beliΞfanx.eth

Posted on May 22, 2022Read on Mirror.xyz

Fundamental Insights Weekly 2022.05.22

Fundamental Insights

#1 ZK VM

Bobbin Threadbare made a great talk introducing Miden VM at ZK Summit. We learned a lot about ZK VM from it.

In general, we can write ZKVM as initial state + programs → ZKVM → final state and proofs.

ZKVM takes initial state and programs related data as input and outputs final state related data and proofs. At the same time, ZKVM also accepts private input as witness. The witness can be signature or tx data. Validators can verify the ZKVM execution without the witness. The public input is not the same. Some ZKVMs use initial state and programs commitment and take initial state and programs as witnesses.

Limitation

ZKVM has some limitations:

  • Not every proving system can make a ZKVM
  • The performance of ZKVM is not as good as specialized ZK circuits

Instruction set

Normally ZKVM uses WASM, RISC V, EVM, or customized ZK-friendly instruction set. WASM and RISC V are popular. EVM is more blockchain-focused. Customized instruction set is more flexible.

Design

Machine Type

  • Stack Machine
  • Register Machine
  • Memory-memory Machine

Flow Control

  • Harvard architecture
  • von Neumann architecture
  • Merkelized abstract syntax tree (MAST)
    • All programs can be reduced to a single hash (the root of the MAST)
    • Every internal node is itself a MAST of a smaller program
    • Leaves of a program MAST are linear programs(no control flow)

Program initialization

  • Program hash
  • Public inputs

Function calls

  • Static call targets
  • Dynamic call targets

Memory model

  • Write-once memory
  • Read-write memory

Native data types

  • Field elements
  • Regular integers
  • Structs, arrays, sets...

Native field

  • 64-bit
  • 128 bit
  • 256 bit

Designers need to balance performance, developer-friendly, ZK-friendly and proof cost. For example, memory operation is cheap on ZKVM. Memory-memory machine ZKVM will be a good choice if you want a cheap ZKVM.

#2 How to find the next Bluechip NFT?

In the conventional definition, Blue Chips are traditional industrial and financial stocks with long-term stable growth. For "Bluechip NFT", we prefer to simply understand it as an NFT that has built up volume and can retain its value over bull and bear time.

How to define a Bluechip NFT from the data? We can see from the market capitalization distribution that there is a clear head effect in the bluechip NFT market. BAYC and CryptoPunks account for about three quarters of the top ten in total market cap, and among the NFT data for all public chains, the top ten NFTs are almost all above $500 million in sales volume.

The traditional blue-chip definition of "sustained growth attributes" is more pronounced for NFTs. Non-blue-chip NFTs have price curves that are more volatile in the early stages and less volatile thereafter, while blue-chip NFTs have more stable growth curves with intermittent bursts of trading volume. Compared to the former, bluechip NFTs have more sustained activity.

The growth of bluechip NFTs

In terms of holder distribution, the higher the “Bluechip degree", the higher the distribution of wallets. A wider distribution of holders indicates a higher degree of decentralization, avoiding the continuous trashing of the market, which will facilitate healthy long-term growth, and the more holders, the more active the community will be.

Long-term holders are known as "Diamond Hand", and the confidence they hold can often keep a project going for a long time. This is reflected in the bluechip NFTs, where the number of long-term holders has not only stabilized but also increased slightly since then, except for the initial mint phase when there were a lot of changing hands, indicating that more holders believe the project will appreciate over time.

In the traditional bluechip NFT, whether the team has sustainable operation ability, enabling NFT holders to have profits through continuous empowerment, such as airdrop, equity gift, etc., which is also the main reason to support the steady price increase. At the same time, there are many ways to empower the on-chain behavior, such as: reducing the circulation of NFT in the market by opening the staking function and turning short-term holders into long-term holders. BAYC, as a well-known bluechip NFT, takes empowerment to a new level by establishing DAO financing and issuing APE tokens. Transforming NFT into an ecology further enhances sustainable value growth.

#3 Why is decentralized storage important in a Web3 world?

Web3: Consensus, Storage and Computation

There are countless definitions and abstractions of what Web3 is and what Web3 strives to be. While definitions vary in terms of scope and depth, one definition by Nader Dabit, Developer Relations Engineer at Edge & Node, captures where most of these definitions intersect:

“Web3 is the stack of protocols that enable fully decentralized applications.” – Nader Dabit

Blockchain technology is what set off this decentralized revolution and brought about the emergence of the concept of Web3, representing the idea to not only decentralize consensus, but to use this technology to decentralize the rest of the internet too. Simply put, Web3 is the next stage of the internet and represents an on-going advancement from the information age to the age of decentralization.

Just like the Web2, Web3 is a complex amalgamation of a wide array of technologies that together form the Web3 ecosystem. Despite its complexity, we can break down the ecosystem into three key infrastructural pillars that need to be developed to achieve full decentralization of the internet: consensus, storage, and computation.

Figure 1: Illustrative slice of projects enabling each of the Web 3 pillars

Arguably the hardest part was to figure out how to decentralize consensus, leaning on many decades of research, trial, and error. The first successful protocol to achieve decentralization on a global scale was Bitcoin – and that was only 13 years ago.

Over time attempts at decentralizing storage and computation have emerged and aim to complement the systems of decentralized consensus, or even subsume and combine their concepts to build something greater. Although in above illustration consensus, storage and computation are portrayed as three distinct pillars, they are in fact far more closely intertwined: Ethereum, Solana and Arweave are smart contract enabled (and soon FileCoin too), meaning they have certain computational abilities. Dfinity (Internet Computer), Holochain and Stratos also have storage and decentralized consensus mechanisms built into their protocols. Despite having capabilities across all three pillars, they are each champions in their specific domains, applying unique solutions to achieve decentralization of key infrastructural elements of the Web3 ecosystem.

In the next section, we will take a closer look at the decentralized storage pillar.

The need for decentralized storage

Decentralized consensus protocols were not designed for storing large amounts of data. On the Bitcoin network one block has a theoretical maximum size of 4 megabytes, and a new block is verified every 10 minutes on average. That means that every day a theoretical maximum of 576 megabytes of data can be stored on the Bitcoin network. In contrast, Facebook alone generated 4 million gigabytes of data per day in 2020. Adding to this the cost of submitting a single transaction with a theoretical maximum of 4mb to the Bitcoin network, very quickly costs start to escalate. Using blockchains to store data is not a feasible solution.

Given the maturity of cloud storage solutions such as AWS, Google Cloud and Azure, you might be wondering why use decentralized storage solutions in the first place. The answer is simple: immutability and permanence. In systems with centralized authority, they have the power to, at any time, alter your data, restrict access to your data or use your data for other means. Decentralized storage solutions provide owners of data with additional safeguards to ensure data is provably unaltered over a period of time, and in some cases, can be made to persist permanently.

NFTs (non-fungible tokens) and dApps (decentralized applications) are two innovations that benefit from decentralized storage in the Web3 ecosystem. If the metadata of an NFT is stored on a centralized storage service, they run mutability and impermanence risks: imagine having an NFT, but without the underlying data (such as images or files) that give it context.

For dApps, decentralized storage can enable permanent service availability. A dApp can be thought of an interface to more easily interact with smart contract of a blockchain, and consists of two parts: the smart contracts that live on the blockchain, and the dApp frontend that needs to be hosted somewhere. On decentralized storage networks where data is replicated many thousand times there is no single point of failure that attackers can use to take down a dApp front-end.

Closing

If the past is an indication of the future, we will see further expansion into increasingly niche use-cases, which will continue to expand the web of interconnected protocols and solutions across all three pillars of the Web3 ecosystem. We are still early on our journey to unlocking the full potential of decentralization and a decentralized internet, and we are optimistic to see what the future holds.

Stay tuned for an in-depth piece on decentralized storage, exploring the above concepts and digging deep into various decentralized storage technologies, to be released in the coming weeks!

Week's Recap

Hot Topics

StepN

StepN has recently become one of the "most successful" DApps in the Move to Earn craze. StepN now has 240,000 total users and 10,000 new users per day. Its Token market value has reached $850 million.

source: FootPrint

Indicator Tracking

MVRV

source: Coin Metrics

Crypto Fear & Greed Index

source: Alternative

Data of NFT Market

source: NFTGo

Total Value Locked of DeFi

source: DeFiLlama

Total Value Locked All Chains

source: DeFiLlama

Protocol Total Revenue

source: Token Terminal