think back to the early days of the Internet, we couldn’t really interact with many different actors, just received information from providers. However, with the introduction of identity models that could characterize individuals, a certain degree of mutual trust was established and the internet began to experience a wave of innovation embracing a wider range of interactions.

Like this, identity is a key enabler of interaction by allowing people to better recognize each other. Therefore, for a wide range of interactions to take place in the Web3 space, a holistic identity must be defined according to a Web3 grammar — However, there is no clear mechanism for doing so at the moment.

The inconvenience and risk from this lack of a holistic identity falls entirely on the end-users. In addition to being unable to interact with different web3 services, they must go through a series of complex processes required by those services in order to use them, and the accounts at the center of all of these must be securely managed.

We often refer to this as an ‘UX issue’, and it is a major bottleneck in the adoption of Web3 by the general public, even though Web3 is currently receiving a great deal of attention. Therefore, if we assume that something like an identity-middleware that can define a holistic identity could emerge and solve these inconveniences, it would have to have the following properties.

1. It should support permanent identity and guarantee different types of interactions with each application.
2. It should adopt a scalable and flexible storage structure that can support a variety of data required for defining identity.
3. The data included in the identity should be controlled in a voluntary and flexible way.

Account Abstraction Unlocks Identity-Middleware

‘Account Abstraction(AA)’, which is being actively discussed these days, can satisfy a significant portion of the properties mentioned above and can be a great help in building an identity-middleware in Web3 stack — The main idea of account abstraction is to convert existing wallet to ‘smart contract wallet’ by enabling the sending of transactions and signatures for Contract Accounts — In other words, existing accounts, which were only able to initiate and send transactions, can now be embedded with different features to make them as easy as managing any other Web2 identity accounts.

The functions here include not only advanced transaction logics, but also the control method of the account through key management service, and defining data linked to the account, etc. This means that the security logic and data required for identity can be integrated into the underlying account — or more simply, account abstraction allows us to inject a concept of identity into an existing account.

Specifically, smart contract wallets can support a variety of features, including the following, to ensure a diverse and flexible user experience.

Customized Signature Schemes

• Currently, only the ECDSA signature algorithm is used in EOA. However, if account abstraction is implemented, other signature algorithms such as Schnorr, BLS, or quantum-resistant alternatives can be used in addition to ECDSA.

Flexibility in Paying Gas Fees

• Flexibility to user-service interactions can be added by delegating the payment of gas fees to other people (or applications) or allowing users to pay gas fees with a variety of tokens.

Batch Transaction

• Costs in the execution of contracts can be saved by implementing logic to process multiple transactions at once.

Signature Abstraction

• Custom logic can be implemented to make verification valid under certain conditions — This allows users to utilize whitelist/blacklist management to restrict transactions to specific targets, time-locked features to set the lifetime of signatures, and logic such as requiring multiple signatures for transactions over a certain amount of money.

Authentication & Key Management

• Users can freely add, remove, and change permissions for multiple accounts, and also use MFA(multi-factor authentication) and Session Keys to improve the existing transaction experience.

and more.

ERC4337 with Brief History of AA

In fact, the concept of account abstraction has been around for a while. Ethereum’s standard proposal “Abstraction of transaction origin and signature (EIP 86)” in 2016 was the first step towards account abstraction. Since then, there have been many other proposals that have further developed the concept, and the keyword “Account Abstraction” has been born — Of the different standards candidates, the relatively new ERC 4337 proposal has become the most talked about and advanced as it implements account abstraction without the need to modify the consensus layer. And leading up to Vitalik’s presentation on the importance of this implementation aspect, account abstraction and ERC4337 became more popular together, and a significant number of projects are now moving to adopt ERC4337.

Below is a brief process of a smart contract wallet implemented with ERC4337.

1. Users send a smart contract called UserOperation to the UserOperation Mempool through an smart contract wallet — Here, UserOperation can include various logics beyond simple token transfer, and UserOperation Mempool is a separate mempool different from the existing one that the network usually uses to process transactions.
2. Bundlers bundle the highest paying UserOperations and package them into a Bundle Transaction.
3. Bundlers take this Bundle Transaction and call a smart contract called an EntryPoint — Bundlers simulate the validity of each UserOperation’s signature and ability to pay for gas, and submit the Bundle Transaction along with the fees included in the Bundle Transaction to the EntryPoint.
4. (Optional) Users can optionally utilize a feature called Paymaster for other entities to pay for gas on their behalf, or to pay for gas in other currencies, such as ERC20 tokens.
5. EntryPoint is responsible for validating each UserOperation and executing it on the network — Whereas the bundler simply simulates UserOperations, the EntryPoint actually validates UserOperations. If the validation is successful, bundlers are paid a predetermined fee by the entities who generated UserOperations, and if there are any gas fees remaining after the execution, they are also refunded to the bundler.

To summarize, account abstraction is to bring the existing principle of smart contracts running on the network to the application level, configure specific logics by each account(steps 1–4 above), and verify & execute all of these logics with a contract called EntryPoint on the network(step 5 above).

The flexibility of ERC4337 allows smart contract wallet to be combined with a variety of existing technologies, in addition to the functions mentioned above — For example, it can strengthen identity or data privacy by using DID(Decentralized Identifier) & VC(Verified Credential), and ZKP (Zero-Knowledge Proof)*. As such, ERC4337 has seen a renewed interest with a wide range of projects quickly adopting or experimenting with it since the code audit by OpenZeppelin,

*To be covered later in this article

CyberConnect, Web3 Social Middleware with AA

CyberConnect is leading the Web3 space by being the first to adopt ERC4337 and are experimenting combining it with other existing technologies such as ZKP. CyberConnect is maximizing the use cases for smart contract wallets by implementing the full range of ERC4337’s features and agilely applying them to its broader user base and ecosystem.

Primer on CyberConnect

CyberConnect is a Web3 Social middleware project where various social-based apps can be built on top of the middleware that CyberConnect provides. It is preparing for the V3 update at the moment, including the application of ERC4337. With this update, CyberConnect envisions a social network that allows users to create a chain-agnostic account and centers that identity around a Web2-like user experience and participant-driven creative economy — Creators and consumers can move around different platforms with the holistic identity, realize a content-based community economy with the best user experience, and developers get all these data to build a more prosperous ecosystem.

Three Pillars of CyberConnect

CyberConnect has three core elements for all of these — 1) Identity, which is the center of all interactions 2) Social graph, which represents each entity’s connections 3) Network, which supports data solution and smooth communication

CyberAccount

CyberAccount is the core identity for creators and consumers in the CyberConnect ecosystem. Its compatibility with ERC4337 and ERC6900 enables users to enjoy an enhanced user experience when interacting with applications, as they can use the functions of account abstraction such as Paymaster, social recovery, permission management, relay transaction, etc. CyberAccount integrates data from Web3 as well as Web2-based social media platforms like Twitter and Discord. This gives users completely control over all their social data. There is also a special account for teams, organizations, and brands called ‘Organization Account’, which offers advanced account management features like multi-signature and customizable access control rules.

CyberAccounts can have separate handles called ‘CyberID’ — Since the CyberAccounts can store so much information, simple profiles showing only the necessary information are required for each interaction. cyberIDs are represented by ‘.cyber’ and are dedicated to interacting with applications with some identity information, such as basic personal information, username, avatar, metadata, etc. However, CyberIDs must renew their registration at regular intervals. If this is missed, the handles will be put up for public auction.

CyberGraph

CyberGraph is a social graph that receives metadata and activities of CyberAccounts, records them in storage, and represents the necessary data. In addition to identity information, the data flowing through CyberGraph includes (ERC721 based) collections, subscriptions, contents, W3ST(Web3-Status Token, ERC1155 based SBT) that defines the status of each user within a specific community, and other off-chain-managed information.

With traditional Web2 social media platforms, it is difficult for social interactions (or social contexts) to span multiple worlds without being connected to a centralized entity, forcing participants to either lock themselves into the platform or give up their identity and assets and start with a blank slate on a new platform. With CyberGraph, participants can continue to interact with the information they have accumulated through their CyberAccount in a variety of worlds with little or no switching costs between these platforms, and developers can support the expansion of the ecosystem through data APIs and various toolkits.

CyberNetwork

CyberConnect aims to encompass data from both Web2 and Web3 spaces(or off-chain and on-chain) and operate efficiently and seamlessly. CyberNetwork is a gas-efficient and scalable L2 network designed for the purpose of smooth sourcing of such data. Additional upgrades will integrate the network into CyberConnect in the coming months.

Comparison with Other Protocols

Protocols that are often compared to CyberConnect in the industry include Lens Protocol, Farcaster, and DeSo. While they commonly share problems for existing social media platforms and build user-owned profiles and social graphs, they have slightly different approaches and protocol designs. This article provides a brief comparison of the key features that influence each approach and discusses CyberConnect’s differentiation.

• All four protocols share a common mission of emphasizing scalable and flexible interactions through identity-based accounts, and currently only CyberConnect has fully implemented account abstraction through ERC4337 — CyberConnect was one of the earliest adopters of account abstraction as soon as the concept was discussed, and the largest number of ERC4337-based smart contract wallets in the industry are onboarded to use CyberConnect to create chain-agnostic interaction use cases across multiple chains(i.e., 450k+ as of August 16th). DeSo, on the other hand, does not directly adopt the standards around account abstraction, but instead introduces some of the functionality of account abstraction through ‘Derived Keys’, which can be used to authorize other entities to sign transactions. The remaining protocols are leveraging NFTs to define identity accounts, but have yet to fully implement account abstraction.
• One of the other most notable differences is the ‘Base Chain’. When a protocol becomes dependent on a specific chain, it can quickly absorb the chain’s user base, which is a benefit, but it also has drawbacks, such as the inability to customize infrastructure configurations and limited interoperability. While Lens and Farcaster rely on a single chain, CyberConnect supports multiple EVM-compatible chains such as Polygon, Linea, Arbitrum, and Base — Optimism, opBNB, and more will be supported soon. DeSo recognizes the limitations of implementing Web3 on a general-purpose chain, so it has built its own Layer 1 that specializes in Web3 Social graphs.
• ‘Data Storage’ is another area of significant difference. CyberConnect and Farcaster have taken a hybrid approach that leverages both on-chain (including decentralized storage solutions) and off-chain — CyberConnect stores identity and major interactions data on-chain and other interaction data off-chain solution. Farcaster, on the other hand, tries to store as little data on-chain as possible and utilize off-chain solutions for the rest. Other protocols store all data on-chain by default, but Lens stores some data other than identity on decentralized solutions like Arweave or IPFS, rather than all on the base chain. These strategic differences stem from tradeoffs between the value and convenience each protocol wants to provide to its users.
• CyberConnect and DeSo do not have a high barrier to entry for users to create accounts, while Farcaster and Lens restrict accounts (or profiles) to be obtained by default through invitations from specific users or special events. As a result, the number of registered accounts differs significantly between these protocols. Interestingly, this fact also seems to affect the nature of each community — CyberConnect’s community is heavily organized around sharing information about events, introducing and interacting with applications, while Lens is more artist & creator oriented, and Farcaster is more about participants having productive discussions. In the case of DeSo, it doesn’t seem to have any clear trend.

While some of the main differences are mentioned, the truth is that perceived UX of each protocol will vary from user to user, so it is best for them to try out each application and choose the protocol that works best for themselves.

• CyberConnect — Link3, Phaver, CyberTune, etc.
• Lens — Buttrfly, Orb, Phaver, etc.
• Farcaster — Jam, Warpcast, etc.
• DeSo — Diamond, etc.

The other most important point to consider is the macro direction of the project, specifically the direction in which the project is being refined — the infrastructure to build Web3 is still in its infancy, so comparing protocols in the current timeframe may be pointless. Therefore, rather than analyzing technical stacks, statistics, business models, etc. in the moment, it is much more important in the long run to judge how fast each protocol can adopt the latest technologies and be future-proofed in a flexible way.

From this perspective, CyberConnect is likely to be the most agile in that it quickly adopts the latest technologies, builds a diverse ecosystem to disseminate relevant experiences to as many users as possible, and enhances their core values(i.e., holistic identity) based on their feedback.

Ecosystem, Tokenomics, etc.

Ecosystem with Some Statistics

The combination of diverse interactions and synergies from the broader ecosystem can enhance the impact of identity and drive adoption by more users.

Having successfully raised two rounds of funding totaling $25M since launch, CyberConnect has now a broad ecosystem where the use cases for smart contract wallet with account abstraction can be fully realized — recording 1.4M+ account onboarding, 1.8M+ contents, 11M+ collections, and hovering around 60k+ WAUs and 400k+ MAUs (with some variation). In particular, as mentioned earlier, CyberAccount has achieved the largest adoption of 450k+ within about 3 weeks of its introduction, demonstrating its dominance in the industry for account abstraction-based smart contract wallets.

The applications that this large number of users can interact with within the CyberConnect ecosystem vary ranging from social media platforms, memberships, entertainment, NFT marketplaces/issuance platforms, etc. With over 50+ applications in the CyberConnect ecosystem, CyberAccount can be interacted with in different ways fully utilizing the features of the account abstraction.

The diversification of interactions in the CyberConnect ecosystem can also be seen in the graph above. Until the beginning of the year 2023, the interaction rate was dominated by Link3 — the 1st party community application created by the CyberConnect team that allows users to easily create and manage their profile, as well as engage in communities to share and learn different sources of information. However, since March, various applications such as CyberTune (a platform for music NFTs) and Atticc (a community-based NFT platform) have continued to emerge and grow on CyberConnect alongside Link3 on CyberConnect.

CyberAccount users can further explore CyberConnect’s ecosystem by participating in the cyberTrek’s rewards program or Link3’s FanClub events with different interactions on multiple applications.

Tokenomics of CYBER and CyberConnect’s Milestone

CyberConnect’s native token, CYBER, not only serves as the essential currency for users to make chain-agnostic interactions within CyberConnect’s ecosystem, but also functions as the primary voting right for the ecosystem to prosper.

Governance — CYBER token holders have voting rights to improve the CyberConnect protocol and can delegate their voting rights to other entities. The scope of governance is as follows.

• Change in payment methods and fee structures
• Adoption of new base chain
• Allocation of funds for ecosystem development, such as grants and community contribution programs
• Organizing online and offline events to revitalize the ecosystem
• Etc.

CyberAccount Gas Token — CYBER is used as a payment method for gas for all transaction types within CyberConnect’s multi-chain ecosystem.

Payment methods for CyberID

The most important update for CyberConnect in Q3 will be the launch of CyberDAO, which discusses how to improve multi-chain interactions. In addition, there is also a plan to continue CyberTrek, a collaboration with six chains (Ethereum, BNB, Optimism, Arbitrum, Polygon, and Base) to educate on the use cases of CyberAccount.

Q4 will see the release of the Login-SDK and CyberWallet app, and next year will see CyberNetwork and the launch of the grant program for developers.

Potential Opportunities with DID & VC, and ZK Proofs

There are several so-called ‘Game Changers’ in Web3 — DID(Decentralized Identifier) & VC(Verified Credential) and ZKP(Zero-Knowledge Proof) are some of them. DID & VC is a technology or specification that implements the concept of SSI (Self-Sovereign Identity) in a decentralized way, where users control their own identity information and ensure their privacy. ZKP is a cryptographic technique that can prove the validity of information without disclosing the details of the information.

They are attracting attention particularly in Web3, as their technology is well-suited to blockchain and enhances the autonomous nature of identity — Here, enhancing the autonomous nature of identity means that people can have more sovereignty over their identity information, and it allows users to express themselves better, eliminates the information imbalance between users by mitigating the data reliance on third parties, and thus enables different types of interactions based on greater trust.

with DID & VC

Currently, the environments where individuals can interact on-chain and off-chain are separated. However, if the DID & VC stack is connected to CyberAccount, it can provide a truly new user experience as the stack encompasses the entire digital and real world where individuals operate.

For example, if the government (or a high-trust entity) in the real world has a smart contract wallet such as CyberAccount and grants individuals citizenship (or corresponding means of expressing an individual’s status) on the blockchain as a VC, individuals can interact in compositive ways on the blockchain that they can do in the real world — We have already observed the explosive adoption of various types of interactions on the web since the introduction of individual identity models and federated identity models in the early days of the internet. Conversely, if on-chain activities are also stored in DID in the form of VC, individuals can use them to create a more dimensional representation of their identity, enabling them to engage in a variety of activities with a specific purpose in the real world. Ultimately, these cases can promote a great synergy between the real world and the Web3 space, as well as make each space more reliable.

with ZK Proofs

Although smart contract-based identity vehicles such as CyberAccount claim to be able to control their data and thus curate externally disclosed data, all on-chain data is public, which means, there is a risk of data exposure to unintended parties or being tracked of a series of on-chain activities. Moreover, in CyberGraph, an individual’s identity information includes not only static information, but also includes contextual or dynamic information such as claims and interactions. Therefore, by embedding ZKP into its infrastructure, CyberConnect will be able to provide users with a fully contextualized experience, where the social context they are exposed to is tailored to the specific application they are using. This whole new experience is previously impossible on existing social media platforms due to privacy concerns.

Or conversely, CyberAccount can drive adoption of ZKP. It is prohibitively expensive to validate ECDSA signatures, the conventional signature for EOAs, through ZKP. If smart contract wallets adopt ZKP-friendly signature schemes through account abstraction, these limitations can be overcome and the benefits of ZKP can be more fully utilized across the network.

To summarize, smart contract accounts can not only strengthen the holistic identity, but also have excellent external scalability as they can synergize with different existing technologies. In this regard, CyberConnect is the best place to see how such external scalability is implemented in practice and what interactions and use cases are created. Because it is the most agile protocol in adopting new technologies, having the largest smart contract wallet user base with a wide ecosystem.

Beyond Tech, Towards New Experience

As digital technology develops in the future, interactions between individuals online will become more sophisticated, and the importance of the creative economy may become even more important. As such, blockchain-based Web3 is gaining a lot of attention as an alternative for a sustainable creative economy in the future. In particular, the concept of smart contract wallet based on account abstraction can remove much of the complexity and limitations of existing blockchain-based accounts, help define the concept of web3 identity, and its synergy with other technologies reinforces the innovative natures of Web3.

However, even these innovative natures cannot truly realize their value unless they are eventually chosen by the public. If the adoption of technology is defined as the moment when a new experience spreads among people, then Web3 has spent a lot of time defining new experiences, but relatively little time discussing how they can spread.

We have seen that different interactions on social media platforms, such as Facebook(Meta), are being converted into new experiences and adopted by a large number of people. AI, also has long been considered a promising technology, but it was only with the GPT model that it was able to achieve widespread adoption so quickly thanks to the dissemination of a variety of experiences. Therefore, it is not convincing to simply attribute the gap between Web3 adoption and innovation to technical accessibility or UX issues, so that focusing on improving these is not the only way to achieve widespread adoption of Web3.

In that sense, CyberConnect is the closest protocol to achieve widespread adoption of Web3 — CyberConnect not only rapidly grasps and integrates cutting-edge technologies that can enhance the core value of Web3, but also transforms and shares these synergies into new experiences based on its largest user base and ecosystem. In other words, CyberConnect is the status quo for Web3, and the place where better technologies, values, and experiences are evangelized.

About Us

Four Pillars is a global crypto research firm based in Seoul, consisting of the most influential blockchain researchers in Korea. Through robust research and governance skills, it helps various market players easily onboard to the blockchain industry by offering high-quality research articles while supporting protocols in their expansion into Korean and global markets.

Four Pillars Links

• Website (EN/KO)
• Twitter (EN/KO)
• Telegram (KO)

CyberConnect, The Experience Hub for Web3 Adoption was originally published in Four Pillars on Medium, where people are continuing the conversation by highlighting and responding to this story.

면책 조항: 본 글은 일반적인 정보 제공을 목적으로 작성되었으며, 법률, 사업, 투자 또는 세금 관련 조언을 제공하지 않습니다. 본 글을 기반으로 투자 결정을 내리거나 이를 회계, 법률, 세무 지침으로 사용해서는 안됩니다. 특정 자산이나 증권에 대한 언급은 정보 제공의 목적일 뿐, 투자에 대한 추천이 아님을 밝힙니다. 본 글에서 표현된 의견은 저자의 개인적인 의견이며, 관련된 기관, 조직 또는 개인의 견해를 반영하지 않을 수 있습니다. 본 글에 반영된 의견은 사전고지 없이 변경될 수 있음을 유의하십시오.

Author: Jay : : FP (@JayLovesPotato)

Key Takeaways

• 웹3는 블록체인 기술을 기반으로 참여자들이 자신의 데이터와 콘텐츠를 주도적으로 통제하고 공유할 수 있는 참여자 중심의 대칭적인 시장을 만들 수 있는 잠재력을 가지고 있음에, 큰 주목을 받고 있다.
• 하지만 신원은 주체들간의 상호작용을 가능케하는 핵심 요소임에도 불구하고, 현재 웹3는 다양한 어플리케이션과 많은 종류의 상호작용을 할 수 있는 통합된(holistic) 신원이 부재하다.
• CyberConnect는 계정 추상화를 통해 통합된 신원을 정의하며 웹2와 같은 UX & 참여자 중심의 크리에이티브 경제를 구축하는 소셜 그래프를 만들고 있다.
• 웹3 영역에서 최신 기술에 가장 민첩하게 대응하는 CyberConnect는 가장 많은 ERC4337 기반의 스마트 컨트랙트 지갑을 온보딩시켰음에 계정 추상화의 실질적인 유스케이스 관찰이 기대되는 바이며, DID & VC, 그리고 ZKP 등 기존의 기술들과 통합은 또다른 혁신을 불러올 잠재력이 있다.
• 더불어, CyberConnect는 넓은 생태계와 최대 규모의 사용자층을 바탕으로 실질적으로 기술을 새로운 경험들 전환하고 공유함에, 웹3의 채택에 가장 근접한 프로토콜 중 하나라고 사료된다.

웹3 채택을 위한 계정 추상화

웹3의 부상

IT 기술이 발달하고 사람들이 온라인 공간에서 더욱 머무르게 되면서 우리는 다양한 형태의 디지털 상호작용을 관찰하게 되었다. 특히 소셜 미디어 플랫폼은 애초에 소셜 네트워킹을 위해 등장하였지만 점차 개개인의 라이프스타일에 깊이 침투하여 어느새 마케팅, 정보검색 등 거의 모든 종류의 상호작용이 공존하는 강력한 채널로써 자리잡게 되었다. 이에, 기존의 상위 소셜 미디어 플랫폼들은 더욱 거대해졌고 “컨텐츠 창작자”와 “인플루언서”, 그리고 “크리에이티브 경제” 라고 하는 새로운 유형의 참여자들과 시장이 출현하게 되는 계기를 만들어내었다. ‘돋보이는 것이 중요한’ 이 시장에서 신원은 곧 스스로의 가치를 높일 수 있는 수단이자 사회적 영향력(Social Impact)를 의미하게 되었고, 개별 컨텐츠까지도 단순히 상호작용을 위한 하나의 소비재 이상의 의미를 갖게 되었다.

하지만 이러한 시장이 점점 커짐에 따라 크리에이티브 경제는 그것의 지속가능성에 대해 점차 의문이 제기되었다. 중앙화된 미디어 플랫폼들은 사용자들의 정보를 활용하여 광고 수익을 극대화하는 한편, 컨텐츠 창작자로부터도 그들의 수익의 상당부분을 회수하였다. 그다지 효과적이지 않은 인센티브 구조임에도 불구하고, 컨텐츠 창작자들은 대안이 없었기에 거대해진 미디어 플랫폼 내에서의 정체성을 포기할 수 없었다. 사람들을 연결하는 것보다 브랜드와 사람을 연결하는 단방향적인 소통을 우선시하게 된 미디어 플랫폼들은 점차 편향된 컨텐츠로 채워지게 되며 사용자들의 상호작용 경험은 훼손되었다.

결국 소셜 미디어 상에서 이뤄지는 사람들 간의 원활한 상호작용 및 크리에이티브 경제를 본래의 목적대로 되돌리기 위해서는 참여자들 간의 상호작용 방식에 대해 새로운 접근법과 함께, 그것이 지속가능할 수 있는 새로운 시장구조의 필요성이 대두되었다. 그리고 현재, 웹3는 이에 대해 유력한 대안으로 언급되고 있다 — 블록체인을 기반함으로써 사용자들은 어떠한 중앙 기관에 대한 의존없이 자신의 데이터 혹은 컨텐츠를 주도적으로 통제하고 공유되는 방식을 결정할 수 있다. 이 때 창작물의 유통 경로는 투명하게 추적될 수 있음에, 컨텐츠 창작자는 자신에게 합리적인 수익구조를 설계할 수 있다. 또한 창작자는 창작물 자체에 속성을 주입하거나 자체적인 토큰을 발행하여 커뮤니티와 상호작용하는 방식을 새로이 정의하고 인센티브 구조를 일치시킬 수 있다. 요컨대, 기존의 크리에이터 경제가 플랫폼 중심의 비효율적인 시장으로 이루어졌다면, 웹3의 크리에이티브 경제는 비로소 실질적으로 사용하는 주체(i.e., 창작자와 소비자) 중심의 대칭적인 시장으로 이루어질 수 있는 것이다.

웹3 내 신원 미들웨어의 부재

그럼에도 불구하고, 웹3 시장에 대한 채택은 아직 갈 길이 멀어 보인다. 몇몇 이유가 있겠지만, 가장 직관적인 것은 불편함과 다양한 상호작용의 부재이다. 초창기 인터넷을 떠올려보면 우리는 인터넷을 통해서 주어진 정보를 받아볼 뿐, 여러 주체들과 그다지 다양한 방식의 상호작용을 하지 않았던 것 같다. 하지만 개개인을 식별할 수 있는 신원 모델이 도입되기 시작하면서 상호간에 신뢰라고 하는 것이 어느 정도 형성이 되었고, 인터넷은 다양한 상호작용을 수용하게 되며 변화의 물결을 맞이하기 시작하였다.

이처럼, 신원은 서로를 더욱 잘 식별하게 함으로써 상호작용을 가능케하는 핵심 요소이다. 따라서 웹3 공간에서도 다양한 상호작용이 만들어지기 위해서는 웹3 문법에 맞는 신원이 필수적으로 정의될 수 있어야 한다. 하지만 불행히도, 현재 웹3 공간에는 신원의 기능을 할 수 있는 매커니즘이 뚜렷하게 존재하지 않는다.

이러한 통합적인(holistic) 신원이 부재하다면, 그로 인해 야기되는 불편함과 리스크는 온전히 최종 사용자들의 몫이다. 다양한 상호작용은 할 수 없음은 물론이거니와, 이들이 여러 웹3 서비스들을 이용하기 위해서는 기본적으로 해당 서비스들이 요구하는 일련의 복잡한 과정들을 거쳐야하며, 이 모든 과정을 실행함에 있어 중심이 되는 계정 역시 안전하게 관리돼야한다.

우리는 종종 이를 ‘UX 문제’라고 표현하기도 하는데, 이는 현재 웹3가 지대한 관심을 받고 있음에도 불구하고 실질적으로 대중들에게 채택되는 데에 있어 병목이 되는 주요 요인으로 꼽히고 있기도 하다. 따라서 만일 통합된 신원을 정의하는 신원-미들웨어같은 것이 등장하여 이러한 불편함을 해결할 수 있다고 가정한다면, 그것은 다음과 같은 속성들을 가져야 할 것이다.

1. 영구적인 신원을 지원하며, 각 어플리케이션과 다양한 방식의 상호작용을 보장한다.
2. 신원의 정의에 필요한 다양한 데이터를 지원할 수 있는 확장가능하고 유연한 스토리지 구조를 채택한다.
3. 신원에 포함된 데이터들은 자발적이고 유연한 방식으로 제어되어야 한다.

신원 미들웨어의 구축에 기반이 될 수 있는 계정 추상화

최근 활발히 논의되고 있는 ‘계정 추상화(AA, Account Abstraction)’ 는 위 속성들 중 상당부분을 충족하며 기존 웹3 스택에서 신원-미들웨어의 구축에 큰 도움을 줄 수 있다 — 계정 추상화의 주요 아이디어는 블록체인 네트워크의 다양한 기능 수행을 담당하는 컨트랙트 계정도 EOA처럼 트랜잭션 발행과 서명을 가능케함으로써, 기존의 지갑들을 ‘스마트 컨트랙트 지갑’으로 전환하는 데에 있다. 간단히 말해서, 트랜잭션을 발행하고 보내는 주체에 불과했던 기존의 계정이 이제 다양한 기능들을 내장할 수 있게 되면서 여느 웹2의 신원 계정을 관리하는 것 만큼이나 그 사용성이 증대될 수 있는 것이다.

여기서 기능들이라 함은, 단순히 고도화된 거래 로직을 포함하는 것 뿐만 아니라 키관리를 통한 계정의 제어방식, 계정과 연동되는 데이터의 정의 등을 포함한다. 이는 곧, 신원에 필요한 데이터 및 보안 로직이 기본 계정에 적용할 수 있다는 것을 의미하며, 또 다른 말로는 계정 추상화를 통해 웹3 공간에서도 기존의 계정에 신원의 개념을 불어넣을 수 있다는 것이다.

구체적으로, 스마트 컨트랙트 지갑은 아래를 포함한 여러 기능들을 지원하며 사용자의 다양하고 유연한 상호작용을 보장한다.

사용자 지정 서명 방법

• 현재 EOA에서는 ECDSA 서명 알고리즘만 사용이 가능하다. 하지만 계정 추상화를 구현한다면 ECDSA뿐만 아니라 슈노르(Schnorr), BLS, 혹은 양자컴퓨팅에 대비한 다른 대안을 서명 알고리듬을 사용할 수 있다.

가스비 결제의 유연성

• 가스비의 지불 의무를 다른 사람(혹은 어플리케이션)에게 위임하거나 다양한 토큰으로 가스비를 결제할 수 있게 함으로써, 사용자-서비스 간의 상호작용에 유연성을 더한다.

배치 트랜잭션

• 여러 트랜잭션을 한번에 처리하는 로직을 구현하여 컨트랙트를 실행함에 있어 비용을 절약한다.

서명 추상화

• 특정 조건에 따라 검증이 유효하도록 사용자 지정의 로직을 구현할 수 있다. 이를 통해 사용자는 특정 대상에 대해 거래를 제한하는 화이트리스트 / 블랙리스트 관리, 서명의 수명을 설정하는 Time-locked 기능, 일정금액 이상의 트랜잭션을 수행하고자 할 때는 다중 서명을 의무화하는 등의 로직 등을 활용할 수 있다.

권한 및 키관리

• 사용자는 여러 계정 별로 권한을 자유로이 추가/제거/변경할 수 있으며, 다중인증(MFA, Multi-Factor Authentication), 그리고 세션키(Session Key)를 도입하여 기존의 트랜잭션 경험을 완화할 수 있다.

기타

ERC4337

사실, 계정 추상화에 대한 개념이 최근에 갑자기 주목을 받게 된 건 아니다. 해당 이슈는 2016년 ‘Abstraction of transaction origin and signature’ 라는 제목의 이더리움 표준 제안 (EIP 86) 이후 많은 제안들을 거듭하며 계정 추상화라는 키워드를 탄생케 하였다 — 다양한 후보들 중에서도 특히 비교적 최근에 제안된 ERC4337가 합의 계층을 수정할 필요 없이 계정 추상화를 구현할 수 있기 때문에 가장 많이 언급이 되며 발전하게 되었다. 그리고 이러한 구현 측면의 중요성을 주제로 하는 비탈릭의 발표까지 이어지며 계정 추상화와 ERC4337는 함께 더욱 널리 알려졌고, 현재 점점 더 많은 프로젝트들이 ERC4337을 채택하려는 움직임을 보이고 있다.

ERC4337로 구현한 스마트 컨트랙트 지갑의 간략한 프로세스는 아래와 같다.

1. 사용자가 스마트 컨트랙트 지갑을 통해 일반 트랜잭션이 아닌 유저오퍼레이션(UserOperation)라고 하는 별도의 스마트 컨트랙트를 유저오퍼레이션 멤풀(UserOperation Mempool)에 보낸다 — 여기서, 유저오퍼레이션에는 토큰 전송 이상의 다양한 로직들이 포함될 수 있으며, 유저오퍼레이션 멤풀은 일반적으로 네트워크가 트랜잭션들을 처리하기 위해 사용하는 멤풀과는 다른, 별도의 멤풀이다.
2. 번들러(Bundler)라고 하는 주체들은 수수료를 가장 많이 지불하는 유저오퍼레이션을 묶어 번들 트랜잭션(Bundle Transaction)으로 패키징한다.
3. 번들러는 이 번들 트랜잭션을 가지고 엔트리포인트(EntryPoint)라고 하는 스마트 컨트랙트를 호출한다 — 이때 번들러는 각 유저오퍼레이션의 서명 및 가스비 지불 능력이 유효한 지 시뮬레이션하고 번들 트랜잭션에 포함된 수수료를 엔트리포인트로 함께 제출한다.
4. (Optional) 페이마스터(Paymaster)라고 하는 기능을 선택적으로 활용하여 다른 주체로하여금 가스비를 대납하거나 ERC20 토큰 등 다른 통화로 가스비를 지불할 수 있도록 상호작용한다.
5. 엔트리포인트는 각 유저오퍼레이션을 검증하고 네트워크에서 실행하는 역할을 맡는다. — 번들러가 유저오퍼레이션을 단순히 시뮬레이션하기만 하였다면, 엔트리포인트는 실질적으로 유저오퍼레이션을 검증한다. 검증이 성공하면 유저오퍼레이션의 실행주체(즉, 개별 이용자)가 미리 기재한 수수료는 번들러에게 납부되며, 만일 실행까지 마치고나서 남은 가스비가 있다면 번들러에게 환불된다.

요컨대, 계정 추상화란 기존의 스마트 컨트랙트가 네트워크에서 실행되는 기존의 원리를 어플리케이션 단으로 그대로 가져와 특정 로직들을 구성하고(위 1~4 과정) 엔트리포인트라고 하는 컨트랙트가 이 모든 로직들을 검증하고 네트워크 상에 실행시키는 것이다(위 5 과정).

ERC4337이 가지는 유연성으로 인하여 스마트 컨트랙트 지갑은 위에서 언급한 기능이외에도 다양한 기존 기술들과 함께 결합될 수 있다 — (본 아티클의 후반부에 더욱 살펴볼 것이지만) 예를 들어, DID(Decentralized Identifier) & VC(Verified Credential), 그리고 ZKP(Zero-Knowledge Proof)와 함께 신원 혹은 데이터 프라이버시를 강화할 수 있다. 이에, ERC4337은 Openzeppelin으로부터의 코드 감사 이후 실로 다양한 범위의 프로젝트들에 의해 빠르게 적용 혹은 실험되면서 다시금 큰 주목을 받고 있다.

CyberConnect, AA를 통해 구현한 웹3 소셜 미들웨어

웹3 영역에서도 가장 먼저 ERC4337을 채택하고 ZKP 등의 기술들을 함께 결합하는 실험을 하며 새로운 미래를 그리고 있는 프로젝트가 있었으니, 바로 CyberConnect이다. CyberConnect는 ERC4337을 통해 구현할 수 있는 기능들을 최대한 구현하고, 그것들을 넓은 사용자층과 생태계에 민첩하게 적용해보며 스마트 컨트랙트 지갑의 유스케이스를 극대화하고 있다.

CyberConnect 개요

ERC4337 적용 등을 포함하는 V3 업데이트를 앞둔 CyberConnect는, 자체적으로 제공하는 소셜 그래프를 통해 다양한 소셜 기반 어플리케이션이 구축될 수 있는 생태계를 조성하는 것을 목표로하는 웹3 소셜 미들웨어 프로젝트이다. 이번 업데이트를 통해 CyberConnect는 사용자들로하여금 Chain-Agnostic한 신원 계정을 생성케하고, 해당 신원을 중심으로 웹2와 비교하여 손색이 없는 사용자 경험 및 참여자 중심의 크리에이티브 경제가 갖춰진 소셜 네트워크의 구축을 꿈꾼다 — 창작자 및 소비자들은 통합된 신원으로 다양한 플랫폼을 돌아다니며 최상의 사용자 경험과 함께 컨텐츠 기반의 커뮤니티 경제를 실현하고, 개발자는 이 모든 데이터들을 가져와 생태계를 더욱 풍요롭게 빌딩할 수 있다.

CyberConnect의 3가지 핵심요소

CyberConnect 상에서 이 모든 것을 가능하게 하는 데에는 3가지 핵심요소가 있다 — 1) 상호작용의 중심이 되는 신원 2) 정보를 표현하는 소셜 그래프 3) 원활한 통신 및 데이터 솔루션을 지원하는 네트워크

CyberAccount

CyberAccount는 창작자 및 소비자들이 CyberConnect 생태계에서 활동하는데에 가장 핵심이 되는 신원이다. ERC4337와 함께 ERC6900이 호환될 수 있도록 하였기 때문에 사용자들은 어플리케이션들과 상호작용을 함에 있어 앞서 언급하였던 페이마스터, 소셜복구, 권한 설정, 릴레이 트랜잭션 등 계정 추상화의 기능들을 실질적으로 활용하며 극대화된 사용자 경험을 누릴 수 있다. CyberAccount는 웹3 뿐만 아니라 트위터나 디스코드와 같은 웹2 기반의 소셜 미디어 플랫폼의 데이터까지 통합한다 — 이에, 사용자는 자신의 모든 소셜 데이터를 완벽히 통제할 수 있다. 조직 전용 계정(Organization Account)이라고하는 팀, 조직, 브랜드를 특별한 계정 또한 존재하는데, 이 계정은 다중 서명 및 사용자 지정 가능한 액세스 제어 규칙과 같은 고급 계정 관리 기능을 제공한다.

이 CyberAccount는 CyberID라는 별도의 핸들을 구성할 수도 있다 — CyberAccount가 워낙에 많은 정보를 저장하고 있으니, 구독 등의 상호작용을 위해서는 필요한 정보만을 보여주는 간단한 프로필이 필요하다. CyberID는 ‘.cyber’로 표기되며 기본 개인 정보, 사용자 이름, 아바타 및 메타데이터 등 신원 정보를 일부 저장하여 어플리케이션과의 상호작용을 전담할 수 있다. 단, cyberID는 일정 주기마다 그것의 등록을 연장해야 한다. 만일 이를 놓치면 해당 핸들은 공개경매절차로 넘어가게 된다.

CyberGraph

CyberGraph는 CyberAccount의 메타데이터 및 활동들을 전달받아 스토리지에 기록하고 필요한 데이터를 표현하는 소셜 그래프이다. CyberGraph에서 흐르는 데이터는 신원 정보 외에도 (ERC721 기반으로 구현된) 창작자의 수익화에 필요한 수집, 구독정보 및 컨텐츠, 특정 커뮤니티 내에서 각 사용자의 상태를 정의하는 ERC1155 기반의 SBT인 W3ST(Web3-Status Token), 그 외 오프체인 상에서 관리되는 팔로우 정보 등을 포함한다.

기존의 웹2 소셜 미디어 플랫폼의 경우, 중앙화된 주체와 연결이 되어 있지 않으면 소셜 상호작용 (혹은 소셜 컨텍스트)이 다양한 세계관으로 이어지긴 어려웠다. 이에, 참여자들은 플랫폼에 종속되거나 신원을 비롯한 자산을 포기하고 새로운 플랫폼에서 백지 상태로 시작하는 수 밖에 없었다. CyberGraph를 활용하면 참여자들은 이러한 플랫폼의 전환비용을 거의 들이지 않고 다양한 세계에서 CyberAccount를 통해 축적한 정보를 바탕으로 상호작용을 이어갈 수 있으며, 개발자들 역시 데이터 API 및 다양한 툴킷을 통해 폭넓은 생태계 확장을 지원할 수 있다.

CyberNetwork

CyberConnect는 웹2와 웹3(혹은 오프체인과 온체인) 공간의 데이터를 포괄하며 효율적이고 원활하게 동작해야한다. 이에 CyberNetwork는 해당 데이터들의 원활한 소싱을 위한 목적으로 고안중인, 가스효율적이고 확장성있는 L2 네트워크이다. 추후 몇 달 내로 추가 업그레이드를 통해 해당 네트워크가 CyberConnect에 통합될 예정이다.

다른 프로토콜과의 비교

CyberConnect와 업계에서 자주 비교되는 프로토콜은 Lens Protocol, Farcaster, 그리고 DeSo 가 있다. 이들은 기존의 소셜 미디어 플랫폼에 대해 각각 비슷한 문제점을 인식하고 사용자 소유(User-Owned)의 프로필 및 소셜 그래프를 구축하지만 문제 해결 방식 및 프로토콜 디자인이 조금씩 상이하다. 이번 아티클에서는 문제 해결 방식의 차이에 영향을 주는 주요 특징들만 간단히 비교해보고, CyberConnect가 가지는 차별점에 대해 더욱 깊이 이해해본다.

• 네 프로토콜은 모두 공통적으로 신원 기반의 계정을 통해 확장가능하고 유연한 상호작용을 중시하는 미션을 공유하고 있는데, 현재 ERC4337을 통해 스마트 컨트랙트 지갑을 온전히 구현한 프로토콜은 CyberConnect가 유일하다 — CyberConnect는 계정 추상화의 개념이 논의되자마자 가장 빠르게 도입한 프로토콜 중 하나로, 산업 최대 규모의 ERC4337 기반 스마트 컨트랙트 계정들이 온보딩되어 다양한 체인들 위에서 chain-agnostic한 상호작용 유스케이스들을 만들어나가고 있다 (i.e., 8월 16일 기준 450k+). 반면 DeSo의 경우, 계정 추상화와 관련된 표준을 직접적으로 채택하는 대신에 트랜잭션 서명의 권한을 다른 주체에게 부여할 수 있는 Derived Key를 통해 계정 추상화 기능의 일부를 도입한다. 나머지 프로토콜들은 NFT를 활용하여 신원 계정을 정의하고 있는데 아직까지 계정 추상화를 완전히 구현하진 않았다.
• 또 하나 가장 두드러지는 차이점 중 하나는 ‘Base Chain’ 이다. 프로토콜이 특정 체인에 의존하게되면 해당 체인의 사용자층을 빠르게 흡수할 수 있다는 장점이 있지만, 인프라 구성을 커스텀할 수 없으며 상호운용성에 제약이 존재할 수 있다는 단점 역시 존재한다. Lens와 Farcaster는 단일 체인에 의존적인 반면, CyberConnect는 EVM 호환이 되는 체인에 한해서 점점 더 많은 채택을 하려고 계획중이다 — CyberConnect는 이미 Polygon, Linea, Arbitrum, 그리고 Base를 지원하고 있으며 Optimism, opBNB이 곧 지원될 예정이다. DeSo는 일반적인 목적의 체인으로 웹3를 구현하는 데에는 한계가 있다고 지적하며, 아예 웹3 소셜 그래프에 특화된 자체 체인을 구성하였다.
• ‘Data Storage’ 영역도 큰 차이를 보인다. CyberConnect와 Farcaster는 (탈중앙화 스토리지 솔루션을 포함한) 온체인과 오프체인을 모두 활용하는 하이브리드 방식을 택하였다 — CyberConnect는 신원 및 수익화에 필요한 정보는 온체인에, 그 외 더욱 다양한 상호작용 데이터까지 오프체인 솔루션에 저장하는 방식을 취한다. 반면에 Farcaster는 최대한 적은 정보를 온체인에 저장하고, 나머지는 오프체인 솔루션을 활용하려는 방식을 취한다. 나머지 프로토콜들은 기본적으로 모든 정보를 온체인에 저장을 하는데 Lens는 신원 외 일부 데이터는 베이스 체인이 아닌, Arweave나 IPFS와 같은 탈중앙화 솔루션에 저장하기도 한다. 이러한 전략들의 차이는 각 사용자에게 제공하려는 가치와 편의 사이에서 발생하는 트레이드오프에서 비롯된다.
• CyberConnect와 DeSo는 사용자들로하여금 계정을 만드는 데에 진입장벽이 높지 않은 반면, Farcaster와 Lens는 계정(혹은 프로필)을 기본적으로 특정 사용자의 초대나 특별 이벤트를 통해서만 얻을 수 있게 제한한다. 따라서 해당 프로토콜들 간 등록된 계정의 수는 현저히 차이가 난다. 흥미로운 점은, 이 사실이 각 커뮤니티에 흐르는 정보들의 성격에도 영향을 미치는 것처럼 보인다는 것이다 — CyberConnect의 커뮤니티는 이벤트 정보, 어플리케이션들과의 상호작용 및 소개에 관한 것들이 가득 차 있는 반면, Lens는 조금 더 아티스트 & 크리에이터 위주의, Farcaster는 조금 더 생산적인 토론을 하는 참여자들 위주의 커뮤니티로 구성되어 있다. DeSo의 경우, 뚜렷한 경향은 관찰되지 않는다.

여러가지 주요 차이점들을 언급하긴하였지만 사실 사용자들이 각 프로토콜마다 체감하는 UX의 정도는 모두 다르기 때문에 각 사용자들은 어플리케이션을 직접 활용해보고 자신에게 맞는 프로토콜을 선택하는게 가장 정확하다.

• CyberConnect — Link3, Phaver, CyberTune, etc.
• Lens — Buttrfly, Orb, Phaver, etc.
• Farcaster — Jam, Warpcast, etc.
• DeSo — Diamond, etc.

이와 더불어 가장 유의미하게 생각해야할 또 다른 포인트는 프로젝트의 거시적인 방향성, 즉 프로젝트를 다듬어가는 방향이다 — 아직 웹3를 구축하는 인프라 자체가 초기단계이기 때문에 현 상태에서 각 프로토콜을 비교하는 것은 무의미할 수 있다. 따라서 지금 당장의 기술 스택, 통계치, 비즈니스 모델 등을 분석하는 것보다는 되려, 프로토콜들이 얼마나 기술들을 빠르게 채택하여 유연하게 미래에 대비할 수 있는 지 가늠하는 것이 장기적으로 훨씬 중요하다.

이러한 관점에서는 가장 빠르게 최신 기술을 수용하고, 생태계를 다양하게 구성하여 관련 경험들을 최대한 많은 사용자들에게 전파하고, 이들로부터의 피드백을 바탕으로 프로토콜의 핵심가치를 강화해나가는 CyberConnect가 상대적으로 가장 민첩하고 할 수 있을 것이다.

생태계, 토크노믹스 및 기타

생태계 및 통계치 일부

광범위한 생태계로부터 형성되는 시너지와 다양한 상호작용의 조합은, 신원의 임팩트를 향상시키는 동시에 더욱 많은 사용자들의 채택을 촉진시킬 수 있다.

21년 출시이후 두 차례에 걸쳐 총 $25M 상당의 펀드 레이징을 성공적으로 마친 CyberConnect는 현재 계정 추상화를 활용한 스마트 컨트랙트 지갑의 유스케이스가 최대한으로 실현될 수 있는 광범위한 생태계를 가지고 있다 — 1.4M+의 계정 온보딩, 1.8M+의 컨텐츠, 11M+의 콜렉션(혹은 상호작용)을 기록하였고, (편차가 존재하지만) 현재는 대략 60k+ WAU, 400k+ MAU 수준의 사용자 수를 유지하고 있다. 특히, 앞서 언급하였듯 CyberAccount는 도입된 지 약 3주만에 약 450k+ 로 업계 최대의 규모의 채택을 이뤄내며 계정 추상화 기반의 스마트 컨트랙트 지갑에 대한 독보적인 우위를 입증하였다.

이렇게 많은 사용자들이 CyberConnect 생태계 내에서 상호작용할 수 있는 어플리케이션은 소셜 미디어 플랫폼, 멤버십, 엔터테인먼트, NFT 마켓플레이스/발행 플랫폼 등 매우 다양하다. 50+가 넘는 어플리케이션으로 이루어진 CyberConnect 생태계에서 CyberAccount는 계정 추상화의 기능들을 최대한으로 활용하며 다양한 상호작용을 할 수 있다.

CyberConnect 생태계 내에서 상호작용이 다각화되어가고 있는 것은 위 그래프를 통해서도 확인할 수 있다. 2023년 초창기까지는 Link3 의 상호작용 비율이 지배적이었다 — Link3는 CyberConnect 팀에 의해 만들어진 1st Party 커뮤니티 어플리케이션으로, 사용자들은 프로필을 간편하게 만들고 관리할 수 있을 뿐만 아니라 커뮤니티에 참여하여 다양한 정보들을 공유하고 배울 수 있다. 하지만 3월 이후에는 CyberTune(; 음악 NFT를 위한 플랫폼)과 Atticc(; 커뮤니티 기반의 NFT 플랫폼) 등과 같은 다양한 어플리케이션들이 계속해서 등장하며 Link3와 함께 CyberConnect에서 성장하고 있다.

프로필 사용자는 계정을 통해 여러 어플리케이션들과의 다양한 상호작용을 퀘스트로 제시하는 cyberTrek 보상 프로그램 혹은 Link3의 FanClub 이벤트를 통해 CyberConnect의 생태계를 더욱 풍부하게 탐험할 수 있다.

CyberConnect 생태계 내에서 상호작용이 다각화되어가고 있는 것은 위 그래프를 통해서도 확인할 수 있다. 2023년 초창기까지는 Link3 의 상호작용 비율이 지배적이었다 — Link3는 CyberConnect 팀에 의해 만들어진 1st Party 커뮤니티 어플리케이션으로, 사용자들은 프로필을 간편하게 만들고 관리할 수 있을 뿐만 아니라 커뮤니티에 참여하여 다양한 정보들을 공유하고 배울 수 있다. 하지만 3월 이후에는 CyberTune(; 음악 NFT를 위한 플랫폼)과 Atticc(; 커뮤니티 기반의 NFT 플랫폼) 등과 같은 다양한 어플리케이션들이 계속해서 등장하며 Link3와 함께 CyberConnect에서 성장하고 있다.

프로필 사용자는 계정을 통해 여러 어플리케이션들과의 다양한 상호작용을 퀘스트로 제시하는 cyberTrek 보상 프로그램 혹은 Link3의 FanClub 이벤트를 통해 CyberConnect의 생태계를 더욱 풍부하게 탐험할 수 있다.

CYBER의 토크노믹스 및 마일스톤

CyberConnect의 네이티브 토큰인 CYBER는 사용자가 CyberConnect의 생태계 내에서 chain-agnostic한 상호작용을 하기 위한 필수적인 통화 역할을 할 뿐만 아니라, 생태계를 번영시키기 위한 주요 의결권으로써도 기능을 한다.

거버넌스 — CYBER 토큰 보유자는 CyberConnect 프로토콜 개선을 위한 의결권을 가지며, 다른 주체에게 자신의 투표권을 위임할 수 있다. 토큰 보유자가 행할 수 있는 거버넌스의 범위는 아래와 같다.

• 지불수단 및 수수료 구조 변경
• 새로운 프로토콜 채택
• 그랜트 및 커뮤니티 기여 프로그램 등의 생태계 발전 기금 할당
• 생태계 활성화를 위한 온/오프라인 이벤트
• 기타

CyberAccount 가스 토큰 — CyberConnect의 멀티체인 생태계 내에서 모든 거래 유형에 대한 가스비 지불 수단으로 사용된다.

CyberID에 대한 지불수단

오는 3분기에서 CyberConnect의 가장 중요한 업데이트는 멀티체인 상호작용 방식의 개선을 지속적으로 논의할 수 있는 CyberDAO의 런칭이다. 이 외에도, 6개 체인(Ethereum, BNB, Optimism, Arbitrum, Polygon, 그리고 Base)과의 협력을 통해 CyberAccount의 유스케이스를 교육하는 CyberTrek 을 지속적으로 운영할 계획도 있다.

4분기에는 로그인을 위한 SDK 및 CyberWallet 어플리케이션 출시를, 그리고 내년에는 개발자를 위한 그랜트 프로그램 및 CyberNetwork를 계획하며 CyberConnect의 기술적 UX를 강화한다.

DID & VC, 그리고 ZK Proof 와의 시너지

한편, 웹3에서는 소위 ‘게임 체인저’ 라 불리우는 개념이 몇 가지가 있다 — DID(Decentralized Identifier) & VC(Verified Credential), 그리고 영지식증명(ZKP, Zero-Knowledge Proof)이 바로 그것들 중 일부이다. DID & VC는 사용자가 스스로 자신의 신원 정보들을 제어하고 프라이버시를 보장하는 자기주권신원(SSI, Self-Soverign Identity)의 개념을 탈중앙화된 방식으로 구현한 기술 혹은 스펙이며, ZKP는 정보의 상세 내용을 공개하지 않고 정보의 유효성을 증명할 수 있는 암호학적 기법이다.

이들이 특히 웹3 상에서 큰 주목을 받는 이유는 그것의 기술을 구현함에 있어서 블록체인과의 궁합이 매우 잘맞고, 이를 통해 신원의 자주적인 속성을 강화하기 때문이다 — 여기서 신원의 자주적인 속성이 강화된다는 것의 의미는 사람들이 더욱 신원 정보에 대한 주권을 가질 수 있다는 뜻이다. 이는 자기 자신을 더욱 잘 표현할 수 있게 하고, 제3자에 대한 데이터 의존도를 낮춤으로 인해 서로간의 정보의 불균형을 제거하며, 따라서 더욱 큰 신뢰를 바탕으로 다양한 방식의 상호작용들을 가능케한다.

with DID & VC

현재 온체인과 오프체인 위에서 개인이 각각 상호작용할 수 있는 환경은 분리되어 있다. 그런데 DID & VC 스택은 개인이 활동하는 디지털 및 현실 세계 전체를 포괄하는 개념이기 때문에 이 스택들이 CyberAccount에 연결된다면 진정으로 새로운 UX를 제공할 수 있다.

예를 들어, 현실 세계에서 정부(혹은 신뢰도가 높은 주체)가 CyberAccount와 같은 스마트 컨트랙트 지갑을 가지고 있어서 개인에게 시민권(혹은 이에 상응하는 개인의 지위 표현 수단들)을 온체인 상에 VC로 발급해준다면, 개인은 온체인 공간에서도 현실 세계에서 할 수 있는 복합적인 상호작용이 가능하다 — 우리는 이미 초창기 인터넷에 개별 신원 모델 및 연합 신원 모델이 도입된 이후 인터넷에서 다양한 종류의 상호작용들이 폭발적으로 수용되는 과정을 관찰했다. 반대로, 온체인 상의 활동들 역시 VC의 형태로 DID에 저장이 된다면, 이를 활용하여 개개인의 신원은 더욱 입체적으로 표현될 수 있음에 현실 세계에서도 특수한 목적을 가진 다양한 활동들을 하는 데에 도움을 줄 수 있다. 결국 이러한 사례들은 현실세계와 웹3 공간 간의 큰 시너지를 촉진할 수 있을 뿐 아니라, 각 공간 자체를 더욱 신뢰할 수 있는 공간으로 만든다.

with ZK Proofs

아무리 CyberAccount와 같은 스마트 컨트랙트 기반의 신원 수단이 데이터를 스스로 제어할 수 있어서 외부로 공개되는 데이터도 선별이 가능하다고하지만, 온체인 데이터는 모두 공개되어 있으므로 의도하지 않은 주체에게 데이터가 노출되거나 일련의 온체인 활동들이 추적될 위험이 있다. 더욱이 CyberGraph 내에서, 개인의 신원 정보는 정적인 정보도 포함하지만 클레임 혹은 상호작용 정보 등과 같이 컨텍스트 기반의 동적인 정보도 포함한다. 따라서 ZKP가 CyberConnect의 인프라에 함께 접목된다면, 사용자들은 프라이버시 이슈없이 어플리케이션마다 노출될 소셜 컨텍스트 설정을 달리하며 완벽한 컨텍스트 기반의 경험(Contextual Experience)을 할 수 있게 된다. 이는 기존의 소셜미디어도 프라이버시 이슈 때문에 달성하지 못한 사용자 경험이다.

혹은 역으로, CyberAccount가 ZKP의 채택을 촉진할 수도 있다. 현재 ZKP를 통해 기존 EOA의 서명방식인 ECDSA 서명을 검증하는 것은 비용이 매우 많이 든다. 하지만 계정 추상화를 통하여 CyberAccount와 같은 스마트 컨트랙트 지갑들이 ZKP 친화적인 서명 체계를 채택한다면 이러한 한계점을 극복하고 네트워크에는 ZKP의 이점들이 더욱 적극적으로 활용될 수 있다.

요컨대, 스마트 컨트랙트 계정은 블록체인 네트워크에 통합된 신원을 더욱 강화할 뿐만 아니라, 기존의 다양한 기술들과 시너지를 일으킬 수 있음에 외부 확장성 역시 매우 뛰어나다고 할 수 있다. 그런점에서, 특히 CyberConnect는 최신 기술을 수용함에 있어 가장 민첩하고, 가장 큰 규모의 스마트 컨트랙트 지갑 사용자층 및 넓은 생태계를 가졌기 때문에 그러한 외부확장성을 실제로 이행하고, 어떠한 상호작용과 유스케이스들이 만들어질 수 있는 지 관찰하기에 가장 적합한 공간이라 할 수 있다.

기술을 넘어, 새로운 경험을 향해

앞으로 디지털 기술이 발달함에 따라 온라인 상에서 개인 간의 상호작용은 더욱 고도화 될 것이고, 크리에이티브 경제의 중요성은 더욱 부각될 수 있다. 이에, 블록체인을 기반한 웹3는 향후 지속가능한 크리에이티브 경제를 위한 대체제로써 많은 주목을 받고 있다. 특히 계정 추상화 기반의 스마트 컨트랙트 지갑은 기존의 블록체인 기반 계정의 복잡성 및 한계를 상당부분 제거하고 웹3 신원의 개념을 정립하는데에 도움을 줄 수 있으며, 다른 기술들과의 시너지는 웹3만의 혁신적인 특성을 강화한다.

하지만 이러한 혁신적인 특성들도 결국 대중들로부터 선택받지 못하면 그 가치가 진정으로 발휘될 수 없다. 기술의 채택을 사람들 간 새로운 경험이 확산되는 순간이라고 정의한다면, 웹3에서는 그간 새로운 경험을 정의하는 데에 노력을 들인 것에 비해 ‘그 경험이 확산될 수 있는 방법론’에 대한 고민은 상대적으로 부족해왔다.

우리는 기존의 Facebook(Meta)과 같은 소셜미디어에서 다양한 상호작용이 새로운 경험으로 전환될 때 해당 미디어들이 대중들에 의해 대량으로 채택되는 것을 관찰하였다. AI 역시 수십년이 넘도록 잠재적으로 혁신적인 개념으로 인식되었지만, GPT 모델을 통해 단번에 다양한 경험이 전파되며 대량 채택이 이루어지는 것을 관찰하였다. 따라서 웹3 채택도 혁신과의 괴리를 단순히 기술의 접근성 혹은 UX 문제만으로 치부하는 것은 설득력이 떨어지며 이들의 개선에 집중하는 것이 채택의 유일한 돌파구는 아닌 것이다.

그런 점에서 CyberConnect는 웹3의 채택을 실현하기에 가장 근접한 프로토콜이라 할 수 있다 — CyberConnect는 웹3 가치를 강화할 수 있는 최신 기술들을 빠르게 채택하여 집약될 수 있도록 할 뿐만 아니라, 넓은 생태계와 최대 규모의 사용자층을 바탕으로 해당 기술들의 시너지를 실질적으로 새로운 경험들로 전환하고 공유한다. 즉, CyberConnect는 웹3에 대한 현재 상황(Status Quo) 그 자체이자 더 나은 기술, 가치, 그리고 경험이 새로이 전파되는 공간인 것이다.

About Us

Four Pillars is a global crypto research firm based in Seoul, consisting of the most influential blockchain researchers in Korea. Through robust research and governance skills, it helps various market players easily onboard to the blockchain industry by offering high-quality research articles while supporting protocols in their expansion into Korean and global markets.

Four Pillars Links

• Website (EN/KO)
• Twitter (EN/KO)
• Telegram (KO)

CyberConnect, 웹3 채택을 향한 경험의 허브 was originally published in Four Pillars on Medium, where people are continuing the conversation by highlighting and responding to this story.

Disclaimer: This article is intended for general information purposes only and does not constitute legal, business, investment, or tax advice. It should not be used as a basis for making any investment decisions or relied upon for accounting, legal, or tax guidance. References to specific assets or securities are for illustrative purposes only and do not represent recommendations or endorsements. The opinions expressed in this article are those of the author and do not necessarily reflect the views of any affiliated institutions, organizations, or individuals. The opinions reflected herein are subject to change without being updated.

Author: Steve : : FP(Steve_4P)

Key Takeaways

• Following the Terra and FTX incidents, the blockchain industry has undergone a market reformation centered around Ethereum and rollup networks.
• Amidst this market situation, Sei Network claims a fast monolithic Layer 1 chain.
• Ultimately, for a blockchain to succeed, it needs to be easily onboarded by a large number of users, requiring scalability and chain stability.
• The Sei Network has developed the Twin Turbo Consensus to achieve the fastest finality in the world while being stable.
• Now, with the mainnet launch just around the corner, there is keen interest in seeing the performance that the Sei Network can deliver.

1. Latest Trend in Blockchain Industry, but Sei.

The mainnet for SEI Network, set to become one of the fastest blockchains in the world and a leading example of monolithic blockchain, is right around the corner. SEI Network, whose mainnet schedule has been delayed until now, announced on August 1, 2023, that it would open a launch pool for the SEI Network on Binance, the world’s largest exchange. Furthermore, Binance revealed on August 15th that it would list SEI, the token of the SEI Network, officially confirming that the mainnet launch of the SEI Network is not far off. As the mainnet launch of the SEI Network approaches, let’s take a brief look at the SEI Network that was previously covered by Four Pillars, and understand the philosophy of the SEI Network and the recent accomplishments of the network.

1.1 The Era of Rollups

Following the Terra and FTX incidents, the blockchain industry has undergone a market reformation centered around Ethereum and rollup networks. As Terra and Solana, which positioned themselves as Ethereum competitors, have either collapsed or severely faltered, the sentiment that “there’s no longer a need for new Layer 1 blockchains” has become the dominant opinion. This market situation has only further strengthened the already robust Ethereum ecosystem, and with more power being funneled into the rollup-centric multi-chain ecosystem adopted by Ethereum, the period from 2022 to 2023 can rightly be called the “Era of Rollups”. During this time, rollup chains, Layer 2 solutions, and several modular blockchains have received significant attention. Below are a few representative projects among the numerous rollups:

1.1.1 Optimism and OP Stack

The OP Stack received significant attention when it was revealed that Coinbase, one of the largest exchanges in the world along with Binance, had implemented its next blockchain, the BASE Chain, through OP Stack. OP Stack is an open-source code that provides the ability to build Optimistic Rollups. By using it, rollup chains can be very easily created. It can be seen as a kind of Software Development Kit (SDK) for Optimistic Rollups. Recently, Worldcoin, co-founded by Sam Altman who is known as the father of Chat GPT, also launched on the Optimism mainnet, indicating their support for OP Stack’s multi-chain vision. As these two renowned companies, Coinbase and Worldcoin, actively utilize Optimism and OP Stack, the demand for constructing Optimistic Rollup chains with OP Stack continues to grow. (For more details, please read Xpara’s OP-Stack article).

1.1.2 ZK Sync and ZK Stack

While Optimism is a rollup that builds Optimistic Rollups, which validate transactions through fraud proofs, ZK Sync builds ZK Rollups, which validate transactions through validity proofs. Even though they are both rollups, there is a clear difference in their methodologies. Currently, as a result of its technical advancements over a long period of time, ZK Sync has not only become the third largest Layer 2 rollup in the Ethereum ecosystem but also announced ZK Stack, similar to Optimism’s OP Stack, in June 2023, thereby entering the SDK war. Of course, there haven’t been notable use cases yet, since it isn’t launched yet. However, the fact that there is now another alternative to OP Stack for those looking to build rollups indicates how intense the rollup market could become in the future. (For more details, please read Xpara’s ZK-sync article).

Aside from these two projects, the blockchain market was filled with discussions about rollups, Layer 2, and modular blockchains as various infrastructures, such as RaaS (Rollup as a Service), were developed to simplify the creation of rollups.So, is building Layer 2 now the definitive solution, and has attempting to create a new Layer 1 become a meaningless endeavor? It seems like at least the SEI Network might not think so.

1.2 SEI’s Counter-Betting, Monolithic L1 chain: Why Didn’t They Choose Layer 2?

In any industry, when a particular technology gains attention, there is inevitably a concentration of talent towards that technology. However, it is also necessary to think critically before blindly accepting these technologies. Jay’s opinion that constructing rollups is not an immediate solution to the problems currently faced by blockchain technology is a statement worth contemplating by those working in the industry. Perhaps, in some ways, rollups might be a term abused to cunningly deceive those who are repulsed by the term ‘Layer 1 blockchain’.

Of course, rollups and Layer 1 blockchains are not exactly the same in all aspects. But depending on the types of applications and assets being built, Layer 1 and rollups might have similar attributes. Sicne all assets issued on Optimistic rollups are not protected by Ethereum, if the measure of chain’s success is “many users should be able to use it without inconvenience,” building Layer 1 appears as valuable as building rollups from the perspective of the SEI Network.

It’s said that Sei Labs initially considered building the Sei Network as a Layer 2 on Ethereum. However, they opted for Layer 1 due to the clear limitations in scalability and the issue of a single sequencer inherent in Layer 2 rollups.

Wait a minute, an issue with scalability? Weren’t rollups introduced as a solution to the inherent scalability problems Ethereum faced?

While proponents of modular blockchains saw rollups as a solution to scalability issues, from Sei Labs’ perspective, rollups remained an ambiguous alternative. After all, rollups still need to record processed data onto Ethereum’s blocks. And since Ethereum blocks don’t have infinite space, there’s a clear upper limit to the number of transactions they can handle.

1.2.1 Let’s Calculate the Maximum TPS (Transactions Per Second) That a Rollup Can Handle.

The target size for an Ethereum block is approximately 15 million gas. To validate 1 byte of data in Ethereum, 16 gas is required. This means that Ethereum verifies and processes roughly 937,500 bytes of data (15,000,000/16) per block. If we assume a very ‘simple’ rollup transaction uses 12 bytes when batched for submission to Ethereum (this number varies by rollup and by transaction type, but this estimate is based on Vitalik’s calculation for an ETH Transfer executed within a rollup), then the number of rollup transactions an Ethereum block can handle is 78,125 (937,500/12). Given that Ethereum’s block time is around 13–15 seconds, dividing 78,125 by 13–15 seconds results in a maximum of 6,000 TPS and a minimum of 5,208 TPS. However, this calculation assumes that Ethereum’s block space is used exclusively for rollups. Let’s not forget that the Ethereum blockchain can also operate as a monolithic blockchain in its own right. Applications like Uniswap and Opensea are currently operating on the Ethereum base layer. This means that rollups are also in competition with Ethereum native apps for block space. Therefore, it’s highly likely that the upper limit for TPS that rollups can achieve is less than 5,000. Even with the introduction of Proto Dank Sharding, the TPS upper limit wouldn’t dramatically increase. While Dank Sharding might offer a significant increase, its implementation would take considerable time.

For these reasons, Sei chose to build a Layer 1 solution with its own consensus mechanism and block time, rather than relying on rollups(and ETH or Celestia). Of course, for many users to use it seamlessly, the network must be exceptionally fast while still ensuring the security inherent in a blockchain. So, how exactly did Sei construct its network, and just how fast is it?

2. Sei Network In a Nutshell

The SEI Network is often viewed as a Layer 1 blockchain that specializes in trading, but it is actually a General Purpose Blockchain. While SEI has its own “order matching engine,” it is more accurate to describe the SEI Network as a General Purpose Blockchain with enhanced trading infrastructure. This is because trading assets is considered one of the fundamental features of blockchain technology. Moreover, in addition to its internally enhanced trading infrastructure, SEI boasts an impressive speed that surpasses many other blockchains in terms of ‘time to finality’. Therefore, it is more significant to examine the consensus mechanism that SEI uses to achieve this speed.

2.1 Twin-Turbo Consensus: What Makes Sei Fast?

Twin Turbo Consensus can be understood as a consensus mechanism equipped with two turbos, which are the Intelligent Block Propagation and Optimistic Block Processing that SEI Network emphasizes. What roles do these play and how do they increase SEI’s speed?

2.1.1 Intelligent Block Propagation

In traditional blockchains, block propagation involves two steps:

1. When a user initiates a transaction, the node receives and verifies the transaction, adds it to its mempool, and propagates this transaction to other nodes.
2. The verified transaction is included in a block by the leader (a mining node in PoW, a block proposer in BFT). This block is then propagated across the entire network. Only when all nodes receive this block can the transaction be considered successfully processed.

However, this traditional method is inefficient because the same transactions are transmitted twice: first during transaction propagation and then again during block propagation. This results in a waste of network bandwidth.

To address this problem, the SEI Network introduces a mechanism where the block proposer propagates a proposal containing the hash values of the transactions and a Block ID that references the complete block, rather than the actual transaction data. A hash value is a summary of the original transaction data processed through a hash function, which is smaller in size. The block proposer first propagates the block proposal to the network and then gradually propagates the complete block in parts.

Through this efficient propagation process, nodes can reconstruct their local mempools to form the block without having to wait for the complete block to arrive if they verify that they hold the transactions corresponding to the hash values. If any transactions are missing, they wait until the entire block arrives to process the transaction.

SEI has successfully improved network performance by about 40% using this block propagation process. (Additionally, SEI has reduced finality to 210ms through Intelligent Block Propagation, as mentioned later.)

2.1.2 Optimistic Block Processing

Before SEI’s Optimistic Processing, consensus on the block had to be established before the block could be processed. However, SEI starts processing as soon as it receives a block, handling both Prevote and Precommit simultaneously. This structure might seem potentially risky, but as the Prevote and Precommit processes are also carried out during the block processing, if the block is deemed invalid during this process, it can simply be discarded and the process restarted. Therefore, it’s not a significant problem. SEI Network has not omitted the traditional consensus process, but rather parallelizes these processes.

SEI has successfully improved network performance by approximately 33% through Optimistic Block Processing.

2.2 Transaction Parallelization

SEI also performs transaction parallel processing along with the Twin Turbo Consensus, and the important thing here is to find out which transactions can be processed in parallel. Not all transactions can be processed in parallel, as there are clearly transactions that affect each other. Ultimately, to enable safe transaction parallel processing, it’s very important to find transactions with potential conflicts. SEI pre-identifies the relationship between transactions through a Directed Acyclic Graph (DAG), and if these transactions are confirmed to be independent of each other, they process the transactions in parallel. Through this, SEI has become capable of handling 20,000 orders per second.

3. Still Hungry For Speed: How Sei Additionally Reduces Finality Time.

3.1 What Is Finality? Why Is It Important?

Many people often consider TPS (Transactions per Second) when calculating the performance and scalability of a blockchain, but in fact, TPS can be a rather ambiguous metric. Here’s an example to explain why. Let’s assume that Blockchain A can contain a total of 600,000 transactions in one block. But if it takes 1 minute for Blockchain A to create a block, then the TPS ( = number of tx in a block / block time in seconds) would be 10,000, but in reality, since it takes 1 minute for 600,000 transactions to be included in the network, it cannot be seen as processing 10,000 transactions per second in real time. TPS is merely a figure dividing the blockchain’s processing capacity by seconds, so the apparent TPS can feel different to users.

My example may not be extreme, but even if you increase the number of transactions per block drastically and extend the block time to 1 hour, the TPS can still be extremely high (although users will have to wait an hour for the transaction to be included in the network). For this reason, these days when discussing the scalability of a blockchain, it seems that the TTF (Time to Finality), a metric that indicates the time it takes for a transaction to be confirmed, is used more than the TPS. Especially for Layer 1 chains like the SEI network, which focus on financial transactions, it’s extremely important for transactions to be processed instantly and reflected in the network, so it would have been important to make the TTF as fast as possible.

3.2 Compare Sei With Other Blockchains

Through the Twin Turbo Consensus explained above (especially through intelligent block propagation), the SEI network successfully reduced the finality to about 250ms. This is the fastest finality among existing blockchains, and can practically be considered as near-instant finality as perceived by users. Of course, this is finality in a very positive situation and the most stable finality time when multiple attempts were made on the SEI testnet, Atlantic-2, was 410ms. They succeeded in reducing finality to 250ms, but certain nodes had difficulty maintaining the network, and infrastructure players like RPC nodes were lagging behind. This is inevitable, as by nature, the blockchain nodes are scattered all over the world, so there can certainly be difficulties in propagation and communication. So how can all these nodes be included while reducing finality?

3.3 Sei Optimization: Fastest Even In The Most Conservative Scenarios

To solve the above problem, the SEI Network added approximately 100ms to the commit timeout during the consensus process.

environment). The commit timeout period could be shorter in the future depending on the network conditions, but for now, in this testnet, the added timeout was 100ms. With this, they were able to achieve a faster finality of 380ms compared to the stable 410ms they could achieve without the timeout. What’s interesting is that even with a 100ms timeout, SEI has implemented the fastest finality among existing blockchains. Depending on network conditions in the future and adjustments to the added timeout, SEI’s finality could decrease further. The fact that SEI’s finality, implemented in a conservative environment, is faster than the finality of other existing blockchains is very intriguing.

4. The Future of SEI

Actually, the launch of the SEI mainnet, which is nearly confirmed, has been continuously delayed over the past year and has only just been finalized. As I am not affiliated with the SEI Foundation, I can’t precisely explain why the SEI mainnet was delayed, but one fact I learned during my research on SEI is that SEI has continually strived to improve network performance. It seems they kept delaying the mainnet launch to achieve their ideal finality. Of course, reducing the finality time is essential, but what’s more crucial is the extent to which the network operates in a stable environment. SEI has focused on network improvements for the past year, postponing the mainnet to ensure both speed and stability.

Of course, after the mainnet comes out, it can be considered a different environment from the testnet, with many external variables, so it’s impossible to confirm that the performance achieved in the testnet can be carried over directly. However, as the testnet environment was set as conservatively as possible, there is a high likelihood of implementing fast finality immediately after the mainnet launch.

High stability and speed — if the SEI network can achieve these two, it can create an environment where many users can use the blockchain without inconvenience. Ultimately, whether it’s Layer 1 or Layer 2, a large number of users need to use it for it to succeed, so the launch of SEI’s mainnet is of great interest to participants in the blockchain industry. As someone who prefers the approach of monolithic blockchains more than modular blockchains, I sincerely hope that SEI will succeed and bring a fresh start to the Layer 1 competition.

About Us

Four Pillars is a global crypto research firm based in Seoul, consisting of the most influential blockchain researchers in Korea. Through robust research and governance skills, it helps various market players easily onboard to the blockchain industry by offering high-quality research articles while supporting protocols in their expansion into Korean and global markets.

Four Pillars Links

• Website (EN/KO)
• Twitter (EN/KO)
• Telegram (KO)

Sei: How does it ensure peak performance even in the most challenging scenarios? was originally published in Four Pillars on Medium, where people are continuing the conversation by highlighting and responding to this story.

면책 조항: 본 글은 일반적인 정보 제공을 목적으로 작성되었으며, 법률, 사업, 투자 또는 세금 관련 조언을 제공하지 않습니다. 본 글을 기반으로 투자 결정을 내리거나 이를 회계, 법률, 세무 지침으로 사용해서는 안됩니다. 특정 자산이나 증권에 대한 언급은 정보 제공의 목적일 뿐, 투자에 대한 추천이 아님을 밝힙니다. 본 글에서 표현된 의견은 저자의 개인적인 의견이며, 관련된 기관, 조직 또는 개인의 견해를 반영하지 않을 수 있습니다. 본 글에 반영된 의견은 사전고지 없이 변경될 수 있음을 유의하십시오.

Author: Steve : : FP(Steve_4P)

Key Takeaways

• 테라 사태와 FTX 사태 이후로 블록체인 시장은 이더리움과 롤업 체인들을 중심으로 시장이 개편되는 움직임을 보였다.
• 그러한 시장의 상황속에서 세이 네트워크는 빠른 모놀리틱 레이어1 체인을 표방한다.
• 결국 블록체인이 성공하기 위해서는, 많은 유저들이 쉽게 온보딩할 수 있어야 하는데 그러기 위해서 필요한 것은 확장성과 체인의 안정성이다.
• 세이 네트워크는 이 두 마리 토끼를 잡기 위해서 Twin Turbo Consensus를 개발하여 안정적이면서도 세상에서 가장 빠른 파이널리티를 구현해냈다.
• 이제 메인넷 런칭이 코 앞으로 다가온 세이 네트워크가 어떤 퍼포먼스를 보여줄 수 있을지 그 귀추가 매우 주목된다.

1. 블록체인 업계의 트렌드, 그리고 세이가 이들과 다른 점.

세상에서 가장 빠른 블록체인 중 하나가 될, 앞으로 모놀리틱 블록체인을 대표할 세이 네트워크의 메인넷이 정말 코앞으로 다가왔다. 여태까지 메인넷 일정이 밀려오던 세이 네트워크는, 2023년 8월 1일, 전세계 최대 규모의 거래소인 바이낸스에서 세이 네트워크에 대한 런치풀을 오픈할 것이라고 밝혔고 오는 8월15일엔 바이낸스가 세이 네트워크의 토큰인 SEI를 상장하겠다고 밝히면서 세이 네트워크의 메인넷 런칭은 이제 얼마 남지 않은 것으로 공식 확인되었다. 세이 네트워크의 메인넷 런칭이 다가온 만큼, 포필러스가 예전에 다뤘었던 세이 네트워크에 대해서 간략하게 알아보고, 세이 네트워크가 가지고 있는 철학과 최근에 이뤄낸 네트워크의 성과들까지 알아보도록 하자.

1.1 바야흐로 롤업의 시대

테라 사태와 FTX 사태 이후, 블록체인 업계는 이더리움, 그리고 롤업 네트워크 중심으로 시장이 개편되었다. 이더리움의 경쟁자 포지션을 내세웠던 테라와 솔라나가 망하거나 크게 흔들리면서 “이제 더 이상 새로운 레이어1 블록체인은 의미가 없다.”는 의견이 주를 이뤘다. 이러한 시장의 상황은 원래도 견고했던 이더리움 생태계를 더욱더 강화시켰고, 이더리움이 채택한 롤업 중심의 멀티체인 생태계에 더 많은 힘이 실리면서, 2022년부터 2023년은 말 그대로 “롤업의 시대”라고 말해도 될만큼 롤업 체인들과 L2 그리고 몇몇의 모듈러 블록체인들이 큰 주목을 받았다. 아래는 수많은 롤업중에서 이들을 대표하는 프로젝트를 가져와봤다:

1.1.1 Optimism and OP Stack

바이낸스와 함께 세계 최대 규모의 거래소인 코인베이스가 차제 블록체인인 BASE 체인을 OP Stack을 통해 구현했다는 사실이 밝혀지면서 큰 주목을 받았던 OP Stack은 옵티미스틱 롤업을 구축할 수 있도록 제공되는 오픈소스 코드로서 이를 활용하면 롤업 체인을 매우 간단하게 만들 수 있다. 일종에 옵티미스틱 롤업의 SDK라고 볼 수 있는 것이다. 최근엔 Chat GPT의 아버지라고 불리는 샘 알트만이 공동으로 창립한 월드코인도 옵티미즘 메인넷에 런칭하면서 OP Stack의 멀티체인 비전을 지지하는 것으로 알려졌다. 코인베이스와 월드코인이라는 두 유명 기업이 Optimism과 OP Stack을 적극 활용하면서 OP Stack으로 옵티미스틱 롤업 체인을 구축하려는 수요는 계속해서 늘어나고 있는 추세이다. (더 자세한 글은 Xpara의 OP-Stack글을 참고하시라)

1.1.2 ZK Sync and ZK Stack

옵티미즘은 옵티미스틱 롤업(사기 증명으로 트랜잭션을 검증하는)을 구축하는 롤업인 반면, ZK sync는 유효성 증명을 통해 트랜잭션을 검증하는 ZK 롤업을 구축하기 때문에 같은 롤업이라고 할지라도 방법론에서의 차이가 엄연히 존재한다. 현재 ZK Sync는 오랜 시간에 걸쳐 기술을 발전시킨 결과, 이더리움 생태계 전체에서 세번째로 큰 레이어2 롤업이 되었을 뿐만 아니라, 2023년 6월에 ZK Sync도 Optimism의 OP Stack과 비슷한 ZK Stack를 발표하면서 SDK 전쟁에 뛰어들었다. 물론 Op Stack에 비해서 늦게 출시되는 것이기 때문에 아직까지 이렇다할 사용사례는 없지만, 앞으로 롤업을 구축하는 입장에선 OP Stack말고도 또 다른 대안이 생겼다라는 점에서 앞으로 롤업 시장이 얼마나 치열해질지 가늠할 수 있는 부분이다. (더 자세한 글은 Xpara의 ZK-sync 글을 참고하시라)

이 두 프로젝트를 제외하고도, RaaS(Rollup as a service) 와 같이 롤업을 쉽게 만들어주는 다양한 인프라들이 개발되면서 블록체인 시장은 한동안 롤업과 레이어2 그리고 모듈러 블록체인에 대한 논의로 가득차있었다. 이제 정말 레이어2를 만드는 것이 정답인 것이고, 새로운 레이어1은 무의미한 시도가 된 것일까? 아마 적어도 세이 네트워크는 그렇게 생각하지 않고 있는 거 같다.

1.2 세이의 역베팅, 모놀리틱한 단일체인 구성: 왜 레이어2를 선택하지 않았는가?

어느 업계던간에 특정 기술이 주목을 받으면 그 기술로 인재들의 쏠림 현상이 일어나는 것은 어쩔 수 없는 거 같다. 하지만 해당 기술들을 무조건적으로 받아들이기 이전에, 비판적인 사고를 해보는 것 역시나 필요하다. 롤업을 구성하는 것이 지금 블록체인 기술이 마주한 문제들을 단번에 해결해주는 해결책이 아니라는 Jay의 의견은 업계에 종사하고 있는 사람들이라면 한 번씩 곱씹어볼만한 주장이다. 어쩌면 롤업은 ‘레이어1 블록체인’이라는 단어에 반감을 가진 사람들을 교묘하게 속이기 위해서 악용되고있는 단어일지도 모른다. 물론 롤업과 레이어1 블록체인이 모든 부분에서 똑같다고 주장하는 것은 아니다. 하지만 자신이 구축하는 어플리케이션과 자산의 종류에 따라서 레이어1과 롤업은 비슷한 속성을 가지고 있을지도 모른다. 옵티미스틱 롤업에서 발행된 모든 자산들이 이더리움에 의해 보호되고 있지 않은 것처럼, 결국 롤업이던 레이어1이던 “많은 유저들이 불편함 없이 사용해야하는 것”이 성공의 척도라고 한다면, 세이 네트워크의 입장에선 아직 레이어1을 구축하는 것이 롤업을 구축하는 것만큼이나 가치있는 시도라고 보여지는 것이다.

실제로 Sei Labs는 처음에 세이 네트워크를 이더리움의 레이어2로 구축할지에 대한 고민도 있었다고 한다. 하지만 레이어2 롤업이 가지고 있는 단일 시퀀서의 문제나, 확장성의 한계가 명확하게 존재하고 있기 때문에 레이어1을 선택한 것이다. 잠시만, 확장성의 문제라니. 롤업은 이더리움이 고질적으로 가지고 있었던 확장성의 문제를 해결하기 위해서 등장한 대안이 아니었던가?

물론 모듈러 블록체인을 옹호하는 집단들은, 롤업이 확장성 문제를 해결했다고 보았지만, Sei Labs의 입장에선 롤업은 여전히 애매한 대안이었다. 결국 롤업도 연산한 데이터를 그 베이스 레이어인 이더리움의 블록에 기록해야하고, 이더리움의 블록은 무한대의 공간을 가지고 있지 않기 때문에 트랜잭션을 소화하는 것에 대한 명확한 상한선이 존재한다.

1.2.1 한 번 롤업이 가져갈 수 있는 최대의 TPS를 계산해볼까?

이더리움 블록 크기의 타깃은 약 1500만 Gas이고, 이더리움에서 1byte의 데이터를 검증하기 위해선 16gas가 필요하다. 즉 이더리움은 한 블록당 대략 937,500bytes의 데이터(15,000,000/16)를 검증해서 처리한다는 이야기다. 만약 아주 ‘간단한’ 롤업 트랜잭션이 이더리움에 배치로 제출될 때 사용하는 용량을 12bytes라고 했을 때(이 숫자는 롤업마다 다르고 트랜잭션의 종류마다 변동되지만, 비탈릭이 계산했던 ETH Transfer 를 롤업에서 실행했을 때 데이터 용량을 기준으로 하였다) 이더리움 블록당 처리할 수 있는 롤업 트랜잭션의 수는 78,125개다(937,500/12). 그리고 이더리움의 블록타임은 약 13~15초니까 78,125에서 13~15초를 나누면? 최대 6,000 tps에서 최소 5,208 tps가 나온다. 하지만 이러한 계산은 이더리움의 블록 공간을 오직 롤업에만 사용했다는 가정이다. 이더리움 블록체인은 그 자체로도 모놀리틱 블록체인의 역할을 수행할 수 있음을 잊지 말자. 지금도 Uniswap이나 Opensea같은 어플리케이션이 이더리움 베이스레이어에서 작동된다. 즉, 롤업은 이더리움 네이티브 앱들과도 블록 공간에 대한 경쟁을 해야한다는 뜻이다. 그렇다고 한다면 롤업이 가져갈 수 있는 TPS의 상한선은 5,000미만일 가능성이 높다. 프로토 댕크 샤딩이 도입된다고 하더라도 TPS의 상한선은 드라마틱하게 증가하지 않을 것이고, 댕크 샤딩이 도입된다면 꽤나 유의미한 증가가 생길 수 있겠지만 시간이 오래 걸린다.

이러한 이유로 세이는 롤업 대신 자체적인 컨센서스와 블록타임을 가지고 있는 레이어1을 구축하게 된 것이다. 물론, 많은 유저들이 불편함 없이 사용하기 위해선 네트워크의 속도가 굉장히 빨라야하면서도, 블록체인으로서의 보안이 보장되어야만한다. 그렇다면 과연 세이는 어떤 방식으로 네트워크를 구축했고, 얼마나 빠를까?

2. 세이 네트워크 소개: 세이 네트워크란.

세이 네트워크는 트레이딩에 특화되어있는 레이어1 블록체인이라고 볼 수 있지만, 엄밀히 말해서는 General Purpose Blockchain이다. 세이가 “자체 오더 매칭 엔진”을 내장시켰다고 하더라도, 블록체인이 가지는 가장 큰 특징이 “자산을 거래할 수 있게 만든다”는 점에서 가장 기본적인 인프라라고 생각하기 때문에 세이 네트워크는 트레이딩 인프라를 강화한 General Purpose Blockchain으로 불리는 것이 더 맞다고 생각하고 있다. 또한, 자체적으로 강화한 트레이딩 인프라를 제외하고도 파이널리티 측면에서 이미 다른 블록체인들과 비교했을 때 압도적인 속도를 가지고 있기 때문에 해당 속도를 어떻게 달성했는지에 대해서 살펴보는 것이 좀 더 의미가 있을 것이라고 생각했다. 세이는 어떠한 컨센서스를 사용하고 있을까?

2.1 Twin-Turbo Consensus: what makes Sei fast?

트윈 터보 컨센서스는 말 그대로 두 가지 터보를 장착한 컨센서스 메커니즘으로 이해할 수 있는데 여기서 세이 네트워크가 내세우는 터보는 Intelligent Block Propagation과 Optimistic Block Processing이다. 이들은 각각 어떠한 역할을 수행하고 어떻게 세이의 속도를 끌어올리는 것일까?

2.1.1 Intelligent Block Propagation

기존 블록체인에선 블록을 전파할 때는 약 두 번의 과정을 거친다: 1)우선 유저가 트랜잭션을 발생시키면 노드가 트랜잭션을 받아서 검증한 후 해당 노드의 멤풀(mempool)에 추가하고, 다른 노드들에게 해당 트랜잭션을 전파하는 과정을 거친다. 2) 이렇게 검증된 트랜잭션은 나중에 리더(PoW의 경우는 마이닝 노드일 것이고, BFT에선 블록 제안자가 될 것이다)에 의해서 블록에 담기게 되는데, 이 블록 역시나 네트워크 전체에 전파해서 모든 노드들이 해당 블록을 받아내야 비로소 트랜잭션이 성공적으로 처리가 되었다고 할 수 있다. 이러한 기존의 방식은 매우 비효율적이라고 할 수 있는데, 그 이유는 이미 노드들이 트랜잭션을 전파하는 과정에서 자신들의 로컬 멤풀에 대부분의 트랜잭션을 가지고 있을 것이 분명한데, 이 트랜잭션이 담긴 블록을 또 전파해서 공유해야하는 상황이기 때문이다. 이러한 이중 전파의 문제(Double Transmission Problem)은 네트워크의 대역폭을 낭비하고 있는 것이라고 볼 수 있다.

해서, 세이 네트워크는 블록 제안자가 블록 제안서(Proposal)에 트랜잭션 데이터가 아닌, 1) 트랜잭션들의 해시값과, 2) 온전한 블록의 레퍼런스인 Block ID를 담아서 전파한다. 트랜잭션의 해시값은 기존 트랜잭션 데이터를 해시함수로 요약한 값이기 때문에 용량이 작다는 장점이 있다. 블록 제안자는 아래 그림과 같이 처음에 블록 제안서를 네트워크에 전파하고, 그 후에 온전한 블록을 조금씩 나누어서 전파한다.

이런식으로 전파의 효율적인 과정을 통해 블록 제안자가 블록 제안서를 전파하였을 때, 다른 노드들이 해시값에 해당하는 트랜잭션을 보유하고 있다는 것이 확인되면 그 노드는 모든 블록이 올 때 까지 기다리지 않고 자신의 로컬 멤풀을 재구성해서 블록을 구성할 수 있다. 만약, 이 과정에서 빠진 트랜잭션이 있다면 그 경우엔 블록이 전부 올 때 까지 기다려서 트랜잭션을 처리한다.

세이는 이러한 블록 전파 과정을 통해서 네트워크의 성능을 약 40%나 향상시키는데에 성공했다(이후에 언급하겠지만, Intelligent Block Propagation을 통해 파이널리티를 210ms까지 줄이는데에도 성공했다).

2.1.2 Optimistic Block Processing

세이의 Optimistic Processing 이전에는 블록을 처리하기 위해서 블록에 대한 컨센서스가 선행되어 이루어졌어야 했는데, 세이는 블록을 받자마자 처리를 시작하며 Prevote와 Precommit을 이와 동시에 처리하는 구조를 가지고 있다. 이러한 구조가 위험할 수도 있다고 생각이 들 수 있지만, 사실 블록을 처리하는 과정에서 Prevote와 Precommit이라는 과정도 같이 진행되기 때문에 해당 과정에서 블록이 타당하지 않다고 여겨진다면 해당 블록은 버리고 다시 시작하면 된다. 그렇기 때문에 크게 문제가 될 것은 없다. 세이 네트워크는 기존 합의 과정을 생략한 것이 아니라, 어디까지나 이들을 병렬로 같이 처리하는 것이기 때문이다.

세이는 Optimisitc Block Processing을 통해서 네트워크 퍼포먼스를 약 33% 개선시키는데 성공했다.

2.2 Transaction Parallelization

세이는 트윈 터보 컨센서스와 더불어서 트랜잭션 병렬처리도 진행하는데, 여기서 중요한 것은 어떤 트랜잭션들을 병렬처리 할 수 있는지 찾아내는 것이다. 모든 트랜잭션들이 병렬처리 될 수는 없는 것이, 각각의 트랜잭션들이 서로 영향을 미치는 트랜잭션들도 분명히 존재하기 때문이다. 결국 안전한 트랜잭션 병렬 처리를 가능하게 하기 위해서는 충돌 가능성이 있는 트랜잭션들을 찾아내는 것이 매우 중요한데 세이는 트랜잭션의 관계성을 DAG로 사전에 파악하고 만약 이들 트랜잭션이 서로 독립적인 트랜잭션이라는 것이 확인되면 병렬로 트랜잭션을 처리하는 것이다. 이를 통해 세이는 초당 20,000개의 주문을 처리할 수 있게 되었다.

3. 아직도 배고프다: 파이널리티를 줄이기 위한 최근의 노력들

3.1 파이널리티는 무엇이며 왜 중요한가

많은 사람들은 블록체인의 퍼포먼스와 확장성을 계산할 때 TPS(Transactions per second)를 많이 따지지만 사실 TPS는 많이 모호한 지표이다. 왜 그런지 예를 들어서 설명을 해보겠다. A 블록체인이 하나의 블록에 총 60만개의 트랜잭션을 담을 수 있다고 가정해보자. 그런데 이 A블록체인이 블록을 생성하는데 걸리는 시간이 1분이라고 한다면 TPS(=number of tx in a block / block time in seconds)는 10,000이 되겠지만 실제로는 60만개의 트랜잭션이 네트워크에 담기기 위해선 1분이라는 시간이 걸리기 때문에 초당 만 건의 트랜잭션을 실시간으로 처리한다고 볼 수 없다. TPS는 어디까지나 블록체인의 처리량을 초단위로 나눈 수치이기 때문에 겉으로 보이는 TPS가 실제로 유저들에겐 다르게 느껴질 수 있다는 것이다. 필자의 예시가 극단적이지 않아서 그렇지, 블록당 담기는 트랜잭션 수를 극단적으로 늘리고 블록타임을 1시간으로 늘린다고 하여도 TPS는 엄청 높은 수치로 나올 수 있다(하지만 유저는 트랜잭션이 네트워크에 담기기까지 1시간을 기다려야 할 것이다). 이러한 이유로 요즘엔 블록체인의 확장성을 논함에 있어서 TPS라는 수치보다는 해당 트랜잭션이 확정되기까지 걸리는 시간을 나타내는 지표인 TTF(Time to Finality)를 좀 더 많이 사용하는듯 보인다. 특히 세이 네트워크처럼 금융 거래에 초점을 맞춘 레이어1 체인의 경우, 트랜잭션들이 즉각적으로 처리되어 네트워크에 반영되는 것이 굉장히 중요하기 때문에 TTF를 최대한 빠르게 만드는 것이 중요했을 것이다.

3.2 다른 블록체인들과 비교

위에서 설명한 트윈 터보 컨센서스를 통해(특히 intelligent block propagation을 통해서), 세이 네트워크는 파이널리티를 약 250ms 까지 낮추는데에 성공하였다. 이는 현존하는 블록체인 중에서 가장 빠른 파이널리티로, 사실상 유저들이 체감하기엔 거의 즉각적인 파이널리티로 여길 정도의 속도라고 볼 수 있다. 물론, 이는 매우 긍정적인 상황에서의 파이널리티이고 세이 테스트넷인 Atlantic-2 상에서 여러 시도를 해봤을 때 가장 안정적인 파이널리티 타임은 410ms이었다. 파이널리티를 250ms까지 낮추는데에 성공하였지만, 특정 노드들이 네트워크를 유지하는데에 어려움을 겪었고, RPC 노드들과 같은 인프라 플레이어들이 뒤쳐지는 모습을 보였다. 이러한 경우는 어쩔 수 없는 것이, 블록체인은 그 특성상 전 세계에 노드들이 흩어져있기 때문에 전파와 소통에 분명한 어려움이 있을 수 밖에 없다. 그렇다면 어떻게 이 노드들을 전부 포함시키면서 파이널리티를 낮출 수 있을까?

3.3 세이의 최적화: 가장 보수적인 시나리오에서도 가장 빠른 블록체인

세이 네트워크는 위와 같은 문제를 해결하기 위해서 컨센서스 과정중에 커밋 타임 아웃을 약 100ms 추가하였다.

사실 100ms가 최적의 시간이 아닐수도 있다(굉장히 보수적인 환경이었다). 해당 커밋 타임아웃 시간은 앞으로 네트워크 상황에 따라서 더 짧아질 수 있지만, 일단 이번 테스트넷에서 추가된 타임아웃은 100 ms이었고 이를 통해서 타임아웃을 추가하지 않았을 때 안정적으로 달성할 수 있었던 410ms보다 좀 더 빠른 380ms의 파이널리티를 구현시킬 수 있었다. 재미있는 것은 100ms의 타임아웃을 두고도 세이는 현존하는 블록체인들 중에서 가장 빠른 파이널리티를 구현했다는 것이다. 앞으로 네트워크 상황에 따라서 추가된 타임아웃의 시간을 조정함에 따라 세이의 파이널리티는 더 줄어들 수도 있다. 보수적인 환경에서 구현한 세이의 파이넡리티가, 현존하는 다른 블록체인들의 파이널리티보다 빠르다는 것은 굉장히 흥미롭다.

4. 앞으로의 세이

사실 이번에 런칭이 거의 확정된 세이의 메인넷은, 지난 1년간 계속해서 밀려오다가 이제서야 확정이 난 것이다. 필자가 세이 재단 소속이 아니기 때문에 세이 메인넷이 밀렸던 이유에 대해서 명확하게 알 수는 없지만, 세이에 대한 리서치를 진행하면서 알 수 있었던 사실 하나는, 세이가 여태까지 네트워크의 성능을 개선시키기 위해서 계속해서 노력했다는 부분이다. 아마 본인들이 생각하는 이상적인 파이널리티를 달성하기 위해서 메인넷 런칭을 계속 밀어왔던 거 같다. 물론 파이널리티 타임을 낮추는 거 자체도 중요하지만, 더 중요한 것은 얼마나 네트워크가 안정적인 환경에서 운영되는지에 대한 여부이다. 세이는 속도와 안정성, 두 가지 속성을 확보하기 위해서 지난 1년간 메인넷을 미루면서까지 네트워크 개선에 집중하였다.

물론, 메인넷이 나오고 나서는 테스트넷과 또 다른 환경이라고 볼 수 있고 외부의 변수들도 많기 때문에 테스트넷에서 달성한 퍼포먼스를 그대로 이어갈 수 있다고 확답할 수는 없다. 하지만 테스트넷 환경 자체도 최대한 보수적이게 설정했던만큼, 메인넷 런칭 직후에도 빠른 파이널리티를 구현할 가능성이 높다.

높은 안정성과 빠른 속도, 세이 네트워크가 이 두 가지를 달성한다면 많은 유저들이 불편함 없이 블록체인을 사용할 수 있는 환경을 조성할 수 있을 것이다. 결국 레이어1이던, 레이어2던간에 유저들이 많이 사용해야 성공할 수 있다는 점에서 세이의 메인넷 런칭은 블록체인 업계의 참여자로써도 매우 관심이 가는 부분이다. 필자는 모듈러 블록체인의 방법론보다, 모놀리틱 블록체인의 방법론을 더 선호하는 사람으로서 세이가 진심으로 성공해서 레이어1 경쟁에 새로운 시작을 알렸으면 좋겠다.

About Us

Four Pillars is a global crypto research firm based in Seoul, consisting of the most influential blockchain researchers in Korea. Through robust research and governance skills, it helps various market players easily onboard to the blockchain industry by offering high-quality research articles while supporting protocols in their expansion into Korean and global markets.

Four Pillars Links

• Website (EN/KO)
• Twitter (EN/KO)
• Telegram (KO)

세이 네트워크가 파이널리티를 줄이는 방법 was originally published in Four Pillars on Medium, where people are continuing the conversation by highlighting and responding to this story.

Disclaimer: This article is intended for general information purposes only and does not constitute legal, business, investment, or tax advice. It should not be used as a basis for making any investment decisions or relied upon for accounting, legal, or tax guidance. References to specific assets or securities are for illustrative purposes only and do not represent recommendations or endorsements. The opinions expressed in this article are those of the author and do not necessarily reflect the views of any affiliated institutions, organizations, or individuals. The opinions reflected herein are subject to change without being updated.

Author: Taeheon Lee (@CryptKachu), Jungwoo Pyo (@hi_jakepyo)

Reviewer: 100y (@100y_eth)

If you want to discuss or share your opinions about the RWA protocol, please send a direct message to the Lano Technology Twitter account or an email to research@lano.im.

In this piece, we will delve into the crucial topic of handling risk management for Real-World Asset (RWA) tokens. We will analyze how various protocols are tackling these associated risks and discuss any troubles they may have experienced so far. Following this, we will explore the ways in which DeFi protocols are penetrating the lending market collateralized by RWAs. To conclude, we will evaluate a range of perspectives about RWAs in the web3 domain, while also considering the unique benefits and constraints these assets may introduce to the Web3 industry.

1. Managing default risk for RWA-backed loans

In the following section, we will be discussing the notable default risk associated with using RWA as collateral and the strategies that have been put in place to mitigate this risk. Before diving into the details, it’s crucial to comprehend the risks tied to the use of token collateral in traditional DeFi and the ways existing DeFi protocols tackle default risk. We will commence with some fundamental context, then shift our focus to how defaults in loans backed by RWA tokens differ, and examine how various RWA protocols approach these challenges.

1.1 Loan position defaults in previous DeFi protocols.

In the previous DeFi model, when a user deposits stablecoins into a protocol, they only need to take care of the mechanisms of the protocol whether they are able to protect the value of the deposits. In other words, it is crucial to evaluate whether the protocol would be able to perform processes such as liquidation when the price of the collateral token of a borrower drops. For example, if you were a depositor of USDC on AAVE, all you had to worry about was that when the price of the ETH collateralized by the lenders dropped, the protocol would liquidate the lenders’ positions and protect your USDC. The same is true for MakerDAO. All it had to do to keep the value of DAI at $1 is to liquidate the borrowers’ collateral positions before the value of the collateral held by MakerDAO fell below the value of the DAI issued as loans.

So from the depositor’s point of view, they only need to check if the protocol is using proper mechanisms from major protocols like AAVE or MakerDAO, and if there were any cases of defaults. From the protocol’s point of view, when defaults occurred, it checks what is wrong with the mechanism, and only modifies the mechanism appropriately to deal with the default.

1.1.1 Examples

In March 2020, MakerDAO experienced a large number of defaults due to a sharp decline in token prices. The problem at the time was that MakerDAO’s liquidation mechanism was poorly designed, preventing liquidators from participating in the liquidation process as required by the protocol. MakerDAO made improvements to the protocol during the recovery phase, including a significant revision to the collateral auction process of their liquidation mechanism.

In addition to liquidation mechanisms, existing DeFi protocols have risk parameters associated with liquidation to prevent defaults. With a help from a token economy modeling firm called Gauntlet, DeFi protocols such as MakerDAO and AAVE have set loan-to-value (LTV) thresholds for liquidations to occur, and adjust these parameters whenever they believe the market is in a bear market to ensure a protection of their collateral assets. For example, in December 2022, the AAVE community increased the collateral required for stablecoin loans in response to a slowdown in the cryptocurrency market.

1.1.2 Summary

In previous DeFi, if a protocol gets the right price for its token assets and disposes of its collateral at that price, it will not default. In other words, in the traditional DeFi model, if a protocol defaults, it is a problem with the protocol design, thus it means that the protocol is poorly designed or has not properly considered the risks that may arise in the protocol.

1.2 Loan Position Defaults in the RWA Protocol

1.2.1 Awareness of Borrower Defaults

Protocols dealing with RWA think about defaults a bit differently: they recognize that each RWA token collateral may default, and they don’t consider it the protocol’s obligation to prevent it from happening.

In conventional DeFi, when the price of the collateralized tokens falls below a certain level, they can be sold to the on-chain market. However, RWA tokens do not have a market where they can be sold instantly at a set price. Therefore, even if the price of RWA falls, it is difficult for the protocol to properly sell the collateral to protect the value of the protocol assets. Furthermore, no matter how tokenized RWA is, it is still unclear whether token ownership on the blockchain reflects ownership of the actual RWA. These two factors make it actually impossible to quickly liquidate collaterals on the blockchain and redeem the borrower’s loan position when the RWA token collateral needs to be liquidated.

1.2.2 Tranche: A Mechanism for Responding to Defaults

For this reason, the RWA protocol assumes that a default can occur at any time, and the focus is not on preventing the protocol from defaulting through liquidation, but rather on how to allocate the loss of assets to Lenders in the event of a default. To be more specific, some Lenders may be willing to make deposits because they want a high-interest return, even if there might be some loss of capital, while others may prefer not to incur any losses, even if the return is low. The RWA protocol categorizes Lenders and their deposits based on their risk preferences, even for the same vault.

This structure is called a tranche in traditional finance, and most RWA protocols adopt the tranche structure when accepting deposits from lenders. Tranches are usually divided into senior tranches and junior tranches. A senior tranche is a form of funding that is protected against losses but acquire limited returns, while a junior tranche is a pool that can receive a higher return in exchange for being the first capital to be exposed to losses.

Vaults that accept lender deposits in a tranche structure are also designed in the same way as tranches used in traditional finance. When a borrower repays the amount they borrowed plus interest, the repayment is first allocated to the senior tranche and then distributed to the Lenders who deposited in the junior tranche. Any remaining amount will be distributed to the junior tranche. In the event of a Borrower bankruptcy, any losses on the outstanding amount will be applied first to the junior tranche. When the junior tranche alone is unable to recover all of the outstanding amounts, the losses will be applied to the senior tranche as well.

1.2.3 GoldFinch Finance’s senior tranche

TrueFi’s Structured Credit Vault and Centrifuge Tinlake’s Tin & DROP both have senior and junior tranches per vault. However, GoldFinch Finance has a slightly different tranche structure. Here, lenders who deposit assets directly into a specific vault are called backers, and all of the backer’s deposits are recognized as junior tranches. Conversely, lenders who deposit money into the GoldFinch protocol rather than into a specific vault are called liquidity providers, and their deposits are recognized as the senior tranche. The distribution of how many backers deposit capital per vault determines how much of the capital pooled in the senior tranche is allocated to each vault. The result of this is that each VAULT is able to borrow more liquidity than the capital provided by the BACKERS, and in that sense, the model is called a leveraged model.

Let’s speculate on why GoldFinch is using a different senior tranche structure than other RWA protocols. First, GoldFinch’s target borrowers are all businesses in the developing world, even if they have different characteristics, so it seems natural to bundle all of the vaults that GoldFinch has lent to and create a deposit product for their risk and return, and then package that senior tranche as a diversified investment in financial products in the developing world and offer it to users. Other RWA protocols, on the other hand, are so different in terms of who they are lending to, that I don’t think it would have made sense to bundle all of the loans in the protocol into a single deposit product.

1.3 Design considerations for the Tranche structure

1.3.1 Funding the Junior tranche is important

As described above, in the RWA token collateralized lending protocol, if the borrower is unable to repay the loan, the protocol is exposed to non-repayment damage, and the lenders of these vaults may not be able to get their tokens back. Considering that lenders are mainly token holders who deposit stablecoins and want to earn a return with relatively little risk of losing their assets, there needs to be a way to limit the size of the lender’s loss in the event of a borrower’s bankruptcy.

In the current RWA token-backed loan protocol, the junior tranche fulfills this role. The fact that the junior tranche acts as a buffer against losses creates a situation where ordinary investors can invest in RWA protocols with relative ease. In other words, in RWA token-backed loans that are not “risk-free,” the junior tranche is acting as insurance for each loan position. RWA protocols need to be concerned with ensuring the size of the junior tranche deposit, monitoring that the junior tranche that exists in each vault is providing a sufficient return to the junior tranche depositor.

1.3.2 The ratio of junior and senior tranche sizes should be set well

If the size of the junior tranche is too small relative to the size of the loans originated in the vault, the junior tranche alone will not be able to absorb enough of the losses; in that respect, the junior tranche should not be too small. Conversely, if the size of the junior tranche is too large, there’s a chance that you won’t receive enough interest income for putting money into the junior tranche because of the structure of the junior tranche, which is distributed to the senior tranche and receives any leftover interest. This means that you need to set the size ratio of the junior tranche to the senior tranche appropriately. This ratio can be set by considering the following factors

• Borrower’s probability of non-repayment and expected losses The higher the probability of non-repayment in your vault, the larger the junior tranche should be. To determine a borrower’s probability of non-repayment, you can use the following information: The Borrower’s past lending and repayment history, the price stability and ease of disposition of the RWA pledged by the Borrower, and overall market conditions.
• Competitiveness of the junior tranche in terms of yield The interest yield that can be expected from the junior tranche should be naturally higher than the senior tranche, and at least moderately higher than the interest yield that can be earned on stablecoins in other DeFi, so that token holders have an incentive to deposit their assets in the senior tranche.

The RWA protocol should constantly monitor the above factors to track whether the junior tranche can maintain its size. If the return on junior tranche deposits is low, it is advisable to reward junior tranche depositors with tokens from the protocol.

1.4 Case Study

1.4.1 Default Cases

In 2022, news of defaults in the vaults of the RWA protocols have been coming out in rapid succession: first, a $3.4M loan from TrueFi to Blockwater Capital defaulted. Invictus Capital, a crypto fund that borrowed money from TrueFi, also failed to repay a $1M loan on time. Maple Finance also failed to get paid back the 2,400 WETH it lent to Auros Global, a crypto fund. In addition, a series of defaults in vaults managed by Celsius, Alameda, Orthogonal Trading, M11, and others led to Maple Finance defaulting on a $54M loan in December 2022. In addition to Maple Finance, $4.4M in defaults occurred at TrueFi and $2.5M at Centrifuge.

One of the commonalities of the defaults that occurred after October 2022 was that the borrowers were crypto funds, who often put their borrowed assets on exchanges like FTX. They planned to use the loans from the RWA lending protocol to execute their strategies, such as market-making and arbitrage, and make profits to repay the loans and interest. However, when FTX collapsed, their loans were frozen on the exchange, and they were unable to repay the loans from the RWA protocol. In addition to FTX, there is also a case of Invictus Capital that suffered large losses when the Terra stablecoin collapsed and was unable to repay its loan.

In the RWA protocol, when a vault defaults, the losses are shared among the Lenders who funded the vault. When Babel Finance defaulted on a $10M loan from Maple Finance in July 2022, the unpaid amount of the default was charged to the vault’s Lenders, resulting in a loss of about 3.2%.

1.4.2 Protocol Mechanism Improvements

The direction in which Maple Finance changed its mechanisms in the wake of last year’s defaults provides insight into additional default risk management methods that the RWA protocol could consider in addition to the mechanisms described above.

Borrower Diversification — One of the reasons why Maple Finance experienced such a large loan default last year (over $50M) was due to the high concentration of crypto funds in its borrowers. Since the borrowers are all players affected by the crypto market, when an event like the FTX collapse happened, many vaults experienced simultaneous non-repayments. From the protocol’s point of view, such a large number of non-repayments at once creates a burden on the protocol to recover the non-repayment through the treasury created by the protocol. Maple Finance has also recognized this issue and has decided to diversify its borrowers from 2023.

Empowering Vault Administrators — Previously, vault administrators could not declare a default in the event of a borrower’s non-repayment until a certain amount of time had passed. On its face, this may not seem like a problem, as the default is triggered by a mechanism defined by the protocol. However, since there is no way to liquidate the borrower’s collateral to force them to repay the loan, the RWA protocol requires the lender to freeze the vault assets as soon as possible to preserve the lender’s assets as much as possible when it is determined that the borrower has lost the ability to repay. With this in mind, Maple Finance has modified the protocol to allow the vault administrator to declare an immediate default if, in the vault administrator’s opinion, the Borrower has lost the ability to repay.

Aligning Vault Administrator Incentives — An issue that arose during the 2022 default was that some vault administrators did not share accurate information about the status of the vaults they were managing and the repayment status of their borrowers. Given the nature of the RWA protocol, which relies not only on the logic of the protocol, but also on the trust relationships between Lenders, Borrowers, and administrators using the protocol, this behavior by vault administrators can lead to greater losses due to defaults. In this respect, rules are also needed to force vault administrators to deposit a large amount of assets in the junior tranche of a vault when they create it.

2. Entry of DeFi protocols into the RWA collateralized loan market

Since the second half of 2022, one of the most talked about issues in the RWA ecosystem has been the emergence of several projects among existing DeFi protocols that have announced their intentions to incorporate RWA. MakerDAO and AAVE are the most prominent examples, while Frax Finance has also been actively discussing RWA-related mechanisms. In this part, we’ll discuss the context in which MakerDAO and AAVE have tried to introduce RWA, and what form it’s taking.

2.1 MakerDAO

2.1.1 Purpose of Introducing RWA

Stable growth of the protocol — One of the main agendas of the MakerDAO community has always been to create a means to stably increase DAI issuance. One of the problems MakerDAO has faced since running the stablecoin since 2018 is that it has been unable to continue to grow DAI with only ERC20 token collateral. MakerDAO recognized that accepting only token collateral would make DAI issuance too dependent on long demand for the cryptocurrency. By incorporating RWA collateral, MakerDAO hopes that DAI will provide utility as a lending instrument for entities looking to borrow on the basis of RWA or credit. Under this strategy, MakerDAO will no longer rely solely on cryptocurrency market conditions, but will also be able to create demand for DAI based on the real economy. As a result, DAI’s issuance will be able to grow more stably.

Monetization — MakerDAO also initially created a demand for DAI through the DAI deposit fees offered by the protocol, known as DSR, and based on this demand, the interest paid by borrowers became the primary revenue stream for the protocol. However, the MakerDAO community soon realized that the yield from tokens was not high or sustainable, and they started to focus on the yield from RWA, which is quite high and sustainable, and they started to issue DAI directly to entities that want to borrow against RWA and make the yield from RWA the main source of revenue for the protocol.

2.1.2 Adoption Criteria

Whenever MakerDAO creates a vault for a particular RWA or makes a loan secured by an RWA to a particular borrower, it relies on input from its community members to make the decision. In making these decisions, the MakerDAO community has decided to use the following evaluation criteria

• Scalability: First of all, MakerDAO’s position is that the RWA must have a high enough issuance volume and a large enough market size to increase DAI issuance.
• Stability: As RWAs are used as collateral for DAI issuance, it is important that the collateral remains stable in value. To that end, the MakerDAO community determines if the price of the RWAs is sufficiently stable and if there is sufficient liquidity in the market to dispose of them. Furthermore, the protocol assess whether there is any risk that this could jeopardize the stability of MakerDAO.
• Demand: Even if MakerDAO accepts the RWA as collateral and issues DAI, it doesn’t make much sense to accept the RWA as collateral on MakerDAO if it’s better to borrow from a traditional financial institution with the RWA in terms of interest and other financial costs. In this regard, the MakerDAO community evaluates the utility of borrowing on MakerDAO from the perspective of the RWA owner, i.e., if MakerDAO and the RWA owner are mutually beneficial.
• Regulatory Risk: In addition to dealing with RWA, the MakerDAO team also considers whether there are additional regulatory or legal entities that need to be involved in the process of incorporating the asset into the protocol, and whether there are additional technical, economic, or regulatory risks.

2.1.3 Steps required for implementation

In addition, MakerDAO has summarized the steps required to adopt RWA as follows.

• How can RWA assets be converted into tradable assets on MakerDAO? In order to hold RWA as collateral on the protocol, users need to be able to tokenize it and have a legal mechanism in place that allows them to dispose of it in the event of a liquidation.
• Is it possible to prevent a DAI borrower from pledging the same asset again in the traditional financial world? Since the legal rights of tokenized RWA are separated from the rights in the real economy, there is a risk that a borrower may borrow from MakerDAO with their RWA and then take out another loan in real finance. If this happens, it is unlikely that the borrower will be able to repay the loan they received from the RWA protocol. In this regard, MakerDAO tries to prevent borrowers from using their RWA as collateral to obtain another loan.
• Can price information, etc. be reliably retrieved? How will the oracle be designed? Even if there are differences in the way collateral is disposed of in the liquidation process, the value of RWA still has a significant impact on the borrower’s repayment potential because RWA acts as collateral. In that regard, MakerDAO is concerned about how to reliably fetch the value of those RWA assets.

2.1.4 Adoption Status

In July 2022, MakerDAO created the Real World Finance Core unit, which aims to increase MakerDAO’s impact on real-world finance. Through this unit, the protocol team has been making RWA-backed loans to actors in real-world finance and continues to create vaults for them. These vaults are working by issuing DAI directly to borrowers needing loans. You can check out the active RWA vaults here. The main vaults are as follows (as of March 2023)

The amount of DAI issued as RWA collateral on MakerDAO has now exceeded $600M, and the pace of DAI issuance has been increasing rapidly, especially since the end of 2022. Vaults like RWA009-A above are particularly significant because they provide funds directly from MakerDAO that would otherwise be lent by a government-regulated bank. In terms of protocol revenue, it’s also helping to improve the protocol’s bottom line, with DAI issued as RWA collateral worth around $650M generating around 75% of the protocol’s total fee revenue as of January 2023.

Against this backdrop, MakerDAO continuously reviews vaults that lend DAI as collateral for RWA, such as short-term bonds, institutional credit, and energy production facilities.

The MakerDAO community has expressed that it is inefficient and slow for each RWA to have to go through governance to decide whether or not to handle the asset. As a result, MakerDAO is looking to partner with RWA tokenization and RWA collateralized lending protocols such as Centrifuge (+ Tinlake) to allow the RWA protocols to take care of the necessary steps in handling RWA assets. MakerDAO is also in talks with Maple Finance about a partnership. Here again, Maple Finance will be utilized as the infrastructure to facilitate RWA-based collateralized loans, and MakerDAO will likely issue DAI to Maple Finance and adjust parameters such as DAI issuance limits and default interest rates for each vault based on the risk of the vault. In a partnership with RWA Protocol, the roles that MakerDAO expects the other protocol to play are as follows: 1) onboarding and managing borrowers, 2) conducting legal proceedings or organizing legal structures, and 3) handling defaults.

2.2 AAVE

2.2.1 Background

The rise of GHO — In July 2022, the Aave Governance Forum, in an agenda item called “Introducing GHO”, proposes a plan to issue a stablecoin called GHO based on the Aave protocol. Like other stablecoins, GHO would be backed by multiple asset classes. Still, instead of being crypto-specific, it was notable that it would be a mix of assets that were less correlated to each other, such as RWA and credit score.

In the GHO mechanism, instead of a single entity managing all collateral, an entity called a facilitator is authorized to issue GHO tokens. Facilitators propose a method for issuing GHO to the community, and if the proposal passes, GHO can be issued in the manner proposed by the facilitator, within certain limits. Interestingly, Facilitators do not have to insist on issuing crypto-collateralized stablecoins; in fact, GHO can be issued/burned in any way that each Facilitator defined, including collateralization against multiple asset classes, algorithmically, credit-based or delta-neutral position-based, and the more diversified the Facilitators, the more stable the price of GHO will be.

The Importance of RWAs in DeFi — In November and December 2022, the Aave community began to discuss the importance of RWAs. In “Helping Aave Think Through RWAs”, Punia from Warbler Labs noted that RWAs could help the Aave protocol in two ways: by providing stability through the diversification of stablecoin collateral outside of crypto and by paving the way for GHO to scale as a global stablecoin by leveraging a larger pool of assets than can be leveraged with crypto.

In “Why RWAs are Important to DeFi”, the author argue that while DeFi can achieve a certain amount of growth (10x~) by focusing on crypto natives, there is a clear limit to the size of the crypto market and that RWAs are necessary to achieve even greater growth (100x~) beyond that. DeFi needs to get some market share from TradFi to enter the mainstream eventually.

• DeFi and TradFi are analogous to the reversal of the ratio between e-commerce and in-store transactions, where people prefer e-commerce to in-store transactions now that the internet is available. The same is true for DeFi and TradFi: as blockchain technology develops and becomes more active, people will favor DeFi over TradFi.
• Finance is a field where user needs are always present, and people always want flexibility in their financial applications and funds, which is a universal phenomenon. DeFi can serve them better in all these ways.
• The visibility and interoperability of DeFi make it safer and less volatile than centralized protocols and better serve the economic system as a whole.

Success on MakerDAO — The success of MakerDAO further strengthens the case for RWA on Aave Protocol. Securing a steady stream of revenue from a DeFi protocol is crucial to its sustainability, and according to the article “Outlining an RWA Strategy For Aave”, as of November 2022, MakerDAO received over 50% of the protocol’s revenue from RWA.

The dashboard above shows the evolution of MakerDAO’s revenue by asset, with RWA accounting for well over 50% of revenue as of March 2023. With RWA accounting for a larger and larger share of total revenue over time, it’s clear that RWA is more important than we might think.

2.2.2 Effects and expected results

According to the article “Outlining an RWA Strategy For Aave”, the risks of Aave including only on-chain assets as collateral for GHO are as follows:

• During market downturns, there is a risk of accumulating liabilities in the protocol if there is a lack of market liquidity in the relevant assets, i.e., there is a risk of protocol default if only crypto assets are treated as collateral.
• The lack of liquidity in a protocol eventually makes it vulnerable to economic attacks or liquidation.
• Cryptocurrency assets are highly sensitive to market conditions, and during market downturns, the cash flow of lending protocols can deteriorate due to lower demand for leverage.

Ultimately, the Aave community expects to see the following effects when RWA assets are offered as collateral for GHO. This overall statement is broadly consistent with the effects the MakerDAO community expected when they decided to accept RWA.

• Because RWA is more stable and less volatile than crypto assets, its risk is more dependent on macroeconomic flows, and it is more robust to the volatility of the narrower crypto market itself.
• RWA borrowers’ financing is often for productive businesses, not speculative trading. This is a much larger scale than crypto speculation and is consistent regardless of bull and bear markets.

2.2.3 Aave’s RWA adoption strategy

RWA is a completely different asset class compared to traditional cryptocurrencies. There’s a lot of discussion in the Aave community about how RWA should be implemented. In the article “Outlining an RWA Strategy For Aave,” the author has broken down the different elements of DeFi to see which ones are being advocated.

• Asset types: The author argues for introducing less risky, lower-yielding RWA first to see how it works in the crypto market to build long-standing trust in RWA facilitators, and then progressively introduce riskier, higher-yielding assets. The asset classes should not be directly correlated to the crypto market, for example, by excluding loans to makers, miners, DAOs, etc.
• Debt Ceiling: In the real world, adopting GHO as a fiat currency will take time, so we must assume that all borrowers will initially exchange GHO for an existing fiat (USDC, USDT, USD, …). This is why the assigned debt ceiling should be proportional to the liquidity available to sell or redeem for existing stablecoins, as, ultimately, it is important for GHO to remain soft-pegged to USD to maintain its value. To summarize, the debt ceiling should be such that it will not depreciate if a significant amount of GHO is sold into the market.
• Interest Rates: The author argues that Aave’s current interest rate model should not be applied to RWA and that interest rates should be based on underwriting and TradFi-like deals. This is because Aave’s facilitators can set interest rates for RWA differently than other asset classes.

2.2.4 Adoption Status

On December 28, 2021, Aave partnered with Centrifuge to launch the RWA market for the tokenization of RWA (https://rwamarket.io/). Offering Centrifuge Tinlake’s TIN & DROP model within the Aave service, RWA market allowed depositors to earn yield on stable, crypto-correlated RWA collateral, and Centrifuge issuers to use the crypto market as an alternative source of liquidity by lending through Aave.

Since the RWA market is a closed market, only users who have been verified through a KYC process to become a depositor can participate as an investor. Issuers (Centrifuge Issuers) can create pools for their assets, and LP tokens correspond to the DROP of Centrifuge Tinlake. Depositors (investors) can stake their deposits with USDC to purchase DROP from that pool.

Immediately after launch, the size of the RWA market grew rapidly, exceeding $20M in April 2022, but later, during the market downturn, the market size shrank, and the total market size was around $7.5M as of March 2023. Fortunately, no capital defaults occurred, and the protocol itself works as intended.

Through the operation of the RWA market, Aave has learned that several key challenges need to be addressed to bring RWA to DeFi in earnest.

• It needs various domain experts related to Risk Assessment (RA) or Underwriting.
• Lenders should consider their strategy carefully, considering the duration and liquidity of RWA collateral, as it is an inherently different class of asset than highly liquid types like crypto.
• RWA’s funding depends on Aave’s stablecoin liquidity, which can often be affected by factors outside of Aave.

RWA facilitators. The example of the RWA market has shown us the potential for bringing RWA to DeFi. We’ve seen how the discussion of an RWA facilitator as one of the facilitators for the issuance of GHO is progressing. As of December 2022, according to the Aave community proposal “Centrifuge’s Perspective of an RWA Facilitator for GHO”, Centrifuge is continuing to attempt to become a full-service RWA facilitator for GHO.

In the previous article, we identified a number of key challenges that need to be addressed to make RWA scalable. In the Aave protocol, we expect the RWA facilitator to have the following roles:

• Compatibility with DeFi: This refers to ensuring that RWA collateral is well integrated into the on-chain ecosystem.
• Trusted infrastructure: Conducting intermediation involving RWA collateral requires a high level of transparency within a sound legal framework. A trusted, professional brokerage framework is required to onboard RWA into the on-chain ecosystem successfully.

Centrifuge is currently developing an index token based on the assets held by the Centrifuge pool. As an index to the entire pool, purchasers of the index token will purchase a portfolio of centrifuge assets, which is the result of risk diversification thought out by the collective intelligence of the credit group within the ecosystem. The RWA pool can also be included as a portfolio composition of the index, and this may be the best case of RWA experience as an Aave GHO facilitator.

3. The state of the RWA mortgage market and what’s ahead

So far, we’ve described the mechanisms used by the protocols that handle RWA tokens and why they are designed the way they are. In this part, we’ll take a quick look at the current state of the protocols that handle RWA tokens. We’ll start with the overall trend in lending volume and then cover how each protocol is doing.

3.1 TVL and overall loan volume trends

3.1.1 The growth and contraction of lending to crypto funds

Below is a graph showing the active loan volume of protocols that are lending against RWA tokens. Before 2022, TrueFi accounted for the majority of RWA collateralized loans, which is partly due to TrueFi being the first RWA collateralized lending protocol. After that, Maple Finance grew rapidly until the second half of 2022, when Maple Finance and TrueFi bisected the RWA collateralized lending market.

The composition of borrowers that Maple Finance and TrueFi lent to during this time was dominated by crypto funds that engaged in crypto-based market making and trading, such as MGNR, Auros capital, Folkvang trading, and Alameda research. Until the second half of 2022, it seems that Maple Finance and TrueFi experienced rapid growth as trading companies needed more funding due to the crypto bull run.

However, as we move through the second half of 2022, events like the collapse of Terra Protocol and the FTX bankrun have taken a toll on major crypto funds, and the volume of lending to crypto funds has dwindled accordingly. As a result, the weight of Maple Finance and TrueFi in the graph above virtually disappears from 2023 onwards.

3.1.2 Stable growth in lending to non-crypto funds

While Centrifuge and GoldFinch haven’t seen the same explosive growth as Maple Finance, they have grown steadily since 2022 and appear to have maintained their overall loan volume through the big events of the second half of last year without much damage. Here, Centrifuge is lending across various asset classes, including loans secured by real estate and loans secured by freight invoices. GoldFinch is lending to FinTech companies and real estate firms in developing countries. The figure below shows Centrifuge’s lending by sector, and they are actually lending to various sectors, including real estate, developing countries, and institutions.

Centrifuge and GoldFinch seem to have been spared much of the damage last year because their lending is not concentrated in crypto companies but rather in non-crypto companies, and in that regard, the demand for borrowing from them remains strong and has been growing steadily regardless of the state of the crypto market.

3.1.3 Increasing RWA Collateralization in DeFi Lending Protocols

In addition, as mentioned above, stablecoins and lending protocols like MakerDAO, AAVE, and FRAX are actively entering the RWA collateralized loan space and are slowly increasing the percentage of RWA assets within their protocols. FRAX has been discussing the allocation of FRAX to RWA collateralized loan pools through governance forum discussions since June of last year and has voted on partnerships with companies like TrueFi and PropFi. While we’re still very new to the RWA collateralized lending mechanism in DeFi, it will be interesting to see how fast this space grows over the course of 2023.

3.2 Benefits of utilizing RWA

3.2.1 Attracting real economy players to the DeFi ecosystem

It’s still true that most of the users of products like DeFi and NFTs are concentrated in a small number of people who have been owning and trading cryptocurrencies for a long time, so to grow the Web3 ecosystem, we need to onboard more users into the Web3 ecosystem. Various initiatives are underway, such as the Web2 X Web3 partnership, and RWA token-backed loans could be one of them.

If RWA collateralized loans can provide value to investment firms and institutions that traditional finance can’t and bring their liquidity into the DeFi ecosystem, it could act as an impetus to grow the overall size of the Web3 ecosystem once again. By slowly bringing more institutions into the Web3 space through the RWA protocols, blockchain can be positioned as an infrastructure for new financial technologies. Bringing RWA tokens into the DeFi ecosystem is a big step towards broadening the user base.

3.2.2 DeFi can have other means of monetization besides tokens

As previously discussed, RWA collateral allows blockchain projects to generate revenue from sources other than cryptocurrency. The problem with traditional crypto-based incentives is that they’re too dependent on market conditions, and they don’t provide meaningful or sustainable returns.

However, by utilizing interest income from RWAs as a new revenue stream, DeFi protocols can have a revenue stream other than their newly issued tokens, which can be attractive from a DeFi perspective, especially if the risks associated with RWAs are well-understood and managed. In fact, if you look at the rwa.xyz webpage, you’ll see that the average interest yield from RWA handled by DeFi protocols is 10%. Since most of MakerDAO’s revenue comes from its RWA vault, protocols can broaden their revenue streams if they get RWA right.

3.2.3 Secure demand that is not affected by the DeFi cryptocurrency market

One of the biggest problems with blockchain projects in any domain, whether it’s DeFi or NFTs, is that they’re all too heavily influenced by the cryptocurrency market, represented by Bitcoin. This problem is exacerbated for DeFi lending protocols because the people who deposit fiat tokens (ETH, UNI, etc.) on DeFi lending protocols and try to lend are usually people who are trying to leverage their long positions by borrowing. Conversely, people who deposit stablecoins on DeFi lending protocols and want to borrow fiat tokens represent short positions.

So if you look at DeFi protocols like MakerDAO or AAVE, you can see that the size of the protocols increases and decreases dramatically depending on market conditions. But if it includes the entities that own RWA in the user base, these protocols have a demand that’s relatively unaffected by the crypto market and that allows them to operate with some degree of predictability in terms of the size of the protocol.

3.3 Disadvantages or Limitations of Utilizing RWA

3.3.1 Trust issues arise

In addition to the advantages in terms of cost and transaction speed, the ability to achieve expected results through blockchain and smart contracts without having to trust a centralized party has been one of the most important values of DeFi. However, most DeFi activities that use RWA tokens as collateral are based on trust. Lenders have to trust a third party in each vault when they deposit their money, and then they have to trust that the borrowers will repay the loan honestly.

Suppose you look at what’s happened in the RWA protocol over the past year. In that case, we’ve had vault defaults and borrowers not redeeming because the third party in the vault was not acting honestly. We’ve had cases like that, and we’re going to continue to have these trust issues until there’s a more solid legal and institutional foundation around the RWA token, and even then, we’re likely to be using this protocol with some level of trust issues compared to fully crypto-native DeFi.

3.3.2 Only certain users are eligible for loans

In the RWA Lending Protocol, we mentioned that a certain amount of trust is required between lenders, borrowers, and third parties, and it is true that RWA token-backed loans are used for under-collateralized lending. However, because of the need for trust between actors, the group of users who can borrow through the RWA Lending Protocol is still concentrated in certain institutions. Look at products such as Centrifuge, GoldFinch, Maple Finance, etc. You can see that the entities that are borrowing or acting as administrators through this protocol are concentrated in institutions with a certain amount of reputation and expertise. In this respect, there is no market where ordinary people can easily borrow against RWA as collateral.

3.3.3 Does it really reduce costs?

Smart contracts can indeed reduce the cost of many arbitrations that occur during financial transactions. However, in the case of RWA protocols, the process of tokenizing the RWA, identifying the borrower, and the risk of the RWA incur additional costs because it requires experts with knowledge of the RWA and entities to handle legal and administrative tasks. In fact, some protocols require establishing a special-purpose vehicle for each vault. In this respect, there are still multiple costs associated with RWA. This limitation will continue to exist until the infrastructure for each RWA is in place and the risk assessment criteria for each asset are somewhat established.

3.4 Personal Opinion on the RWA Protocol

RWA has a lot of value in that it can solve some of the limitations of the blockchain ecosystem and some of the inefficiencies of traditional finance simultaneously. Once we get some administrative guidelines around RWA in 2023, RWA will get a lot more attention from the blockchain market, and that’s why RWA tokens are getting a lot of attention, and this market will grow in the future.

3.4.1 The need to recognize a different identity from traditional financial or Web3 products

But for a protocol that deals with RWA, it’s based on a very different philosophy than what many DeFi or blockchain projects are based on. This is because everything is not handled through smart contracts like the existing Web3 products, but to some extent, you have to trust someone and accept the problems that come with it. This is a very different characteristic from the existing Web3 products that people who have been in Web3 have seen, and the protocols that deal with RWA are located somewhere in between the existing financial system and Web3 products.

In this regard, when protocols deal with RWA, it is necessary for them to clearly understand not only the value they can bring to the existing financial sector and the blockchain ecosystem through their services but also why they are trying to deal with RWA and the various problematic situations and constraints that may arise in the process, both for those in the existing financial sector and the blockchain ecosystem. And protocols should be aware of the problems that may arise due to the transparency and centralization of information as they operate and improve them so that the RWA market can develop into a more mature industry.

3.4.2 The importance of risk management

As I keep mentioning in this article, the protocols that deal with RWA are still centered around lending protocols, and the lending that happens here is in the form of credit. Since the lending itself is often done for leverage purposes, the lending system itself can be destabilized at any time during the process of over-expansion and investment. This risk can be even more devastating to the protocol, especially since the lending is not regulated and monitored like traditional finance but can be done at the discretion of a third party that exists in each vault.

Most RWA protocols are not yet fully equipped to deal with these risks, so protocols are currently forced to rely on opaque information disclosure (compared to traditional Web3 products), the behavior of certain actors, and the unregulated nature of their mechanisms. Against this backdrop, each protocol must consistently take care of the way to manage these risks.

About Us

Four Pillars is a global crypto research firm based in Seoul, consisting of the most influential blockchain researchers in Korea. Through robust research and governance skills, it helps various market players easily onboard to the blockchain industry by offering high-quality research articles while supporting protocols in their expansion into Korean and global markets.

Four Pillars Links

• Website (EN/KO)
• Twitter (EN/KO)
• Telegram (KO)

The Mechanism and Ecosystem of RWA Protocols was originally published in Four Pillars on Medium, where people are continuing the conversation by highlighting and responding to this story.

Disclaimer: This article is intended for general information purposes only and does not constitute legal, business, investment, or tax advice. It should not be used as a basis for making any investment decisions or relied upon for accounting, legal, or tax guidance. References to specific assets or securities are for illustrative purposes only and do not represent recommendations or endorsements. The opinions expressed in this article are those of the author and do not necessarily reflect the views of any affiliated institutions, organizations, or individuals. The opinions reflected herein are subject to change without being updated.

Author: moyed (@samoyedali)

It seems like there are new on-chain games every day. Through this series with Iskra, I’ll be reviewing various on-chain games that are released or available for playtesting that month. Please keep in mind that my thoughts on these games are very personal, and if you have a favorite on-chain game you’d like to see me review, please let me know on Twitter @samoyedali.

1. What The Kart

1.1 Summary

What The Kart by Oh Baby Games has gameplay similar to Mario Kart and uses CT influencers as characters. The community is active, and we’re curious to see if they’ll license more CT Influencers and turn them into collectibles, or if they’ll create their own original IP.

1.2 Game Introduction

Oh Baby Games’ first game, What The Kart, is a Mario Kart game with a few twists.

• Crypto Twitter (CT) influencers as characters
• A customization system for karts
• New modes like ghost matching and player-made bounties.

The first round of alpha testing took place from July 1–3, and information such as which chains will be used and the launch timeline are not yet available.

1.3 Team Background

Pasteur “Pas” Tran, CEO of Oh Baby Games, the creators of What The Kart, previously founded an Australian-based game studio called Mana Wings. Mana Wings worked with big game companies like Activision and Ubisoft, most notably on Crazy Fishing VR. Oh Baby Games raised $6M in seed funding on January 31, 2023. Lead investors include Synergis Capital and eGirl Capital, with participation from gmjp, Merit Circle DAO, and Kevin Lin (Co-founder of Twitch). In addition to What The Kart, Oh Baby Games is also working on two other games, Rugpull Guys, inspired by Paul Guys, and Platform Fighter, inspired by Super Smash Bros.

1.4 Gameplay

The overall gameplay is very similar to Mario Kart, with drift boosts, flying items, and a variety of characters and maps.

Personally, I think the fun ultimately comes from the fact that you get to play a kart game while controlling some of the influencer characters you’re used to seeing on Twitter. For example, picking up Inversebrah and blasting items at Cobie or beating CL. This can be very appealing to CT addicts like me, but I think people who aren’t familiar with CT might not see much of a difference between it and Mario Kart, or might even think it’s worse. As a side note, there are tons of different CT influencers as characters, including CL, inversebrah, ansem, cobie, 3LAU, and Tetranode.

Here’s a gameplay video from @DantheOPMan.

1.5 On-chain elements

Aside from the use of CT influencers as characters and the fact that the game has received funding from Web3-friendly investors, it’s hard to see any connection to Web3. However, in one particular article, when discussing Oh Baby Games’ business model, they mentioned cross-game collectibles.

If it is a digital collectible, it means a collection of NFTs that are compatible across Oh Baby Games’ games, but I wonder if they will license existing CT influencers and collect them, or if they will create their own original IP.

1.6 Economic Model

In the case of Oh Baby Games, there was no mention of its own token, and no mention of its internal economic model. Based on the nature of the upcoming games, it seems like it would be more beneficial to simply allow users to purchase skins and digital collectibles through credit cards, USDC, or ETH rather than building their own economic model through a native token.

1.7 Community

If you look at Oh Baby Games’ Discord server, you can see that it has a fairly active community, with 49 boosts and various stickers and emotes being used without any giveaways or events. Even during the alpha testing period, the #alpha-testing channel was great for getting feedback, patching in real time, and answering questions.

In the case of Oh Baby Games, CT influencers often appear as characters or join as angel investors, so it seems that many people, including me, naturally learn about and are exposed to Oh Baby Games through these influencers.

1.8 Overall

As someone who enjoyed Mario Kart as a kid and is addicted to CTs, I found Oh Baby Games’ What The Kart to be enjoyable enough without the need for token incentives or complicated economic models. However, it’s unlikely to be an instant hit with people who don’t know CT. It would be great to see more CT influencers as characters, and it would be nice to see a variety of fun skins for each character in the future. It would be nice to be able to purchase without tokens, with a card or PayPal, or with USDC or ETH.

1.9 References

• A $6m bet to mix crypto and traditional gaming
• Oh Baby Games Launches Out of Stealth with the Closing of a $6M in Seed Round

2. Rhascau

2.1 Summary

Rhascau is a FOC (Fully Onchain) strategy racing game created by Minters World that uses the Arbitrum Nova chain. The game is played by rolling dice to move units, and can be played by up to four players. The game also uses a “burner account” that is stored in the player’s browser to reduce UX friction and process transactions for gameplay. The game is constantly evolving through an active community and feedback loop on Discord.

2.2 Game Introduction

Rhascau is a Fully Onchain (FOC) strategy board game created by Minters World that uses the Arbitrum Nova chain and was officially launched on July 5. As of July 31, the game has generated approximately 114K transactions and has received an impressive response from the community, with the total number of matches exceeding 250 in less than a week.

2.3 Team Background

The main team members of Minters World, the team behind Rhascau, include CEO @stokasz, game designer @FarathonMW, and full-stack developers @DevGhinter and @grzyslaw. Minters World launched their own 10k PFP NFT collection prior to Rhsacau’s launch, and while they don’t have a clear utility, they seem to have advantages such as early access to the game and in-game item airdrops. According to their website, they are planning to launch a new game called Battle For Blockchain in 2024, but no information about the game has been released yet.

2.4 Gameplay

Rhascau is a board game for up to four players, with a normal mode and a winner-take-all winner-take-all Ethereum mode. Players move four units each, and the objective is to complete a full turn and return to the starting point. Generators act as dice, and special abilities can be used to eliminate opponent units or gain extra turns. New units are placed on the board when the generator rolls a 1 or 6, and can be removed when they land on a space containing an opponent’s unit. If you don’t act for 24 seconds, you go AFK and another player can take your turn.

After playing it myself, I found it to be a fun game to play casually. Of course, it’s not without its drawbacks. The game doesn’t have a lot of strategy depth, and the slow pace of the game can make it feel a bit tedious. However, the rules of the game are easy to understand and intuitive, so even first-time players can enjoy it to a certain extent.

2.5 On-chain elements

On-chain logic

Rhascau’s code is available on GitHub, and everything from managing ongoing games and players, moving units, using skills, and even sending emojis is implemented on-chain. In a board game like Rhascau, one of the things we need to look at is how to generate random numbers. Rhascau uses a commit-reveal scheme, where the rollDice function commits to a hash value of the player's current block number+1, and the makeMove function reveals the value after the block arrives, and then generates a random number again from this hash value, which is then divided by 6 and added to the remainder by 1 to be used as the final generator value. Since the average Arbitrum Nova block time is around 0.5 seconds, this process takes less than a second.

Burner Account

In principle, everything you do in the game requires a signature, but Rhsacau eliminates this UX inconvenience with a burner account. When you first access Rhsacau, you sign to create a burner account that is stored in the player’s browser and sends transactions while playing the game on your behalf. Since the burner account is stored in the browser, if you delete your cookies or use a different browser, you’ll need to create a new buner account.

I’m actually curious about the exact mechanics of how the burner wallet sends and processes transactions on behalf of the user, but I haven’t found any information on it. If anyone knows, please let me know!

2.6 Economic Model

While it is expected to have its own token called $MINT, it is still unclear how it will be used. Based on the information that has been released, it appears that more $MINT will be earned by placing higher on the global leaderboard. As far as we know, the protocol’s own business model is a 2% fee on the rewards taken by the winner in Legendary mode.

2.7 Community

Currently, most Discord members are players of the game, and the server has a total of 14 boosts, which indicates a fairly active community. Since it’s a multiplayer game, it can be difficult to get people together, but Discord is constantly organizing match events every few days to make it easier for people to get together and play. Also, the recent patch 1.1 has been impressive in that they have been very responsive to user feedback on the #feedback channel.

2.8 Overall

Overall, Rhascau breaks my stereotype that FOC games are inferior in terms of UX and boring in terms of gameplay. The UX is very good, even when compared to regular games (I may be biased because I’ve already played Avitrum Nova), and the gameplay is more than adequate for casual enjoyment. Of course, there’s a lot to like. For starters, it’s not available on mobile, and the boardgame-style gameplay can feel tedious and slow. However, based on my personal experience and the numbers, I think it’s safe to say that Rhsacau is a successful start to the Minters World.

2.9 References

• Official Website
• Rhascau Tutorial
• Github

3. Frame Arms Girl: Dream Stadium

3.1 Summary

Frame Arms Girl: Dream Stadium is a 3D subculture roguelite game developed by Kotobukiya, LoadComplete, and Line Next and published on the GAME DOSI platform. FA Girl offers a good mix of strategy and hitting gameplay, as you beat each dungeon on a map with a character of your choice. However, the lack of a story and NFT strategy have been criticized by the community, especially as the large supply of NFT items has led to a drop in overall NFT prices. The game itself and the strategy for selling NFTs need to be improved, and a major marketing push is needed to address the existing issues and attract new users.

3.2 Game Introduction

Frame Arms Girl: Dream Stadium (FA Girl) is a 3D subculture roguelite game produced by Loadcomplete and the first game title released on the LINE DOSI platform. The official launch was on July 6, and naturally, it uses LINE’s Finschia.

3.3 Team Background

The main players involved in FA Girl are Kotobukiya, who provided the IP, Loadcomplete, who created the game, and Line Next, who handled the publishing.

First, Kotobukiya is a well-known Japanese figure maker that created the Frame Arms Girl series, which became quite popular and successful, selling over 500,000 figures in about a year and a half and reaching the #3 spot on Amazon Japan in 2015. The IP goes beyond just figures, with an anime and comic in 2017, and the characters have made occasional appearances in other games prior to FA Girls.

Loadcomplete is a game developer founded in 2009 whose most notable previous game is Crusader Quest. Crusader Quest is a 2D dot-based mobile RPG that has surpassed 25 million global downloads, but announced in early October that it would stop updating for the last time.

FA Girl was released through LINE’s GAME DOSI, which appears to be the part of DOSI that handles Web3 games on LINE Next’s all-in-one NFT platform. Starting with FA Girl, it appears that two other games are in the works, Project GD and Sweet Monster Guardian. Both games are already selling NFTs, although the exact details are not yet available.

3.4 Gameplay

FA Girl is a traditional roguelite where you pick a character, a map, and pick tiles on the floor to beat each dungeon. The gameplay doesn’t seem too physically demanding, but it was difficult for me. On the strategic side, you’ll need to find the right mix of supporter cards and E-units that you can equip to each character, as well as pick and choose your attribute items based on the monster’s attack pattern. At the end of each tile is a boss monster, and defeating a three-tier boss counts towards your ranking score.

Personally, I found the gameplay to be satisfying because it’s not too difficult to control, and it feels good to hit, even though there are some repetitive elements to it.

3.5 On-Chain Elements

FA Girl is a typical Optional Cosmetic Mints (OCM) game where only some items such as skins, support cards, and E-Units are NFTs, and users can trade NFTs on the GAME DOSI Marketplace. However, FA Girl’s NFT strategy leaves something to be desired.

The game launched with 1200 initial NFT skins for sale, but these skins were essentially just color changes, so it’s hard to evaluate their quality. In addition, a free airdrop gave away a common skin, Wind Blue Gourai, which was tradable. As a result, too many of these skins flooded the secondary market, which negatively impacted the overall NFT price. Announcing the sale of secondary NFT skins when the price was already down made the community’s reaction even more negative, and we’ve now paused the sale and are taking a wait-and-see approach.

3.6 Economic Model

There is no token-based economic model for the game.

3.7 Community

The total number of boosts for the GAME DOSI discord containing FA Girl is 8, which is a relatively small number considering that there are approximately 380,000 members. In light of this, it is reasonable to assume that FA Girl’s community is made up of members of the existing DOSI community. Despite the fact that the game uses a well-known IP, Frame Arms Girl, it could have benefited from a more sophisticated game and marketing strategy around this IP to attract new users.

The aforementioned strategy regarding NFTs, the lack of a story, and the unfavorable treatment of existing DOSI CITIZEN holders have not been well received by the community. It is not easy to improve this situation in the short term.

3.8 Overall

FA Girl is an OCM game with an interesting IP, and the gameplay itself is interesting and satisfying within each dungeon. However, there are a couple of things that I was disappointed with, the first being the lack of a story in the game, and the second being what they are doing with NFTs. With a great IP, developer, and publisher, I think they could have done a lot better.

3.9 References

• ‘Frame Arms Girl: Dream Stadium’ Review: A Neat Blend of Mecha Girls and Blockchain Gaming
• Frame Arms Girl: Dream Stadium
• Frame Arms Girl: Experience Dream Stadium
• Load Complete

4. Gran Saga: Unlimited

4.1 Summary

Gran Saga: Unlimited is a PC MMORPG game developed by METAPIXEL that utilizes blockchain elements such as Proof of Play, but without affecting the core game loop. METAPIXEL has an active community and has engaged influencers for content partnerships and playtests. METAPIXEL contributes to the Aptos ecosystem through its AIP-11, tokenized object model, and proxy gas fee payment system, and is a fun game for those who love the MMORPG genre.

4.2 Game Introduction

Gran Saga: Unlimited (“GSU”) is a PC MMORPG game that leverages NPIXEL’s existing Gran Saga IP, and is an OCA (Onchain Asset) game with some elements implemented on-chain. It uses Aptos as its underlying chain, has completed its first and second community playtest, and is expected to launch in late 2023, although an official launch date has not yet been set.

GSU features a player-driven economy, and the ultimate goal for players in the game is to become a System Agent, which is said to have unique powers to influence the GSU world, including the game’s economic model and other areas of the game.

4.3 Team Background

Founded in September 2017, NPIXEL is best known for its first game, Grand Saga, which was released in 2021. Since its domestic launch in January 2021, Gran Saga has generated approximately 140 million USD (approximately 180 billion KRW) in gross revenue through February 28, 2023, with nearly half of that coming from Japan. The company is also preparing to service the Chinese market, having completed the issuance of a Chinese license late last year. NPIXEL became the first Korean game company to become a unicorn, raising a total of $75 billion in Series A funding in 2020 and $100 billion in Series B funding in August 2021. NPIXEL is currently preparing a new open-world MMORPG called Chrono Odyssey.

METAPIXEL is NPIXEL’s onchain game studio, and the METAPIXEL ecosystem will be built on top of the Aptos chain. METAPIXEL’s lineup includes GSU, its own NFT collection, Catcha, and another game, Pixelcraft.

4.4 Gameplay

Thanks to 1mpal, I was able to participate in a second playtest. First of all, I personally haven’t played the MMORPG genre, so it’s hard for me to give a proper evaluation, but the completeness of the game itself seems to be comparable to a normal game. I originally planned to play up to level 15, but I gave up because it became difficult to level up rapidly from level 13. Users who were having fun seemed to be partying together, raiding, and doing other activities. The lack of running and jumping and the quality of the animations in between were a little disappointing, but they didn’t affect the gameplay too much. For a more in-depth review, check out tweet from 1mpal.

4.5 On-chain elements

Proof of Play (PoP)

Proof of Play (PoP) is a mechanism that METAPIXEL uses instead of a random number generator to determine the drop rate of rare items, and the relevant information is recorded onchain and can be viewed by anyone through Explorer. PoP works by inputting user-generated in-game information into a hash function, and if the output value is less than the target hash, it succeeds, and if it is greater, it fails.

Currently, PoP is not used for all in-game items, but it is applied to one of the most important items, the tesseract. These tesseracts have a certain chance of yielding keystones, which can be used to summon some of the most powerful Gran Weapons in the GSU universe. For example, when you open a tesseract, you’ll see a number between 0 and 999, and if the target hash is 750, you’ll only get the keystone if the number is less than 750, otherwise an epic item will be given to you instead. The difficulty of earning Keystones is designed to increase as more players join the server.

When gamers open a Tesseract in-game, they can view various information and data about the results directly through the PoP Explorer. For example, the following Tesseract opened by a user named “Dexler” resulted in a result_value of 81, which earned him the keystone.

Outside the Game

Outside of gaming, METAPIXEL continues to contribute to the infrastructure of the Aptos Chain.

• Tokenized Object Model (AIP-11): In-game items often change and evolve over time, and traditional token models have difficulty reflecting this. AIP-11 enables dynamic representation of in-game items as on-chain assets.
• Proxy Gas Payment System: This is a system that allows game companies to pay for gas on behalf of gamers who are unfamiliar with the concept of gas fees at certain events.

4.6 Economic Model

As stated in the AMA, METAPIXEL has no plans to launch a token yet.

4.7 Community

Through an event called “Gran Saga: Unlimited Idol,” GSU partnered with creators for content and allowed them to participate in the second round of playtesting with their fans, and I participated in the playtest as a fan of 1mpal. It’s a win-win situation, as GSU gets publicity through the influencers, and the influencers get to create their own content. These creators receive in-game Soulbound Tokens (SBT) called “Idol Tokens,” which do not affect game progression but allow them to interact with fans through special features such as growing in size and emitting light.

METAPIXEL’s Discord server has received a total of 124 boosts, indicating the support of many community members. In addition, the team has held three AMAs to date and two community playtests, one in March and the latest in July.

4.8 Overall

Overall, METAPIXEL’s GSU is the closest thing to a AAA game that we’ve seen to an on-chain game to date, and it’s clear that a lot of thought has gone into how to use on-chain elements in a way that doesn’t impact the core game loop, but gives gamers an extra edge. Personally, I’m not a big fan of the MMORPG genre, so I can’t say I enjoyed the game itself, but I’m sure it’s a game for those who already enjoy the genre. Outside of the game, it’s impressive to see a variety of open-source contributions directly to the Aptos ecosystem.

4.9 References

• METAPIXEL Medium
• METAPIXEL: Step Inside a Web3 Fantasy Ecosystem

5. Wildcard Alliance

5.1 Summary

Wildcard Alliance is a game that combines the RTS, battle arena action, and card collecting genres, and uses polygon chain. The game is being developed by Playful Studio, whose CEO, Paul Bettner, is a former developer of ‘Age of Empire’ and the game ‘Words With Friends’. Wildcard Alliance aims to be fun to play and fun to watch, with onchain capabilities that directly connect fans with streamers, creators, and esports players. The Genesis WildPass is currently available for playtest participation, airdrops, exclusive access, and more.

5.2 Game Introduction

Wildcard Alliance is a mix of RTS + battle arena action + card collecting genres, where players take on the role of a champion in the arena, summoning various creatures to destroy the opponent’s main base. It uses polygons as its base chain for the Gensis Wildpass and is currently in closed beta for Gensis Wildpass holders.

5.3 Team Background

Paul Bettner, CEO of Playful Studio, the company behind Wildcard Alliance, started his career as an Age of Empire developer and later created a game called Words With Friends at a studio called New Toy before selling the studio to Zynga. He spent about two years at Zynga as a general manager before leaving to start Playful Studio, where he worked on Super Lucky’s Tale, Creativerse, and his third game, Wildcard Alliance. In June 2022, the company raised $46M in Series A funding from Paradigm, Polygon, and Griffin Gaming Partners.

5.4 Gameplay

I was lucky enough to participate in the Wildcard Alliance Quest event and win a Genesis Wildpass, which gave me the opportunity to participate in the recent closed beta playtest.

It was one of the most fun on-chain games I’ve ever played. The goal of the game is to break the shield of your opponent’s home base in the battle arena and destroy it. Users summon their creatures based on their mana, and you, as the champion, fight alongside them in the arena. The arena has three lanes and several devices like teleports, and it’s a strategy game where you have to think about where, what creatures to summon, and when to summon them to fight alongside you. Later on, when multiple creature cards are released, it will be fun to build these creature packs like in Hearthstone. Also, each champion has different skills and ultimates, so while there are only two champions at the moment, BOLGAR and LOCKE, it will be interesting to see what creatures work best with which champions.

The only downside is that the specs are pretty high, making it difficult to run the game on my laptop. I ended up having to go to a local PC bang to playtest, which was well worth it for me. It’s also still in closed beta, so I’m sure there will be optimizations for lower specs as we get closer to the full release.

Here’s a video of @brycent_ playing Wildcard Alliance.

5.5 Onchain Elements

Why Onchain?

Wildcard Alliance started as a generic web2 game five years before CEO Paul Bettner realized it was web3. Playful Sutdio wanted to make Wildcard Alliance as fun to watch as it is to play, with fun gameplay as a fundamental part of the gameplay. To realize this vision, they approached streaming platforms like Twitch and YouTube, but it was not easy to change the direction of their platforms. After weighing our options, web3 emerged as the solution.

In other words, the Wildcard Alliance’s answer to “Why Onchain?” is that it allows for a direct, disintermediated connection between streamers, creators, and esports players and the fans watching the game. Just like a player giving away jerseys to fans at a real basketball game, the Wildcard Alliance allows for live giveaways from the wallet of Streamer A to the wallet of the fan B watching the stream.

Genesis WildPass

The onchain element within Wildcard Alliance at this point is the Genesis WildPass. Existing on top of the Polygon chain, the Genesis WildPass is a form of battle pass that gives holders access to playtests, airdrops, holder-only access, and early access to the Wildcard League ecosystem. The Wildcard Alliance will have separate web2 and web3 modes, with cards used in the web3 mode expected to be in the form of NFTs.

5.6 Economic Model

There are currently no known plans for a fungible token within the Wildcard Alliance.

5.7 Community

The Wildcard Alliance’s Discord server currently has a total of 30 boosts and is considered to have an active community. Discord regularly hosts closed beta playtests for Genesis WildPass holders, weekly quests with Genesis WildPasses as prizes, and community events to preview game designs and animations. Outside of Discord, the team actively interacts with its audience through its Twitter account, and a character introduction tweet in late 2022 received nearly 14.4 million views.

5.8 Overall

Personally, this was the most fun I’ve had playing this month, and it’s been a lot of fun compared to games outside of web3. Of course, it’s still in playtesting, so there are still some kinks to work out, but Wildcard Alliance is one of the most exciting on-chain games in the OCA format. There’s a good reason why it’s on-chain, and part of it is the format of a card game, so the introduction of NFTs seems natural compared to other games.

5.9 References

• Official Website
• CEO Interview
• Whitepaper
• WolveDAO: Deep Dive #1: Wildcard Alliance

6. Words3

6.1 Summary

Words3 is a Scrabble-style word game turned PvP game, with all logic implemented onchain. The game is played with the word “INFINITE” centrally created and then participants add letters to create valid words, with the cost of each letter determined by the VRGDA mechanism. The contract for Words3 is currently not publicly available, but previous versions of the contract are still available.

6.2 Game Introduction

Words3 is a PvP game that is a slight variation on the Scrabble-style word game where players take an alphabet and create valid words, and is a FOC game where all logic is implemented on-chain. The game is built using a MUD framework and is currently undergoing intermittent playtests. The underlying chain will be Optimism, and playtests are currently being conducted on the Lattice testnet.

6.3 Team Background

Small Brain Games, the creators of Words3, are a community of developers and players who make FOC games, and in addition to Words3, they’ve launched on-chain games like Dark Seas, Underseas, and Ape’s Gambit. They are currently working on an on-chain strategy game called Network States.

6.4 Gameplay

The game begins with the word INFINITE created in the center of the grid. Participants start with this word and add spellings to it, scoring points by creating valid words. The main difference from traditional word games is that each additional letter costs money. The cost of each word is determined by a mechanism called VRGDA based on the demand for that word at that time.

The player’s final profit is determined by Your Word Rewards + Your Prize — Amount Spent.

• Your Word Rewards: If another player uses your word or part of your word to form a valid word, 1/4 of the cost paid by that player is paid to you in World Rewards.
• Your Prize: The remaining 3/4 goes to the treasury, which is distributed at the end of the game based on each player’s score overall.
• Amount Spent: The total amount of money spent on adding letters to form valid words.

I participated in the second playtest on July 22. The dark purple are the words I created, but because I focused on simply creating a lot of valid words, I earned a lot of points but messed up my final PNL. In the end, I think it’s more important to earn the most points at the least cost and get others to use your words than it is to simply earn a lot of points to maximize your final profit.

6.5 On-chain elements

Words3 is a FOC game, meaning that all game logic is implemented on-chain. Previous versions of Words3 were built with MUD v1, and the current playtesting version of Words3 is built with MUD v2. You can read more about the differences between v1 and v2 of the MUD framework in this article, but the main difference is that while MUD v1 enforced an ECS, MUD v2 allows developers to use any data model they want. Words3’s current contracts will be released before the public launch, and previous versions of the contracts are still available on their GitHub.

6.6 Economic Model

There is no token-based economic model.

6.7 Community

Small Brain Games’ Discord server has only had two boosts, so there is not yet an active community. However, Small Brain Games seems to be focusing on creating a FOC game and evolving it through iterative playtests rather than building a community.

6.8 Overall

This is a simple game, but the addition of a cost for each spelling, and the fact that the cost changes based on demand, makes it different from other games. The game is still in the experimental stage, but we are looking forward to seeing how it develops as we continue to playtest.

6.9 References

• Words3｜Overview of examples of evolving existing game models such as word games and crosswords into on-chain games based on “Optimism x MUD”
• Words3 Wiki

About Iskra

Iskra is a community-driven gaming platform that aims to bring together users and game studios in a collaborative ecosystem. Unlike traditional Web3 gaming platforms, Iskra focuses on building a community of gamers and developers who run the platform together. The platform’s ultimate goal is to evolve into a fully decentralized gaming platform governed by the community.

Iskra Links

• Website
• Twitter
• Introducing Iskra

About Us

Four Pillars is a global crypto research firm based in Seoul, consisting of the most influential blockchain researchers in Korea. Through robust research and governance skills, it helps various market players easily onboard to the blockchain industry by offering high-quality research articles while supporting protocols in their expansion into Korean and global markets.

Four Pillars Links

• Website (EN/KO)
• Twitter (EN/KO)
• Telegram (KO)

Onchain Games I Played This Month: July was originally published in Four Pillars on Medium, where people are continuing the conversation by highlighting and responding to this story.