What on earth is DFINITY? Compared with Ethernet Square, what is the positioning of DFINITY?

At present, there are many articles about DFINITY, which are not easy for ordinary users to understand. Well, the most important thing about the project is to be understood and well known. According to this requirement, I wrote this "brief introduction to DFINITY", which mainly let you understand the historical background of DFINITY, but also let you have an overall understanding of DFINITY.

Original title: "DFINITY In A Nutshell-A Non Technical Introduction" original author: Dominic Williams,DFINITY founder, chief scientist source: official account DfinityFun translator: red Army uncle, BlockPunk chain news official account "DfinityFun" and author "mule foraging gulls" authorized reprint

Original title: "DFINITY In A Nutshell-A Non Technical Introduction"Original author: founder and chief scientist of Dominic Williams,DFINITYSource: official account DfinityFunUncle of the Red Army, BlockPunkThe chain is authorized to be reprinted by the official account "DfinityFun" and the author "mule foraging gulls"

Founder and chief scientist of Dominic Williams,DFINITY

DFINITY 究竟是什么?与以太坊相比,DFINITY 的定位如何?

1 Origin

Like many blockchain projects, the origins of DFINITY can be traced back to Bitcoin and the history behind it. Bitcoin stands out from several of the decentralized digital currencies that password punk has been trying to achieve in the past. In 1998, I stumbled upon his b-money proposal while working on a project using the Crypto++ library of Wei Dai (password punk and the main contributor to the cryptocurrency technology base.

In fact, the Bitcoin protocol, released by Satoshi Nakamoto in 2008, built an indestructible shared virtual ledger for stand-alone computers for the first time, an impressive approach (implemented on peer-to-peer networks).

With the token balance at different addresses, anyone can send the token to a new address with the accompanying access control script. His economic incentive mechanism allows any individual who provides computing power and security for the network to be rewarded.

More importantly, you can think of Nakamoto's innovation as the introduction of a new "stateful" decentralized network protocol that allows independent participants to agree to collaborate across the Internet, agree and update shared "status" (in this case, virtual books) without having to be scheduled by anyone.

Bitcoin white paper, the first to propose a viable, stateful decentralized network

DFINITY 究竟是什么?与以太坊相比,DFINITY 的定位如何?

2 Block chain 2.0

Nakamoto's creation of Bitcoin was the trigger for igniting the blockchain, but after a while, some people began to question the monotonicity of the logic it could carry, and Bitcoin had big limitations, so in 2014, Ether Square further established the foundation of blockchain 2.0.

Ethernet Square is no longer a simple shared ledger, but a shared virtual "blockchain computer", an indestructible computer that cannot be tampered with. On this virtual computer, you can install and run a special interoperable body that we call a "smart contract," because anyone who interacts with etherland can see very clearly how etherland computers will handle logic and store data.

640 (1).jpeg

640 (1).jpeg

A blockchain computer, with no specific physical instances, is ubiquitous, which means you can't tamper with data and run results by logging into the operating system.

But writing smart contracts is not for everyone. The performance and capacity of Ethernet Square computers, like previous bitcoins, are limited, making blockchain computers unsuitable for many potentially large commercial applications. Like the economic mechanism of Bitcoin, Ether Square believes in the purpose of "code is the law". This means that there is no review, governance and regulation of the use of the platform or the underlying network itself, and on the contrary, many real-world companies want strong governance.

Why believe in "code is the law"? There are many reasons for this, one of which is that if a governance center is established and integrated into a blockchain computer in cyberspace, the security of the system will inevitably be reduced, and there will be links and conflicts with the governance of local laws and regulations.

3 Block chain 3.0

Block chain nervous system-based on the chain reaction of decision, which is determined by the mutual "link" of neurons.

640 (2).jpeg

Let's imagine what blockchain 3.0 should look like. New encryption techniques and protocols are as important as the PoW and PoS protocols that build the blockchain foundation. Because they can directly solve the limitations of current block chain efficiency, performance and capacity, and provide unlimited methods to increase computing and storage capacity.

In addition, imagine that "code is the law" is replaced by the new "AI is the law". All the content on the chain is subject to a de-mediated algorithm governance system, which combines group intelligence and traditional artificial intelligence technology to freeze malicious intelligence contracts that harm the interests of users (such as the ISIS slave market, with which many companies do not want to share the network), mitigate the impact of hackers on hosting systems, dynamically optimize network security and economic adjustment parameters, and upgrade underlying protocols and software transparently. And does not affect existing users.

Now you probably have enough imagination about DFINITY's third-generation blockchain network-"decentralized intelligent cloud."

Threshold Relay Technology-A New Cornerstone for rapidly expanding Block chain performance

640 (3).jpeg

4 Vision of DFINITY

There are two main aspects to DFINITY's vision:

On the one hand, traditional monopoly centers such as Uber, eBay, social media, chat software and even web search may use autonomous software to be restructured into "open commerce", which will use their own distributed governance systems to update themselves.

On the other hand, we hope to take advantage of the features provided by blockchain computers to reconstruct the existing enterprise IT system on a large scale and cut costs significantly.

The last point does not seem obvious at the moment, because computing on blockchain computers is much more expensive than on traditional clouds such as Amazon Web Services). However, significant cost cuts are possible because the vast majority of the costs involved in running an enterprise IT system come from human costs, not from the calculation itself, and the DFINITY cloud will provide the possibility to build a business system with ultra-low labor costs.

In January 2015, DFINITY focused on its primary goal, "studying how to conduct decentralized web searches."

640 (4).jpeg

DFINITY aims to explore new areas and prove that the possibilities of blockchain are far more ambitious than anyone has realized so far. New encryption technology, unlocking more possibilities.

5 crazy sister of etheric Square

Although DFINITY is an independent project, it is not an extremist idea, but a strong supporter of Etherland. We hope that most of DFINITY's new encryption technologies can be used in the Ethernet Square network. At the same time, we will also work to create more shareable technology components.

DFINITY can be seen in part as a sister network of Etherland, providing developers with unusual features such as "AI is the law". Our goal is to provide compatibility with smart contracts where possible so that developers don't have to learn new skills and make development easier.

6 DFINITY Q / A

DFINITY is a public network of client computers that provides a "decentralized world computing cloud." You can install and run programs that have all the advantages of a "smart contract" system on a traditional blockchain and are designed with the underlying technology in mind to support the construction of a highly resilient tamper-proof private cloud. These private clouds also provide managed software services that can invoke smart contracts on public blockchain clouds, which brings many additional benefits.

DFINITY has introduced new cryptography technologies and protocols designed to provide extreme performance, unlimited scalability, and interoperability. DFINITY has also introduced a decentralized intelligent chain governance system called the blockchain nervous system.

For those interested in decentralized technologies, the technical components used by the DFINITY project are being released step by step (one of the key technologies, the code for the "threshold signature relay" technology has been released). We hope to launch the fully available test network by the end of the second quarter of 2019, as well as our unique software framework, AI algorithm governance and other features, and to launch the main network in the second half of 2019.

DFINITY can be traced back to the spirit of password punk and the ideal of decentralization. Back in 1999, Dominic Williams (Dominic Williams), president and chief scientist of DFINITY, stumbled upon his bMoney proposal while using David's crypto++ library. In 2013, Dominique gave up his past work to focus on decentralized technology and devoted himself to theoretical research in 2014. The name DFINITY is a combination of the words decentralized (decentralized) and infinite (infinity), designed to create a decentralized application platform of unlimited scale.

DFINITY maintains a close relationship with academia. Our team is located near Stanford University, and Dominic's design relies heavily on the BLS algorithm to create randomness, which was designed by Professor Dan Boneh and his PhDs. DFINITY is also a sponsor of the Stanford Block chain Research Center. In addition, we would like to support students who are willing to do distributed systems research through our scholarship program. The DFINITY project welcomes a wide range of scholars to contribute to us. If you are interested, you can contact us.

So far, the only way to make a large number of client nodes work together in any secure anti-attack network that can establish a "virtual computer" is to generate random numbers with the help of cryptography. Of course, the workload established by Satoshi Nakamoto proves that the process of allowing miners to compete with each other and find qualified answers can only be achieved by calculating random numbers by critical strike. Finally, it is allowed to find a random number of miners and add trading blocks to the chain.

DFINITY needs a more powerful random scheme with less process operation, which can produce randomness more effectively in a fixed time. The importance of randomness plays an important role not only in ensuring the consensus rights of miners, fair distribution of block incentives and other matters. Turing's complete blockchain, such as DFINITY, requires a higher level of randomness, because smart contracts can automatically rely on large transactions under any condition, including random numbers. Therefore, through the manipulation of relevant conditions, the potential benefits can be very considerable.

The solution we found was the threshold signature relay, which uses encryption to create randomness and select the appropriate network participants on demand in an almost indestructible, uncontrollable, unpredictable manner. Using the "threshold signature relay", the participants of the DFINITY network generate a deterministic verifiable random function (VRF), to support the cooperation and operation on the whole blockchain network.

On an open blockchain cloud platform, secure and stable random sources are essential. For many years, randomness has been a key component of software systems, and the implementation of many applications depends on random numbers. In addition to simply implementing applications such as fair raffle and game systems, randomness is a key goal of DFINITY, ensuring that developers who deploy code to DFINITY can trust that their code will work as expected (trusted computing).

In autonomous systems, randomness can be used to reach consensus.

Block hashing can be predicted or manipulated by attackers, so it is not a secure and trusted random source. True randomness must not be affected and predicted by any third party. Block hashes can actually be manipulated and predicted to attack the network and steal assets.

In a workload proof system, in order to "mask" a block hash, you need to create a number of candidate blocks, which costs a lot of money. In the equity proof system which does not involve strong calculation, the verifier can easily modify the content of the block to get its hash value, so that the block hash is ineffective. In either case, block hashing is not enough to meet the randomness we need.

DFINITY is a distributed operating system. It is a network composed of "clients", usually referred to as "nodes" in P2P networks. Each "client" must sign broadcast messages in its own identity, which are recorded in the globally maintained network state. Each "client" should provide some standardized computing resources, that is, data processing capacity, network bandwidth, and storage. The computing resources section uses mechanisms such as USCID, which we will continue to explain later in FAQ.

The network of clients cooperates in the manner described in the FAQ above. The threshold signature relay produces an internal random number beacon. The value of each output random number defines a random network of multiple nodes (nodes can also spontaneously form a "threshold network").

The composition of each network is completely random, so there is a crossing between the networks, and a node can exist in multiple networks. In DFINITY, each network consists of 400 nodes. When a network is established, members of the network will spontaneously use the distributed key generation protocol to set up the BLS threshold signature protocol.

If these nodes succeed in networking in a limited number of blocks, the nodes will send a special transaction to register the public key and create an identity for their own networking throughout the network in order to become an online networking and participate in the next process.

Genesis Network will contain some preset networks, one of which will be nominated to create signatures that contain default values. Such a signature must be random, if not, the message signature in the group is likely to be forged, and the threshold signature relay system will become insecure-because random values will be used to select the next round of operation of the network.

Then, the current group selected by the previous group signs the previously generated random value, produces a new random value, and then selects another network. Through the infinite relay between the group networks, a series of random values are generated. In the encryption threshold signature system, the network can generate messages through the cooperation of the network nodes with the minimum threshold, and then sign the messages. The threshold set in the DFINITY network is 51%.

To generate a threshold signature, group members sign the message separately (in this case, the threshold signature for the previous group), create a "shared signature," and then broadcast it to other group members.

Enough shared signatures can be combined into threshold signatures for the group. For example, if the group size is 400 and the threshold is set to 201, messages from enough nodes will be collected to be able to build the group's signature.

Each "shared signature" can be verified by the nodes of other networks, and the client can verify the threshold signature of each group by using the public key of the group.

The magic of the BLS scheme lies in its "uniqueness and certainty", which means that as long as enough shared signatures are collected from any subset of the same network, the threshold signatures created are always the same and unique values.

Therefore, the generated random value sequence is completely definite and uncontrollable, and the process of intergroup relay signature produces a verifiable random function, that is, VRF.

Although the sequence of random values is predetermined by a given set of participating groups, each new random number can only be generated when the current network achieves minimum consistency (that is, 51% confirmation).

On the contrary, in order for the relay process not to stop because there is no random number, the number of nodes that are not operating properly must be below the threshold. The setting of the threshold makes this almost impossible.

For example, if the group size is set to 400 and the threshold is 201, there must be 200 or more process errors to stop generating random numbers. If there are 10000 processes in the network, of which 3000 are wrong, the probability of relay aborting is less than that of 10e-17.

This system not only has incredible robustness, but also is very efficient. In the network of broadcasting gossip protocol, a network with 400 nodes only needs to forward about 20KB communication data to generate threshold signature.

At the same time, with the help of DFINITY's BLS threshold encryption library, under modern hardware conditions, all operations can be calculated in less than one millisecond.

You can learn more about threshold signature relay by reading the DFINITY consensus White Paper.

DFINITY's network is designed to store unlimited state when used, so it is impossible for a node to maintain copies of all data-after all, the amount of data may be in EB (1EB is about 1 billion GB).

Therefore, it is necessary to partition the state storage across nodes, which will naturally lead to some problems, that is, what kind of replica factor is needed to ensure sufficient security.

In order to provide a practical guarantee of data security, networks such as DFINITY need to do more to determine the underlying replica factors involved.

This also ensures that the file copied by the node is not too large-after all, on a 1m-sized client node, it is absurd to make a large copy.

USCIDs is the solution that requires each client to maintain a copy of the unique state data assigned to them-hence the acronym "Unique State Copy ID." Each client is required to store all the data encrypted by its own derived key, and the identities of all other clients must be visible.

The use of a specially tuned symmetric encryption algorithm makes encryption relatively slow and decryption relatively fast. The algorithm is designed to encrypt data when it is updated and written, but it is unrealistic to encrypt all data shard states in a limited amount of time.

The USCID system requires the client to authenticate its unique encryption status during protocol communication.

For example, when a node generates a candidate block on a threshold relay chain PSP, it must contain such a proof. In order for this block to be included in the chain and for the node to be rewarded, the block must be broadcast in a limited number of seconds, and the malicious node will have a problem.

The authentication process is a random walk on the specific encryption state of the node, and the output of the hash chain is generated-starting with a random block indicated by a random number beacon in the relay network, the block is added to the hash summary, and then another random block is selected, which continues until the data of all the blocks in a random chain of the desired length are added to the summary.

The blockchain nervous system (BNS) can access special opcodes in the virtual machine. This allows BNS to freeze, thaw, and modify other independent software objects (smart contracts).

It can also configure the DFINITY client software that the user is running, such as upgrading the network protocol to the latest version.

Dfinity's BNS is not a traditional AI like neural networks or Bayesian classifiers. On the one hand, it needs to make decisions on proposals with the help of inputs from "neurons" controlled by humans, but the other part of the decisions arise from the decentralized "follow" relationship between neurons and non-deterministic computing processes.

BNS improves the decision-making ability of neurons because when new information appears and feedback is received, the neurons are reconfigured by the owner. The actual process behind the decision is unknowable: the neuron following relationship exists only in the neuron client software that the owner runs on his own computer, and the distribution status of the neuron client software cannot be known.

This process is uncertain because time affects the way neurons communicate decisions by following each other. The purpose of BNS is to use the wisdom and knowledge of the public to determine complex issues wisely.

The voting weight of neurons is proportional to the token (DFINITIES) locked in it. Each neuron can vote under the guidance of its owner, or automatically follow the vote of other neurons on the address assigned by its owner.

This is similar to the concept of "mobile democracy". In BNS, the "follow" relationship exists only on client computers and is unknowable, which is why the system may be better described as an "opaque" mobile democracy.

Bns uses a system called wait to collect enough input and make a decision.

Yes, creating and running neurons will be rewarded for participation. You can create a neuron by locking the token (DFINITIES) of dfinity. The weight of neurons is proportional to the amount of DFINITIES held.

At the end of each DFINITY round, you will receive a reward proportional to the number of DFINITIES you lock in the neuron. The more you vote, the more you reward.

You can create nodes on DFINITY by running client software, each of which must have a "node identity". DFINITY node clients should provide relatively small but stable computing and storage capacity for the network.

Running nodes on DFINITY is very different from the way PoW digs mines. In DFINITY networks, clients play the role of processing data and are rewarded for performance.

Therefore, there is no need to add your client to some kind of mine pool system (which is even impossible), and every client you run will receive a regular reward because it will participate in supporting the network in a variety of ways.

You must mortgage DFINITIES, to the network if your client fails to function properly or attempts to defraud, your deposit may be forfeited. The blockchain nervous system adjusts the current size of the security deposit, specifically, taking into account price fluctuations and other factors.

Dfinities is the token that participates in the operation of the network.

In the current design, there are four clear uses in the network:

1. Run (and install) smart contract fuel on the blockchain cloud.

2. The "node identification" deposit will not allow the client software to join the network until the deposit is paid.

3. Through the block chain nervous system, create a deposit for "neurons" that can participate in decentralized treatment.

4. A deposit that allows a private DFINITY cloud network to connect to a public network.

While dfinities is valuable and can be traded, it is widely believed within the team that currencies need to remain stable and should be created by existing financial institutions using color coin models (for example, banks support tokens issued under their systems) or by next-generation encryption protocols that support economies that use these stable currencies, such as PHI (PHI, which is unlikely to be online until 2020.

In DFINITY, all economic policies are affected by the blockchain nervous system, including inflation. At the beginning, it issues a new dfinities as a reward for running nodes or using neurons to vote.

The specific amount of the issue will be related to fluctuations in the value of dfinities, as well as whether BNS wants more nodes to join the network, as well as a number of other factors.

However, BNS may eventually decide to use stable currencies to pay awards, such as PHI or some other system. Because this can effectively stop dfinities inflation.

Source link: mp.weixin.qq.com

Chain News only provides relevant project information and does not constitute any investment advice.

Does not constitute any investment advice.

Firmly put an end to all kinds of token issuance and hype, if you find that the article contains sensitive information, please click "report", we will deal with it in a timely manner.

Report