Azus

主要收获的点其实是在网络原理吧，包括防火墙，端口转发之类的，顺带也复习了一下Linux文件系统。

Wolf.game

Background

On a quaint farm in the metaverse, a flock of Sheep congregate and produce a healthy supply of $WOOL. They huddle together in a Barn and are sheared regularly by their owners to farm the $WOOL. With more $WOOL, the farmers can purchase more Sheep! But outside lurk dangers the Sheep are terrified of… The Wolves.
The Wolves are on the hunt for Sheep and their precious $WOOL. They’ll take it by any means necessary. They’ll kidnap Sheep or catch them unaware and steal all of their $WOOL. So the farmers struck a deal with the Wolves: they pay the Wolves a tax on all $WOOL production. In return, the Wolves don’t attack Sheep who are safe in the Barn.
But when a Sheep leaves the farm or new Sheep are born… The Wolves don’t hold back.

Wolf Game is a risk protocol for NFTs with novel tokenomics. It shows what’s possible with interactions between the ERC-20 and ERC-721 protocols.For the very first time, your NFT can steal other NFTs (ERC-721 tokens) for you. The rarer your NFT, the more tokens you’ll accumulate probabilistically. Wolf Game is pioneering new types of NFT mechanics. Fully decentralized. No roadmaps or empty promises. Just a game in the metaverse that’s ready to play at launch.

The tl;dr:

 - There will only ever be 10,000 Gen 0, minted for 0.069420 ETH each. The 40,000 Gen 1 are minted by farming $WOOL
 - Sheep can be staked in the Barn to earn $WOOL and pay a tax anytime they claim their $WOOL earnings
 - If a Sheep is unstaked from the Barn, the Wolves try to steal all of its accumulated $WOOL
 - When a new Sheep is born, the Wolves attempt to kidnap it. If they’re successful, it’s given to a randomly selected Wolf, instead of the owner who minted it
Contract Addresses
 - Sheep / Wolf NFT: 0xEB834ae72B30866af20a6ce5440Fa598BfAd3a42
 - Barn / Gang Staking: 0x29205f257F9E3B78bcb27e253D0f3Fad9D7522a2
 - $WOOL Token: 0x8355dbe8b0e275abad27eb843f3eaf3fc855e525

Minting

Token ID Minting Cost
1 to 10,000 (Gen 0) 0.069420 ETH
10,001 to 20,000 20,000 $WOOL
20,001 to 40,000 40,000 $WOOL
40,001 to 50,000 80,000 $WOOL
The total cost to mint all of the Sheep and Wolves in existence will be 1,800,000,000 $WOOL.

Sheep

You have a 90% chance of minting a Sheep, each with unique traits. Here are the actions they can take:
Action Notes Risk
Enter Barn (Stake) Accumulate 10,000 $WOOL / day (prorated to the second) No risk.
Shear $WOOL (Claim) Receive 80% of $WOOL accumulated on your Sheep Wolves take a guaranteed 20% tax on sheared $WOOL in return for not attacking the Barn. Taxed $WOOL is split among all the Wolves currently staked in the Barn, proportional to their Alpha scores.
Leave Barn (Unstake) Sheep is removed from the Barn and all $WOOL is shorn. Can only be done if the Sheep has accumulated 2 days worth of $WOOL to keep it warm. 50% chance of ALL of your accumulated $WOOL being stolen by Wolves. Stolen $WOOL is split among all the Wolves currently staked in the Barn, proportional to their Alpha scores.
​

$WOOL

The maximum $WOOL supply is 5,000,000,000 $WOOL:
When supply reaches 2,400,000,000 $WOOL earned for staking, the staking “faucet” turns off.
The developers will receive 600,000,000 $WOOL
Community Rewards will be allocated 2,000,000,000 $WOOL
Action Notes Risk
Mint a new Sheep using $WOOL There is a 10% chance that the NFT is actually a Wolf! 10% chance of the new Sheep or Wolf being stolen by a staked Wolf. Each Wolf’s chance of success is proportional to their Alpha scores.

Wolves

You have a 10% chance of minting a Wolf, each with unique traits, including an Alpha value ranging from 5 to 8. The higher the Alpha value:
 - The higher the portion of $WOOL that the Wolf earns from taxes
 - The higher chance of stealing a newly minted Sheep or Wolf
Example:
Wolf A has an Alpha of 8 and Wolf B has an Alpha of 6, and they are staked.
 - If 70,000 $WOOL total have been accumulated as taxes, Wolf A will be able to claim 40,000 $WOOL and Wolf B will be able to claim 30,000 $WOOL
 - If a newly minted Sheep or Wolf is stolen, Wolf A has a 57% chance of receiving it and Wolf B has a 43% chance of receiving it
Only staked Wolves are able to steal a sheep or earn the $WOOL tax.
Action Notes Risk
Stake Wolf Earn your share of the 20% tax of all $WOOL generated by Sheep in the Barn No risk.
Claim $WOOL Receive all $WOOL taxes accrued for the staked Wolf No risk.
Unstake Wolf Receive all $WOOL taxes accrued for the staked Wolf No risk.

Why this tech is novel

Protocol-level risk is ripe for exploration in NFTs. Many projects are implementing stake-to-earn, but haven’t cracked the code on users making choices in the face of risk.
In the case of Wolf Game’s Wolf-eat-Sheep world, your NFT can steal ERC-20 and ERC-721 tokens for you. This is entirely new.
Everything in Wolf Game happens on-chain: the decisioning, the results, the generation of the NFTs themselves. It’s split between 4 different smart contracts so that anyone can read through and get a sense of how it all talks to each other.
There are a number of techniques in these contracts that make this all possible while maintaining accuracy, keeping gas costs down, and not exceeding the EVM contract size limit.
Data Packing:
Storage is expensive on the blockchain! One 256 bit integer costs 20,000 gas to save, which at today’s costs is about $7.50… for one number! Efficiently packing data types by designing around smaller max values can save users hundreds of thousands of dollars over the lifetime of the project
Constant Time Algorithms & DeFi Math:
A lot of contracts use architectures that require significant looping to accomplish tasks. Those make contracts inefficient and add to gas costs. It’s imperative to remove things like the need to search through arrays or loop through stakes. This means you can do things like unlimited simultaneous stakes while keeping things cheaper.
As an example, the minting contract uses AJ Walker’s Alias Method to efficiently select traits in constant time.
Another example: tracking Wolf $WOOL earnings (taxes) proved to be very complex, requiring math used by DeFi liquidity pool protocols.
100% On-chain:
Not the first, and certainly not the last. But as long as Ethereum is running, your Sheep and Wolves will survive. Always available and always yours. Your traits and all the pixel art reside in the contracts themselves, nowhere else.
UI at launch:
You can use the entirety of Wolf Game by directly interacting with the blockchain (like through Etherscan). But Wolf Game is launching with a custom UI to make it easy for people to enjoy the game on day 1.

In conclusion…

With Wolf Game, everyone can play with their own level of risk… Pay the tax? Stay liquid on the market? Take a chance to keep all your $WOOL, but have a chance to lose it all? The choices and this game have to be engaging.
Wolves preying on Sheep feels almost like a parallel to the NFT community itself: A select few with alpha among all of us, and many others following the pack.

Cat.game

In the middle of the vast METAVERSE, there is a house full of cheese. Inside the house, a lot of mice live one after another inside the wall. They search around the room every day to collect cheese, so that more mice wandering in the METAVERSE can live in it!
However, there are more cats inside the METAVERSE who are watching for them! They are so clever and cunning that they steal countlessheese the mice find every day!
They watched the room every day, not letting any mouse take the cheese out of the room, sometimes even kidnap new mice that are sneaking in!

Definition

The cat game aims to create an iterable, large-scale METAVERSE social experiment using popular decentralized projects and novel tokenomics.
[MISSING](Technological details.)
You get to decide how to interact with the community using your NFTs and $CHEESE Tokens and all the profit or responsibilities are your own.

Quick Guide

Stage 1:

• NFT Mechanism: MISSING
• Mouse can be stacked to go out the Hole to safely collect $CHEESE everyday
• If a mouse wants to leave the house with the cheese he has collected, the cat will seize the opportunity to harvest all those free cheese.
• NFT Stoleing Mechanism: If you want to bring a new mouse to the house, cats will try their best to catch our new fella - and when they succeed, it’s given to a randomly selected Cat, rather than whoever minted it.

Contract Addresses

MISSING

Minting

MISSING

Mouse

You have a 90% chance of minting a Mouse and they all have unique characteristics. They can do:

Stack: Find CHEESE
found $CHEESE everyday

Claim CHEESE through the Hole
lost 20% due to cats’ harrassment

Unstack: Leave the House with all the CHEESE
50% chance of ALL of your accumulated $CHEESE being rugged by Cats. Rugged $CHEESE is split among all the Cats currently staked in the House, proportional to their Skill levels.

$CHEESE

The maximum $CHEESE supply

When supply reaches 2,400,000,000 $CHEESE earned for staking, the staking “faucet” turns off.

The developers will receive 600,000,000 $CHEESE

Community Rewards

Actions:
Mint new Mouse:
10% chance you’ve found another cat in the METAVERSE!
10% chance of the new Cat or Mouse being stolen by a staked Cat. Each Cat’s chance of success is proportional to their Skill Levels.

Cats

You have a 10% chance of minting a Cat and they all have unique characteristics and Skill Levels.
The higher Skill Levels indicats:

• Higher portion of $CHEESE that cat earns from harrassing mice taking CHEESE back to the Hole.
• Higher chance of catching a newly minted Cat or Mouse.

They can do:
Stack Cat:
Earn share of 20% from mouse claiming the CHEESE

Claim Cat:
Receive all $CHEESE accrued for the staked Wolf

Unstake Wolf: Receive all $CHEESE accrued for the staked Wolf

Key Features:

To be written.

Conclusion:

To be written.

DAO

前言

自发zi
自治
自动
自治最好解决：投票。(Multisig) 由上至下->由下至上
但是低效，所以自动：
自动是每个项目的核心-是所有区块链项目解决的问题
自发：由于需要从下至上，所以需要自发，自发性需要积极性（自动，即代码做事，更加使得群体化的陪伴感减弱，自发性减弱。）
想要激励来提高自发性
Discord已经做得很好了，meme也是为了完成这个事情

每天对着最多的东西，还是要好好布置👍

This dictionary aims to briefly explain the most important terms from Andrew.Ng’s deeplearning.ai. It contains short explainations of the terms, accompanied by links to follow-up posts, images, and original papers. The post aims to be equally useful for Deep Learning beginners and practitioners.

基于pytorch的深度卷积神经网络(CNNs)算法原理的初步学习
第一次diveinto人工智能的黑盒里头，让我看看ML的网络里面到底装了什么~

阅读 《图解HTML》笔记。
使用request/bs4进行爬虫等项目时遇到的http协议相关问题，做一个记录。

解析Repository bupt-gym-reserve