Editor’s note: the text from a chain learning community (ID:yikuailianxishequ), author Liu Yi, the daily planet Odaily authorized reprint.
Liu Yi, a partner at Capital Random, master of Tsinghua University, block chain and big data technology experts; 20 years of investment experience in a variety of capital market, bitcoin investors early.
Click here to read the words “long said through Polkadot architecture (on)”
Thirteen, Polkadot architecture
Next we introduce the Polkadot architecture, Chinese version of Polkadot white paper of the two images are from the translation of Yue Lipeng.
Look at the right side of the figure, the basic structure of Polkadot network is a star, or called the spoke type. The star is the center of Polkadot Relay relay chain, periphery is many Parachain parallel chain.
Look at the small figure on the left side of the Polkadot network, participants have four characters: Validator, Nominator, verify the nominee Collator collection or check, Fisherman fishing or fisherman.
DApp can be deployed in parallel intelligent contracts on the chain, or the chain is a parallel DApp. The user initiates the transaction in parallel chain transactions are collected are collected and packaged into blocks, to a group of people to verify the validation.
This group of people is not from the parallel verification chain, the verifier pool but from intermediate chain management, by grouping randomly assigned to parallel chain. Each parallel chain has an outlet and an entrance queue queue. If the user initiated cross chain transactions, the transaction will be on the import and export queue. Be parallel to the target chain were collected into the entrance queue.
The target parallel chains from the implementation of the transaction, generating block, finalized by the validation group. Polkadot uses a hybrid consensus protocol, protocol English abbreviation is BABE, a child; finalize the agreement is the abbreviation of GRANDPA, grandpa.
Just talk about mixed consensus, some students may ask: a fast, slow wait to finalize finalized, the block will not be more? No, GRANDPA each round can be finalized a number of blocks, so as to catch up with the progress of. So the child is lively, walking light; grandpa is good, with vigorous strides. An old small complement each other.
In addition to the parallel chain, there are two types of peripheral blocks into the relay link chain. One is through the bridge, it can not be directly linked to the existing, block links into the relay chain, such as Ethernet square and bitcoin etc.. From the relay chain perspective, is a parallel chain transfer bridge.
From the etheric Fang or bitcoin’s point of view, through the bridge is a side chain. In addition, in order to make the system have unlimited scalability, can also have two stage relay chain. But the two relay is currently no specific design ideas.
We have introduced the collected and verified the function of the system, then the nominee and the fisherman is what to do? The nominee is Polkadot through DOT based card holders, he hopes to gain DOT pledge. But either because the number of DOT is low, either the lack of operation and maintenance of the verifier node professional skills.
So the system provides another way to participate, is Chibidaigou choice he trusted verifier, to pledge their DOT through verification, verification of revenue sharing. The benefits of doing so is to improve the overall pledge ratio, enhance the security of the system, but also make income distribution more equitable. The economic model on the Polkadot, is a very complicated and interesting topic, we will not say more.
Just introduced the effectiveness of transactions over piece of architecture, as well as reward solutions, so it is not difficult to understand the role of the fisherman. He is the monitoring and reporting of illegal transactions, earn a bonus character. It sounds simple, but it is very complex.
Some students may think the report is this: to the Web3 foundation email: I found someone to pack evidence of illegal transactions, see annex, look forward to your reply. After a few days the Web3 foundation reply: your report has been confirmed, the perpetrators have been executed, bonuses will be sent to your address, thank you for your support to our work.
But the report is not completely block chain like this. The fisherman is a software process, it monitors the network of illegal behavior, once found to block chain report submitted to the transaction. The report also deals through consensus process, through the 2/3 above is verified, packaged into blocks, penalties and rewards are also trading chain blocks.
The whole process is automatically, to the center of the implementation. There are many complicated problems, such as how to provide incentives to the fisherman. The fisherman as the police, you may think that it is very simple, catch a bad hair a bonus.
There are a bunch of police every day staring at the network, no one dare to do evil, the police have been unable to get a bonus. The police have the operation cost, to verify and store a large amount of data, no income will not go dry. The police diverted, the bad guys will appear. You might think, to the police for your salary, basic salary plus commission.
Well, I can claim to be the police, to receive a salary. But I did not verify and store transaction data, the cost is 0, my salary is profit. When a bad guy, I said sorry, I didn’t see, or that I just broke the hard disk, then the system should be how to punish me?
There is not any report, report processing system cost, just report it would be dust vulnerabilities. In addition to report to report whether the transaction should be how to deal with, etc.. So in the decentralized environment, reporting mechanism is very complex. The working mechanism of Polkadot, at present I have yet to see specific instructions.
In Polkadot network, parallel chains assume their respective transactions, data storage, and parallel chain can interoperate with each other, have achieved the target slice. So I put Polkadot as a piecewise expansion scheme. You can compare with Serenity. You will find Polkadot than Serenity more complicated technology.
Slice Serenity is an isomorphism, with the same capacity of uniform consensus protocol. As is to provide standard containers used by the DApp specifications are the same, the developers choose a slice, put their DApp into the well.
Polkadot is the Web3 backbone network, it cannot and should not require parallel chain is uniform. Parallel chain can decide what to use, what kind of economic consensus protocol model and governance model, decide the hardware and network configuration and so on. In parallel chain is autonomous, Polkadot can be regarded as parallel chains or federal union.
Polkadot relay chain to support heterogeneous parallel links, and then realize the interconnection complexity over beacon chain Serenity. The technical complexity of the benefits is that parallel chain flexibility, without side chain 1000, according to the specific requirements and constraints, the optimal design of parallel chain development.
Fourteen, the same Cosmos
Here we introduce Cosmos, starting from the side chain. Look at the top left of the figure, A and B are two chains each side chain. That is to say, A contains the B SPV light chain chain chain to the client, A can verify B chain transactions. In turn, the B chain also contains the A SPV light chain and B chain authentication client, A chain trading.
Each side chain is the result of issuing passes in the A chain and B chain can huzhuan. If the expansion from two strands to multiple chains, A and B into A/B/C/D/E/F, an extrapolation idea is still uses the 22 mutual side chain approach, on the formation of the structure of the bottom left.
But it will come a lot of problems, each chain should be built in all other light chain client, all other synchronous chain area size, this is certainly a great burden. But every increase a chain, need to be modified in all other chain. With the increase of the number of chain blocks, the complexity of n* (n-1) /2, exponential growth, which is obviously not feasible.
The solution is to change the structure of the upper right picture shows, on the middle of a Hub Hub, this is also the block chain, it all are for the side chain chain. That is to say, through each card on the chain can go to Hub, and then Hub to the other chain, the complexity of network interconnection with the block chain number is linear relationship.
This is the practice of Cosmos, in the definition of Cosmos, access to the Hub chain called Zone partition. To access the Hub partition need 2 conditions, the first is in line with the Cosmos standard protocol, IBC chain communication protocol. The second is the requirement that the Zone has the final instant, to ensure consistency across the chain.
The Internet and Cosmos can support multiple Hub. For the existing chain, can access Cosmos Hub protocol adapter, Cosmos protocol adapter gateway called Peg Zone anchored partition. The resulting structure is right below the picture.
We start from the side chain deduction Cosmos architecture. But looking back, Cosmos partition assume their respective transactions, data storage, and can also partition between interoperability, also achieved the target. Therefore I also considered Cosmos slice expansion scheme.
Some people may disagree with this classification of. But Never mind, Cosmos is Cosmos, the classification is in order to better understand it. From a different perspective and interpretation, can use different classification methods, there is no absolute right and wrong.
Cosmos Hub and other Cosmos SDK development of the partition by Tentermint consensus protocol, and finalize the process block is one, as long as the block is obtained above 2/3 to verify the signature verifier. The benefits of doing so is simple and fast, a time to reach the second level or second level, but also has the final instant.
But in the network partition occurs, Tentermint may suspend a consensus. Cosmos and Hub partition has verified respectively, do not need to verify the dynamic random grouping. So how to protect the Cosmos cross chain transaction effectiveness? According to my understanding, Cosmos avoided this question.
Hub cannot verify the validity of the transaction, only to verify the existence of. If a partition is malicious verifier control, then the user assets in this partition is not safe, may be stolen. But this should not be regarded as Cosmos vulnerabilities, but its design choices.
Cosmos is often compared with Polkadot, but Polkadot and Serenity were more similar in architecture. A few days before the orange book article three village analogy is very appropriate. But from the viewpoint of DApp development, especially this year and next year’s heavyweight DApp development, mainly from the Polkadot and Cosmos two.
Technically, Cosmos is better than Polkadot or Serenity more simple. I am here to say simply is not derogatory, under the premise of meeting the demand, technical scheme should be as simple as possible. So a relatively simple scheme with Cosmos slice also achieved the purpose of expansion, is not very good?
Really good, so I also very optimistic about the Cosmos, it will be very suitable for certain types of DApp. But as we have repeatedly emphasized, income will be lost. Cosmos chose simple, but sacrifice in security. The security level of PoS block chain is decided by market capitalisation and pledge proportion.
The line after the Polkadot, the total market value of assuming Dot is $1 billion, half of the pledge on the network, then to double attack on the Polkadot network, the theory requires a minimum of $167 million will be more practical needs. It is a huge sum of money, so the Polkadot network is very safe, the cross chain trading with high security.
After the Cosmos main line, Atom card will be very high market value, but the network pledge Atom, only Cosmos Hub security guarantee. Partition and other Hub will be issued through their own cards to establish economic model, security.
But the partition is usually to the center of the particular application, the application scale and the market value of less than Cosmos/Polkadot such a large platform level. So it can be expected, the security level is lower than the Cosmos Hub Cosmos partition.
You cross chain trading at Cosmos. Initiate partitions need trust transaction, the target partition and Hub, if the intermediate cross chain transactions through multiple Hub, then each Hub after all to be trusted.
On the one hand is the need to trust the blockchain changed much, but the security level of some chain may not be high. So Cosmos do not trust minimization. The safety level is reduced, but decreased after enough to use? May or may not be enough, and due to the application of It differs from man to man..
Cosmos certainly knows its own short board, to provide security for the partition in future versions. It will be very difficult to achieve. To provide security for the partition, needs a large amount of verification, can only modify the consensus protocol, the block and finalized separately, of course, stochastic dynamic grouping and transaction efficiency problems verified. This is a change down the complexity of Cosmos and Polkadot are almost the same.
Fifteen, support the comparison of DApp
Then Serenity, Polkadot and Cosmos which is more suitable for the development of DApp? We compare.
The first DApp development mode, three chain can use smart contracts. Polkadot and Cosmos brought a new style of DApp development, is the development of a specific application block chain. Application of Cosmos chain to provide the tool is Cosmos SDK, currently supports Go language development. Application of Polkadot chain tool provided by Substrate, currently supports Rust language development.
Substrate application development tool chain is complete, with the application of chain model, Gavin Wood demonstrated in a new computer, released a 15 minute application chain. In addition, Substrate all modules can be customized or replaced, powerful and very flexible.
In comparison, Cosmos SDK to be thin, it mainly provides Tendermint consensus engine and IBC link communication protocol and passes the core module, most of the superstructure needs to develop.
On the performance of Serenity, each fragment is about 100tps, is still gas price high priority transaction processing. Polkadot Hub should be able to reach thousands of TPS, parallel chains can decide consensus algorithm, hardware and network, in theory there is no performance limitation. Cosmos Hub and the vast majority of partition by Tendermint, can reach thousands of tps.
About interoperability. Serenity and Ethernet square is the same as the 1, smart contracts can call each other. Polkadot parallel chain through Relay and other parallel chain interoperability, interoperability with other chain by Bridge.
Cosmos partition can huzhuan passes through Hub, by anchoring the partition with other chain huzhuan passes. IBC news and data fields, like e-mail attachments, by extending the data field, can also pass through the partition between card outside data.
Serenity access mode with Ethernet Fang 1, developers to deploy smart contracts. Polkadot Relay access to the auction slot, pledge the amount of money DOT. Cosmos is similar to Polkadot, Atom access pledge auction qualification.
Then the security patch as previously said Serenity is a standard container, put it in the DApp on the line, the safety guarantee system by. The opposite is Cosmos, regardless of whether the application of chain access Hub, are their own security. The application of Substrate chain development is the two option, or access Relay into parallel chain, by Polkadot security; or independent operation, to ensure the safety of their own.
The last is the DApp upgrade, Serenity and Ethernet square 1, does not support smart upgrade contract. Many people may have been used to it, but I think that does not support the upgrade is a big defect, and may lead to serious security problems. The first is not possible without bug code.
Solidity intelligent contract development language is not friendly to even do formal verification, formal verification, achieve the logical path of 100% coverage is not realistic. The second is DApp Internet applications, Internet applications should be flexible, iterative evolution.
Some people will think it is so intelligent agreed contract, can not be changed. In fact, the real world of the contract, have agreed to cancel or modify the terms. About two companies signed the contract, now the two sides have agreed to amend the contract, the contract itself limits can not be changed, the absurdity of the situation.
Don’t pull what code is the law, the law is not immutable and frozen, can amend the code could be abolished, but no, not very strange? The result is very absurd, on the one hand DApp has strong upgrading demand; on the other hand, the platform does not support the upgrade. So developers to find their own way, using the method of delegateCall rudimentary, awkward to achieve scalable, also called scalable design patterns.
With this approach, developers can without the user’s consent, or even modify the smart contract in case the user’s knowledge. Then the code is legal and what meaning? How can the user know intelligent contract or not can be upgraded to upgrade, which will not change what logic? Only to see their own code. It is no wonder that there are so few users of Ethernet fang. Anyway, I can not use the etheric Fang DApp. I have studied the fomo3d contract, he did not see a random number of vulnerabilities.
A small bug multi signature wallet contract Parity, lock up billions of dollars, and a lot of people hang up their. If you want to thoroughly study and use the DApp code to the trust, the world’s DApp target users, which is probably thousands.
To develop to the center of the Internet application competition DApp upgrade is a must. It must be platform support standard upgrade, not DApp gexianshentong. Analog Android platform, mobile phone App often upgrade, but the user must be informed and agreed that the new version of App increased the demand to display to the user permissions.
These are the platform control, mobile phone App can only follow not spare. The DApp upgrade should be more standardized and more stringent management, because the DApp is encrypted assets, and no trusted center. Polkadot and Cosmos allow the application of chain upgrade, Cosmos partition is own upgrade, parallel chain Polkadot can upgrade security.
How to realize the parallel chain security upgrade Polkadot, I also didn’t understand. A few weeks ago, Gavin wood Chinese, about Trust Wormhole trusted wormhole, I don’t quite understand. Who understand this part hopes to enlighten me.
In a word, in to the center of the block chain, realize the normative security system upgrade is very difficult, but there is no other choice, must support the upgrade, but also the safety of the platform security specification upgrade. I think in terms of scalability, only the direction of Polkadot is the right choice.
Sixteen, network topology comparison
Now to talk about, Serenity, Polkadot and Cosmos these three blocks in the full development of the ecological chain, will form a network topology what. Note that the logical network, not the physical network. In addition to the full development of network needs time, is discussed here about 5 years after the picture, some content is based on the architecture design of speculation.
First look at the left side, the Serenity network is like a guard of phalanx. The beacon chain is the standard bearer or team leader, each column is a standard piece. Each partition can carry some DApp. The middle of the map is Polkadot. Polkadot network is a tree, the root is the relay chain, can be connected with a plurality of parallel chains. Parallel chains may be the application of special business chain, can also support intelligent contract DApp platform.
In addition to other types of blocks through the chain bridge access relay. Analysis of Polkadot structure shows that the number of people because of verification and other restrictions, the single relay can support a limited number of parallel chains. This is about ten to one hundred magnitude, it is difficult to reach hundreds of.
Of course, even reach hundreds, also cannot achieve Web3 vision. So Polkadot will support future relay cascade. It is a root relay access relay, two relay access level relay, which has unlimited expansion capability.
On the right side of the map is Cosmos. A Hub Cosmos can be interconnected, each Hub connecting multiple partitions. There are other types of anchor patched block chain. Cosmos network topology and Polkadot rough looks like, is a tree structure. But Cosmos Hub and Hub, no one to whom to provide security problems, so there is no level.
If the hierarchy is regarded as the direction of the connection, then Polkadot is a directed acyclic graph, Cosmos is a directed acyclic graph. In fact, Cosmos network topology can be a ring, should be in order to avoid the cross chain message routing problems, choose the design without ring.
In comparison, I think the Serenity growth mode and resource allocation are rigid. The system is a top-down growth, support more divided by iteration. DApp choose which slice, faced with some uncertainty. For example, a DApp is very successful, it needs to exceed the processing ability of how to do a single slice capacity?
At present there is no way. In addition, when the DApp on-line you choose a relatively free slice. The same piece of soil especially popular D App, then your users can only endure high cost and congestion.
This is a block chain computing resources can not do on DApp distribution according to need. Cosmos and Polkadot growth is a bottom-up, there have been new application chain join, will also have the application of chain exit, resource allocation is more flexible.
Cosmos with the biggest difference between the other two platforms are not shared security, to a certain extent at the expense of trust is minimized, in front of the already mentioned. So Polkadot has shared security and bottom-up organic growth, is the best? Polkadot does have these advantages, but it also has its own disadvantages.
I think the biggest problem is the access threshold of parallel chains will be very high. According to the current published auction plan, by the end of 2020, only 24 access slot. If you develop the parallel chain, hope that next year will be with on-line, many of the world’s team competition this 24 places.
Of course, the future of intelligent edgeware contracts after the line on the platform, to a certain extent, lowering the threshold for DApp. In contrast, deployed on the Serenity DApp no threshold. Cosmos will be much better, because Cosmos Hub can support more slots, also in the ecological system will have multiple Hub, the formation of market competition.
From the bigger picture, Serenity, Polkadot and Cosmos interconnection is feasible, it will happen. EOS uses the DPoS block chains can be connected in, coupled with the side chain of two layer network, the Internet will block the formation of heterogeneous chain.
Seventeen, “on the cross chain
This share is close to the end, we began to discuss the concept of cross chain. I think the reason is “cross chain” concept is ambiguous, misleading. At least I own misguided for a long time. Used to see on the Cosmos and Polkadot materials are used them to introduce cross chain solutions.
I think why the cross chain? The cross chain capable? Although there are now hundreds of thousands of the chain, but there are a few useful? Bitcoin is one, someone will add Ethernet / EOS square, some people will add ZCash/Monero, anyway, is 35.
Just a few useful chain, what can cross? The feeling is your pants off – make an unnecessary move. So before I saw the introduction of Cosmos and Polkadot in the article, glanced at the title of the past, do not point into.
Until last year, Gavin Wood in Munich Substrate demo, I know Polkadot is a new generation of public chain architecture is a new form of divide and rule, DApp. From then on, I began to pay attention to this field.
Polkadot and Cosmos’s vision blockchain picture of the future, not one chain fit all a chain all over the world, nor the one chain rule them all a chain reigns, but formed by numerous heterogeneous chain network connecting block. The Internet is able to cross between block chain chain transactions, so the cross chain is the ability to block Internet heterogeneous chain.
Bitcoin, Ethernet and other public square chain block chain added to the Internet, is one of the results rather than the full content of the cross chain. So in my opinion, the cross chain concept is not wrong, but if you understand and become cross cross, did not seize the key.
The heterogeneous block chain to achieve cross chain trading is defined as the cross chain. Then we say what is not the cross chain, or on the market which is the so-called cross chain in cry up wine and sell vinegar. First, the cross chain must be consistent changes in two or more block chain state is a write operation.
If one is read, the other side is written, or is to modify another block a block chain chain according to the state of the data, this is not called the cross chain. Of course, read data from a plurality of chain, more than the cross chain.
There is a requirement, not only to achieve cross chain transactions and cross chain transactions should be to trust. Or take the Cosmos card through the cross chain transfer for example, from A Hub to 10 after partition passes into the B partition, you need to believe, Hub and B A partition three partition block chain network.
The three net worth of trust? You need to decide if, to the center of the high degree of PoS, the high price of pledge passes, attack is difficult, it is worth to trust. Some of the so-called cross chain scheme, Hub itself is not a block chain, but a gateway gateway, so worthy of trust?
The answer is not known, the gateway is a computing device of single entity operation. The premise is the operating entity trust trust gateway. For example, we put the money in the bank, through Alipay money, nor what is wrong. Ali and the industrial and commercial bank are worthy of trust, so they also trusted gateway operation.
But no matter whether the gateway is trustworthy, I do not belong here about the cross chain. We say that the cross chain, Hub adopts distributed books technology, but also to the center of the operation, so as to realize the minimization of trust.
To see the cross chain project called later, can be resolved, is to the center of the cross chain trading does not support heterogeneous block chain. If you can not support, it said the cross chain with Cosmos and Polkadot are not the same concept.
The development of DApp technology, the next generation eighteen
I personally believe that the DApp has landed only one, is bitcoin. Bitcoin is a store of value type currency to the center, or call the number of gold. Because the value is stored, so it is of very low performance requirements.
In the next few years, to the center of the payment and settlement currency exchange and is likely to fall. Type of payment and settlement currency, should be directly or indirectly linked with purchasing power. Now, usdt, tusd, JP Morgan currency, and the future of the currency, the central bank Facebook encryption is the center of the legal tender. Not to trust, without permission and anti censorship.
Lightning network and MakerDAO is an important attempt, may bring a breakthrough. Money, loans, asset distribution, asset trading, insurance, derivatives… We can change the world, to subvert the traditional financial also only a few really landing DApp. With the block chain expansion, infrastructure is awesome, DApp will usher in a real outbreak.
The programmer how to become DApp goers, DApp development technology we comb the next generation. Note that DApp is an Internet application. The development of the Internet of the back-end, front-end, mobile terminal, browser, desktop technology is still valid, but not within the scope of the discussion, we only talk about the implementation of decentralized technology.
Three expansion ideas, each has a variety of realization, I only list representative projects. On behalf of the representative is EOS and wave field expansion. Hierarchical expansion into state channel and the side chain of two branches, the state channel plan includes Celer Network, Counterfactual, Loom, OmiseGo and other side chain. Three typical program slice, Serenity, Polkadot and Cosmos have been introduced one by one.
So, the development of DApp technology in the next generation seems to have a lot of. In fact, not only for the development of DApp, and is the intelligent contract. Smart contracts have two mainstream programs, EVM and Wasm. We list the basic project, support EVM, now or in the future will support Wasm.
The main development language is EVM Solidity, has been formed around the Solidity ecosystem, including Truffle/Remix/OpenZepplin tools, a large number of technical information, examples, community discussion and answer and so on, but also has a considerable number of programmers in Solidity language. So master Solidity, can ensure that the development of DApp in most public chain.
The only exception is EOS, EOS does not support EVM, but one step using wasm. Wasm is the standard in the future development of smart contracts, can support the development of intelligent programming language with a variety of contracts, including java\c++\go\rust and so on.
Cosmos and Polkadot provide second ways to the development of the DApp, is the development and application of block chain. Compared with the intelligent application of contract, the advantage of the chain is of great flexibility. Developers can choose or customize the consensus algorithm, governance model, economic model and so on, and according to the actual demand of hardware and network configuration.
But on the other hand, the application of chain development, operating costs will be significantly higher intelligent contract. For example, to deploy the Cosmos partition chain, at least 4 or more of the host, but also to pledge a considerable number of Atom card to access the Hub. Can be expected to achieve a certain scale team, have the chain of sufficient resources to develop and operate Cosmos or Polkadot.
A brief summary is: the next generation of DApp development technology of two kinds of intelligent, lightweight and heavyweight contracts, the application of chain. An individual or a small entrepreneurial team will mainly use smart contracts. Large enterprises or have sufficient resources to support entrepreneurial projects, will use the application chain.
There is a reasonable approach is to first use of smart contracts to realize DApp, the verification and feedback, to develop more perfect function and application chain experience better.
How to choose nineteen, developers?
The back-end or full stack engineer to select DApp development technology? I think the blockchain world of the future will be heterogeneous block chain networks, multiple platforms will have their own living space, do not have to say who will replace him.
So my advice is: mainly consider the current conditions to select the technical route, such as the ability of currently available platforms, the skills of team members and so on. Not too tangled the uncertainty of the future.
If the project has been started, it will continue to go along the line. If not, the first clear is the development of public DApp chain or the development of enterprise solutions. If the public is the development of chain DApp, then the need to existing DApp interoperability. If it is, then we should develop public choice in the existing DApp on the same chain.
That is to say you need interoperability of DApp in Ethernet workshop, you will develop in the etheric workshop. In EOS, you have developed in EOS. One might ask, can not cross the chain? The cross chain is very complex, and certainly not free. Under the premise of meeting the demand, implementation should be as simple as possible. So as to avoid the cross chain, of course should be avoided.
That if you do not need to interoperate with existing DApp? It implied a problem is, you will DApp with future DApp interoperability? Or others will DApp to your DApp interoperability.
For example, you develop a pass card contract, as the company’s business integration system. If your company uses a wide range of integration, is not can be traded to the center of the exchange, is not can be used as collateral for loans to assets of DApp? The success of DApp should be as far as possible into the ecological value of the internet. So DApp interoperability is not an option, but the basic needs of DApp, but not necessarily with the existing DApp interoperability.
The next question is: DApp is not only the need for management through the encryption card? If the answer is yes, should give priority to the Cosmos, if the answer is no, you should choose the Substrate. Why? We have seen, Cosmos can cross the chain through transfer, while Polkadot can achieve any form of DApp interoperability.
Someone may say, Polkadot more powerful and flexible. I agree with this, but we should understand that the world is fair, there is no free lunch. A powerful and flexible price is complex and high cost.
In the framework of Cosmos are much simpler than Polkadot. So I can be inferred with certainty, at least in the early stage of Cosmos will be more reliable than Polkadot, and the use of lower cost. So if Cosmos meet the demand, you should choose it.
If the business needs through beyond the encryption certificate category, you can choose Substrate. In front of a branch, the development of enterprises is the block chain applications or development alliance chain should choose what? I think you should choose Substrate.
Why not choose Hyperlegder Fabric or Ethernet square? Because the technology platform business to promote competition, but destined to mainstream open platform. As for Ethernet Fang Development Alliance chain, I think the lack of flexibility.
The business is usually very complex, high performance and manageability are required, and often requires rapid iteration. Substrate is a block chain framework complete, highly modular and can be customized. Rust language focuses on security and performance, but also very suitable for the development of key business system.
Twenty, Rust language
Do not open around the Polkadot/Substrate development, the problem is the Rust language. In my more than 20 years of development, encountered very few programmers to quickly master the new language. But the vast majority of programmers, including myself, to master a language, familiar with the standard library and development framework, the need for a long time.
So learning a new language, for me, is a hard decision. In recent years, the emergence of a lot of development language, the Rust language with a steep learning curve is known. In the end it is worthwhile to spend a lot of time to learn Rust language? As a Rust language for beginners, I talk about my own views.
The first is the Rust language is great, but has not been popular. This is not my personal opinion. A survey in 2019 Stack Overflow, according to popularity rankings, Rust only in the twenty-first row, the picture on the left shows only the top 16, so can not see Rust.
In addition to the Tiobe programming language index ranking, Rust is currently ranked thirty-fourth. Look at the right side of the map, in the Stack Overflow survey most programmers favorite development language list, Rust ranked first. And this is the Rust for four consecutive years ranked first, taking into account the Rust1.0 version 2015 was released, it can be said that the Rust official debut, has been the most popular language programmers.
But you know, a programming language, the pursuit of high performance, and easy to let developers, usually can’t have. For example, Java and Python programmers do not need to consider the memory management, of course, reduce the burden of learning and development, but the system has to deal with garbage collection, performance problems.
On the other hand, c/c++ requires the programmer to manage its own memory, can achieve the best performance, but the premise is to put the program to write, increasing the difficulty of the development. But Rust seems to have both fish and bear’s paw, which achieves the same performance with the c/c++ level, but also by the programmers love.
The second view is that the Rust language suitable for senior programmer development platform project. Rust language attaches great importance to security and performance, it tries to pass the language specification, or compiler, to guide programmers to write efficient code security. Familiar with the use of Rust, can naturally develop high performance and high reliability, it is called the Rust road.
Of course, the quality of the code may be uneven in quality poor maintainability as well, and. Learn and master Rust, need to understand some important programming concepts, such as the ownership of the object. Without a solid development experience, it is difficult to grasp. This is the reason we think that the Rust language learning curve steep.
If not the development of performance and safety requirements of high level platform project, Rust is overkill. Because of the performance and security benefits, can not offset by the cost of the new development of language.
If you want to develop a platform level system or key business system, Rust is worth considering. In addition to learn Rust language can drive their own understanding of memory, threads, asynchronous, closure, function and other advanced development concepts, of great benefit to the development of ability.
It is a summary, if you have some experience in the development of the system in the future may or may not be interested in the development platform, including block chain development, then the Rust language is worth your time to learn.
I use Russell’s words as the share of ending notes diversity is essential to happiness. To truly understand and appreciate Polkadot and Cosmos design concept, is not the maximum doctrine, do not think the goal of Polkadot is to replace the etheric fang.
At least the etheric Fang foundation and the Parity companies do not think so. Parity company is one of the most important pillar of the etheric Fang ecology, they are also actively involved in the development of Serenity.
Earlier, the etheric Fang foundation to Parity company to pay $5 million, which is about Parity years Ethernet square thanks, also funded they continue to develop software maintenance workshop Ethernet node. I have a bit surprising divity, at least I can understand why they think so.
But now the etheric Fang ecology, begin to appear the etheric religion is won’t listen to reason. Open is the foundation of the etheric vision Fang, Fang Ethernet has led us to see the possibility to the center of the value of the internet. So I think the Ethernet square, but against the other block chain, is a contradiction.
On the front, demand of people to the center of the application are diverse. Serenity, Polkadot, Cosmos and EOS, DApp and other public platform chain, do choose different design, or a different compromise. They will be very suitable for certain needs, but not applicable to other needs.
The interconnection is represent the general trend any one if you choose, the ecological development will be isolated, block chain Internet generates huge network effect by extrusion, was eventually eliminated. Therefore, we can expect that the future will block the chain more diversity, All flowers bloom together., hope to the center of the block chain and application become the source of happiness.