Over 90% of data produced in the world today remains locked up in data silos due to concerns about privacy. Once data is shared, owners lose control over how data is used and often jeopardize sensitive information.
在现有基础设施环境下，数据拥有者无法控制自身资产如何被使用。无数的隐私泄露、知识产权窃取事件告诉我们：一旦将数据交由第三方使用，基本等于交出了所有权。 这导致了 90% 的数据被锁在「信息孤岛」中，价值巨大却深埋地下。这些数据中既有隐私，又有重要商业价值。只要我们拥有合理的工具，在保证隐私的情况下挖掘这批存量资源，对数据进行运算、交易，就能让它展现更大的价值。 人类在 1846 年发明了第一座现代化油井。今天的我们，又应该如何解决数据的开采？
Data transferred through the COVA protocol can be protected even after access has been granted. COVA powers Smart Policies which enable data owners to allow their data to be accessed, and transacted while prohibiting undesirable privacy-compromising inferences. By empowering data owners without sacrificing the freedom of data transfer, Smart Policies give data property rights and enable it to be transacted as capital.
COVA 协议首创的 Smart Policy（智能条款）将改变这一点。即便将数据交由第三方使用，用户仍对其保有 100% 控制权，可以规定它应该以何种方式被访问、交易，防止隐私泄露。在保证数据自由流动的情况下，智能条款将赋予数据拥有者真正的物权，从而推动数据通货化交易。
COVA protocol is a new addition to the internet protocol suite that creates a web of secure and trusted execution. On COVA, data can be controlled with a Smart Policy: a data usage policy written in code that defines and enforces how it can be used.
作为现有互联网协议的补充，COVA 将创造一个安全、可信任的数据传输环境。在 COVA 协议下，数据产生者可以附加一个智能条款（Smart Policy），从而规定他人如何使用数据，并确保在该条款上运行的程序遵守该条款。
Data is safely transacted and processed inside a secure enclave by a swarm of TEE nodes equipped with Intel Software Guard Extensions (SGX), which prevent the host from being able to see your data or tamper with execution of code. COVA uses the guarantee from the processor-level security to implement self-protecting data with Smart Policies.
在可信任执行环境（TEE）节点和 Intel® SGX 等技术的支持下，数据可在安全区域（Secure Enclave）中完成传输。第三方使用者将无法查看数据或者篡改用于处理数据的代码。由处理器级别开始，智能条款（Smart Policy）就能帮助数据实现自我保护。
COVA is partnering with several institutions such as hospitals and financial institutions, to open up a market of data previously locked up in data silos due to privacy concerns.
COVA 正在与数家医疗、金融机构展开合作，利用智能条款保证优质数据供给渠道安全畅通。
COVA is the only data privacy project to support the entire scientific computing stack of python—including Numpy, Scipy, Scikit-learn, and Pandas on SGX. Data scientists can organically integrate their existing models with data on COVA and use tools familiar to them, including much of the Python ecosystem.
COVA 充分挖掘技术潜力，使得在 Intel® SGX 上运行包括 Numpy、Scipy、Scikit-learn 在内的 Python 科学计算工具成为可能。对于数据科学家而言，他们可以将现有的数据模型无缝接入 COVA，并使用熟悉的工具，进一步提高分析效率。
This enables data owners to enforce privacy-preserving computation, create transparent, uncircumventable DRM, and define property rights for data, alongside many other applications that have not yet been conceived, all without the lofty impracticality of Fully Homomorphic Encryption (FHE) or Secure Multi Party Computation (MPC).
相比同态加密（FHE）、安全多方计算（MPC），COVA 采取的技术手段拥有更好的落地性。即便经过计算处理，数据拥有者也能确保自身资产及隐私不被泄露。数据物权的确认，更为透明、可靠的数据版权管理（DRM）也将变得指日可待。
Powering the COVA network is a substrate of TEE nodes that run COVA software, and provide the horsepower of secured and trusted execution. All data and code that is processed by the COVA network are securely pieced together, hidden and protected inside an enclave. Smart Policies are written in Centrifuge, and provide data owners with a new tool in their arsenal to prevent their precious data from being abused.
同样是赋予用户数据控制权，COVA 是首个提供软件+硬件结合的项目。其他项目不是只开发硬件环境（TEE 网络），或者只依赖于软件（MPC，ZKP）。 COVA 的硬件环境包括了众多 TEE 节点。基金会利用 Token 机制激励更多的 TEE 节点加入我们的网络，为我们提供可不断扩展的可信算力。这些可信算力确保 「监督程序」的行为没有安全隐患。它们将成为定义和执行智能条款的工具的底层架构。在安全区域（Secure Enclave）中，所有运行在 COVA 网络上的数据、代码将被安全地分割、隐藏，并得到保护。 为了让机器理解自然语言语义，懂得如何根据数据拥有者意志保护数据，COVA 开发了 Centrifuge 程序语言。Centrifuge 是一种条款规范语言（Policy Specification Language，PSL）。它将人类的数据使用规范转化为可执行的代码，即「智能条款」。
Centrifuge bridges semantic notions from natural language to machine-enforceable policy code.
With Centrifuge, you can write smart policies that:
可以让机器理解自然语言语义，将其转化为可执行的代码。通过 Centrifuge，你可以编写智能策略，达成以下目的：
Data silos come about from a large dataset containing sensitive information that can't be leaked. However, these datasets may contain potential for important insights beyond what data owners are able to imagine.
数据中既有隐私，又有重要商业价值。Centrifuge 可以在保证用户隐私的情况下对数据进行运算。由此，数据孤岛将被打破，让潜藏于地下的财富浮出水面。
By opening up data silos to an open market of analysts that have ideas on how to exploit the data, COVA enables data owners to be compensated for their data, and allow them to regulate the types of inferences that people can make on the data, by writing policies that enforce consuming computation to use k-anonymity or differential privacy.
通过编写智能条款（Smart Policy），强制使用者采用 k-匿名性（k-anonymity） 或差分隐私（Differential Privacy），数据拥有者可以可以限制他人对于数据的更改、破坏。
Digital Rights Managements is currently a mess, with thousands of custom solutions created by different companies. It is hard to keep track of what one can and can’t do, as the DRM policies are often restrictive, opaque, and embedded within the actual data itself. COVA lets data owners write policies that are transparent (policies are public), do not corrupt the actual data, can be enforced without fail, and extend beyond just copyrighted music and art.
由于存在多种解决方案，目前的数字版权管理（DRM）生态极为混乱。大多数字版权管理政策具有限制性强、不透明、捆绑于数据等特点，这大大加剧其使用难度。 开放的 COVA 协议，使得数据所有者可以编写透明的智能条款作为数据使用政策。该政策不会破坏数据本身，且失败率极低。
更多内容请查看 帮助文档
More on this tutorial
Executive whitepaper published
发布执行白皮书
Initial PoC of COVA Medical Data marketplace (MEData) completed
完成对于 COVA 医疗数据市场（MEData）的初始概念验证
Initial research of TEE and COVA VM implementation started
开始初步研究 TEE 和 COVA VM
Implementation on TEE and COVA VM started
开始开发 TEE 环境和 COVA VM
COVA community channels launched
COVA 社区频道上线
Work on Covalent Policy Language (Centrifuge) started
开始开发智能条款语言 Centrifuge
Work on Public Testnet started
开始开发公共测试网络
Proof-of-concept of Smart Policies with COVA VM through modifications to CPython 2.7
完成通过 COVA VM 实现智能策略的概念验证
Modification to Graphene Library OS started
开始调试 Graphene Libary OS
Successfully ran full python scientific computing stack inside COVA VM (i.e. SGX core OS)
在 COVA VM（即 Intel SGX 核心 OS）内成功运行完整的 Python 科学计算工具
Public testnet MVP release
发布公开测试网络最小可行性方案（MVP）
Start recruiting compute node hosts among community members
开始在社区成员中招募计算节点主机
Add more smart policies beyond basic policies
增添更多的智能合约
Deploy system to COVA protocol partners for various projects
为合作伙伴部署 Covalent 协议
Start work on scaling of Covalent TEE network layer
开始扩展 Covalent TEE 网络层
Start recruiting compute nodes at scale
开始大规模招募计算节点
Collectively surpass computing power of the most powerful supercomputer on the planet (4000+ compute nodes)
招募超过 4000 个计算节点（此算力将超越目前最高速超级计算机）
Sharding and expand routing nodes for further scaling (i.e. support dataset and compute node provision at scale)
扩展路由节点以进一步扩展（大规模支持数据集和计算节点）
Professor of Computer Science @ Harvard University
Leader of Data Systems Laboratory (DASlab) @
Harvard School of Engineering and Applied Sciences
哈佛大学计算机科学系教授
哈佛大学工程学与应用科学学院数据系统实验室领头人
Professor of Marketing @ Harvard Business School
M.A. in Economics and Ph.D in Business @ Stanford University
Former development engineer @ IBM Research
哈佛商学院市场学教授
斯坦福大学经济学硕士、商学博士
前 IBM 开发工程师
B.A. in Computer Science and Statistics @ Harvard University
Early employee @ Gigster (a16z-invested)
Former engineer @ Citadel
本科毕业于哈佛大学计算机科学及统计系
Gigster（A16Z 投资企业）早期员工
前 Citadel 工程师
M.S in Mechanical and Aerospace Engineering @ Princeton University
B.S. in Applied Physics @ Peking University
Early investor in Conjur (acquired by NASDAQ: CYBR).
硕士毕业于普林斯顿大学机械工程与航天工程系
本科毕业于北京大学应用物理系
Conjur 早期投资人（后被纳斯达克上市企业 Cyberark 收购）
Founder of FBG Capital
FBG Capital 创始人
Investor of Bitfinex & Limited Partner of Bitfund
Consultant of FBG Capital
Consultant of Zcash
Bitfinex 早期投资人，Bitfund 基金有限合伙人
FBG Capital 顾问
Zcash 顾问
In the spirit of transparency, and in the spirit of open-sourcing the Covalent project, we will start crediting major contributors with each technical module. Currently, most of the names here are members of the Covalent Foundation. Going forward, we expect to see more part-time contributors hailing from technical communities around the world.
Private sale (20%)
Primarily for long-term project funding
Team (15%)
Incentive for top developer hires and the founding team
Foundation Reserve (10%)
For marketing, PR, and business development
Huobi NEXT (4%)
Voting rewards distributed to voters as incentives
Block Rewards (28%)
Incentive for computing and routing nodes to join computations
Ecosystem (16%)
Post main-net release ecosystem building reward for the community
Advisor (7%)
Incentives for the world-leading experts in related fields
私募投资（20%）
用于项目发展长期资金支持
团队（15%）
用于激励创始团队与顶级开发者加入
基金会运营（10%）
用于市场推广、公共关系、商业合作
火币创业板投票活动（4%）
用于激励火币创业板投票者
区块奖励（28%）
用于激励算力、路由节点
生态建设（16%）
用于主网上线后的生态建设
>顾问团队（7%）
用于激励世界顶级专家学者加入
